]> git.sesse.net Git - ffmpeg/blobdiff - libavcodec/imgconvert.c
10l for myself, fixing --disable-encoders
[ffmpeg] / libavcodec / imgconvert.c
index 403336812dc683707a95ad9ceba31fb54dcec991..2971afaa89eaebfd0d588ff186ad83c9a1f49273 100644 (file)
 /*
  * Misc image convertion routines
- * Copyright (c) 2001 Gerard Lantau.
+ * Copyright (c) 2001, 2002, 2003 Fabrice Bellard.
  *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License as published by
- * the Free Software Foundation; either version 2 of the License, or
- * (at your option) any later version.
+ * This file is part of FFmpeg.
  *
- * This program is distributed in the hope that it will be useful,
+ * FFmpeg is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU Lesser General Public
+ * License as published by the Free Software Foundation; either
+ * version 2.1 of the License, or (at your option) any later version.
+ *
+ * FFmpeg is distributed in the hope that it will be useful,
  * but WITHOUT ANY WARRANTY; without even the implied warranty of
- * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
- * GNU General Public License for more details.
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * Lesser General Public License for more details.
  *
- * You should have received a copy of the GNU General Public License
- * along with this program; if not, write to the Free Software
- * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
+ * You should have received a copy of the GNU Lesser General Public
+ * License along with FFmpeg; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+/**
+ * @file imgconvert.c
+ * Misc image convertion routines.
+ */
+
+/* TODO:
+ * - write 'ffimg' program to test all the image related stuff
+ * - move all api to slice based system
+ * - integrate deinterlacing, postprocessing and scaling in the conversion process
  */
-#include <stdlib.h>
-#include <stdio.h>
-#include <string.h>
 
-#include "avcodec.h"
-#include "dsputil.h"
+#include "avcodec.h"
+#include "dsputil.h"
+
+#ifdef USE_FASTMEMCPY
+#include "libvo/fastmemcpy.h"
+#endif
+
+#ifdef HAVE_MMX
+#include "i386/mmx.h"
+#endif
+
+#define xglue(x, y) x ## y
+#define glue(x, y) xglue(x, y)
+
+#define FF_COLOR_RGB      0 /**< RGB color space */
+#define FF_COLOR_GRAY     1 /**< gray color space */
+#define FF_COLOR_YUV      2 /**< YUV color space. 16 <= Y <= 235, 16 <= U, V <= 240 */
+#define FF_COLOR_YUV_JPEG 3 /**< YUV color space. 0 <= Y <= 255, 0 <= U, V <= 255 */
+
+#define FF_PIXEL_PLANAR   0 /**< each channel has one component in AVPicture */
+#define FF_PIXEL_PACKED   1 /**< only one components containing all the channels */
+#define FF_PIXEL_PALETTE  2  /**< one components containing indexes for a palette */
+
+typedef struct PixFmtInfo {
+    const char *name;
+    uint8_t nb_channels;     /**< number of channels (including alpha) */
+    uint8_t color_type;      /**< color type (see FF_COLOR_xxx constants) */
+    uint8_t pixel_type;      /**< pixel storage type (see FF_PIXEL_xxx constants) */
+    uint8_t is_alpha : 1;    /**< true if alpha can be specified */
+    uint8_t x_chroma_shift;  /**< X chroma subsampling factor is 2 ^ shift */
+    uint8_t y_chroma_shift;  /**< Y chroma subsampling factor is 2 ^ shift */
+    uint8_t depth;           /**< bit depth of the color components */
+} PixFmtInfo;
+
+/* this table gives more information about formats */
+static const PixFmtInfo pix_fmt_info[PIX_FMT_NB] = {
+    /* YUV formats */
+    [PIX_FMT_YUV420P] = {
+        .name = "yuv420p",
+        .nb_channels = 3,
+        .color_type = FF_COLOR_YUV,
+        .pixel_type = FF_PIXEL_PLANAR,
+        .depth = 8,
+        .x_chroma_shift = 1, .y_chroma_shift = 1,
+    },
+    [PIX_FMT_YUV422P] = {
+        .name = "yuv422p",
+        .nb_channels = 3,
+        .color_type = FF_COLOR_YUV,
+        .pixel_type = FF_PIXEL_PLANAR,
+        .depth = 8,
+        .x_chroma_shift = 1, .y_chroma_shift = 0,
+    },
+    [PIX_FMT_YUV444P] = {
+        .name = "yuv444p",
+        .nb_channels = 3,
+        .color_type = FF_COLOR_YUV,
+        .pixel_type = FF_PIXEL_PLANAR,
+        .depth = 8,
+        .x_chroma_shift = 0, .y_chroma_shift = 0,
+    },
+    [PIX_FMT_YUYV422] = {
+        .name = "yuyv422",
+        .nb_channels = 1,
+        .color_type = FF_COLOR_YUV,
+        .pixel_type = FF_PIXEL_PACKED,
+        .depth = 8,
+        .x_chroma_shift = 1, .y_chroma_shift = 0,
+    },
+    [PIX_FMT_UYVY422] = {
+        .name = "uyvy422",
+        .nb_channels = 1,
+        .color_type = FF_COLOR_YUV,
+        .pixel_type = FF_PIXEL_PACKED,
+        .depth = 8,
+        .x_chroma_shift = 1, .y_chroma_shift = 0,
+    },
+    [PIX_FMT_YUV410P] = {
+        .name = "yuv410p",
+        .nb_channels = 3,
+        .color_type = FF_COLOR_YUV,
+        .pixel_type = FF_PIXEL_PLANAR,
+        .depth = 8,
+        .x_chroma_shift = 2, .y_chroma_shift = 2,
+    },
+    [PIX_FMT_YUV411P] = {
+        .name = "yuv411p",
+        .nb_channels = 3,
+        .color_type = FF_COLOR_YUV,
+        .pixel_type = FF_PIXEL_PLANAR,
+        .depth = 8,
+        .x_chroma_shift = 2, .y_chroma_shift = 0,
+    },
+
+    /* JPEG YUV */
+    [PIX_FMT_YUVJ420P] = {
+        .name = "yuvj420p",
+        .nb_channels = 3,
+        .color_type = FF_COLOR_YUV_JPEG,
+        .pixel_type = FF_PIXEL_PLANAR,
+        .depth = 8,
+        .x_chroma_shift = 1, .y_chroma_shift = 1,
+    },
+    [PIX_FMT_YUVJ422P] = {
+        .name = "yuvj422p",
+        .nb_channels = 3,
+        .color_type = FF_COLOR_YUV_JPEG,
+        .pixel_type = FF_PIXEL_PLANAR,
+        .depth = 8,
+        .x_chroma_shift = 1, .y_chroma_shift = 0,
+    },
+    [PIX_FMT_YUVJ444P] = {
+        .name = "yuvj444p",
+        .nb_channels = 3,
+        .color_type = FF_COLOR_YUV_JPEG,
+        .pixel_type = FF_PIXEL_PLANAR,
+        .depth = 8,
+        .x_chroma_shift = 0, .y_chroma_shift = 0,
+    },
+
+    /* RGB formats */
+    [PIX_FMT_RGB24] = {
+        .name = "rgb24",
+        .nb_channels = 3,
+        .color_type = FF_COLOR_RGB,
+        .pixel_type = FF_PIXEL_PACKED,
+        .depth = 8,
+        .x_chroma_shift = 0, .y_chroma_shift = 0,
+    },
+    [PIX_FMT_BGR24] = {
+        .name = "bgr24",
+        .nb_channels = 3,
+        .color_type = FF_COLOR_RGB,
+        .pixel_type = FF_PIXEL_PACKED,
+        .depth = 8,
+        .x_chroma_shift = 0, .y_chroma_shift = 0,
+    },
+    [PIX_FMT_RGB32] = {
+        .name = "rgb32",
+        .nb_channels = 4, .is_alpha = 1,
+        .color_type = FF_COLOR_RGB,
+        .pixel_type = FF_PIXEL_PACKED,
+        .depth = 8,
+        .x_chroma_shift = 0, .y_chroma_shift = 0,
+    },
+    [PIX_FMT_RGB565] = {
+        .name = "rgb565",
+        .nb_channels = 3,
+        .color_type = FF_COLOR_RGB,
+        .pixel_type = FF_PIXEL_PACKED,
+        .depth = 5,
+        .x_chroma_shift = 0, .y_chroma_shift = 0,
+    },
+    [PIX_FMT_RGB555] = {
+        .name = "rgb555",
+        .nb_channels = 3,
+        .color_type = FF_COLOR_RGB,
+        .pixel_type = FF_PIXEL_PACKED,
+        .depth = 5,
+        .x_chroma_shift = 0, .y_chroma_shift = 0,
+    },
+
+    /* gray / mono formats */
+    [PIX_FMT_GRAY16BE] = {
+        .name = "gray16be",
+        .nb_channels = 1,
+        .color_type = FF_COLOR_GRAY,
+        .pixel_type = FF_PIXEL_PLANAR,
+        .depth = 16,
+    },
+    [PIX_FMT_GRAY16LE] = {
+        .name = "gray16le",
+        .nb_channels = 1,
+        .color_type = FF_COLOR_GRAY,
+        .pixel_type = FF_PIXEL_PLANAR,
+        .depth = 16,
+    },
+    [PIX_FMT_GRAY8] = {
+        .name = "gray",
+        .nb_channels = 1,
+        .color_type = FF_COLOR_GRAY,
+        .pixel_type = FF_PIXEL_PLANAR,
+        .depth = 8,
+    },
+    [PIX_FMT_MONOWHITE] = {
+        .name = "monow",
+        .nb_channels = 1,
+        .color_type = FF_COLOR_GRAY,
+        .pixel_type = FF_PIXEL_PLANAR,
+        .depth = 1,
+    },
+    [PIX_FMT_MONOBLACK] = {
+        .name = "monob",
+        .nb_channels = 1,
+        .color_type = FF_COLOR_GRAY,
+        .pixel_type = FF_PIXEL_PLANAR,
+        .depth = 1,
+    },
+
+    /* paletted formats */
+    [PIX_FMT_PAL8] = {
+        .name = "pal8",
+        .nb_channels = 4, .is_alpha = 1,
+        .color_type = FF_COLOR_RGB,
+        .pixel_type = FF_PIXEL_PALETTE,
+        .depth = 8,
+    },
+    [PIX_FMT_XVMC_MPEG2_MC] = {
+        .name = "xvmcmc",
+    },
+    [PIX_FMT_XVMC_MPEG2_IDCT] = {
+        .name = "xvmcidct",
+    },
+    [PIX_FMT_UYYVYY411] = {
+        .name = "uyyvyy411",
+        .nb_channels = 1,
+        .color_type = FF_COLOR_YUV,
+        .pixel_type = FF_PIXEL_PACKED,
+        .depth = 8,
+        .x_chroma_shift = 2, .y_chroma_shift = 0,
+    },
+    [PIX_FMT_BGR32] = {
+        .name = "bgr32",
+        .nb_channels = 4, .is_alpha = 1,
+        .color_type = FF_COLOR_RGB,
+        .pixel_type = FF_PIXEL_PACKED,
+        .depth = 8,
+        .x_chroma_shift = 0, .y_chroma_shift = 0,
+    },
+    [PIX_FMT_BGR565] = {
+        .name = "bgr565",
+        .nb_channels = 3,
+        .color_type = FF_COLOR_RGB,
+        .pixel_type = FF_PIXEL_PACKED,
+        .depth = 5,
+        .x_chroma_shift = 0, .y_chroma_shift = 0,
+    },
+    [PIX_FMT_BGR555] = {
+        .name = "bgr555",
+        .nb_channels = 3,
+        .color_type = FF_COLOR_RGB,
+        .pixel_type = FF_PIXEL_PACKED,
+        .depth = 5,
+        .x_chroma_shift = 0, .y_chroma_shift = 0,
+    },
+    [PIX_FMT_RGB8] = {
+        .name = "rgb8",
+        .nb_channels = 1,
+        .color_type = FF_COLOR_RGB,
+        .pixel_type = FF_PIXEL_PACKED,
+        .depth = 8,
+        .x_chroma_shift = 0, .y_chroma_shift = 0,
+    },
+    [PIX_FMT_RGB4] = {
+        .name = "rgb4",
+        .nb_channels = 1,
+        .color_type = FF_COLOR_RGB,
+        .pixel_type = FF_PIXEL_PACKED,
+        .depth = 4,
+        .x_chroma_shift = 0, .y_chroma_shift = 0,
+    },
+    [PIX_FMT_RGB4_BYTE] = {
+        .name = "rgb4_byte",
+        .nb_channels = 1,
+        .color_type = FF_COLOR_RGB,
+        .pixel_type = FF_PIXEL_PACKED,
+        .depth = 8,
+        .x_chroma_shift = 0, .y_chroma_shift = 0,
+    },
+    [PIX_FMT_BGR8] = {
+        .name = "bgr8",
+        .nb_channels = 1,
+        .color_type = FF_COLOR_RGB,
+        .pixel_type = FF_PIXEL_PACKED,
+        .depth = 8,
+        .x_chroma_shift = 0, .y_chroma_shift = 0,
+    },
+    [PIX_FMT_BGR4] = {
+        .name = "bgr4",
+        .nb_channels = 1,
+        .color_type = FF_COLOR_RGB,
+        .pixel_type = FF_PIXEL_PACKED,
+        .depth = 4,
+        .x_chroma_shift = 0, .y_chroma_shift = 0,
+    },
+    [PIX_FMT_BGR4_BYTE] = {
+        .name = "bgr4_byte",
+        .nb_channels = 1,
+        .color_type = FF_COLOR_RGB,
+        .pixel_type = FF_PIXEL_PACKED,
+        .depth = 8,
+        .x_chroma_shift = 0, .y_chroma_shift = 0,
+    },
+    [PIX_FMT_NV12] = {
+        .name = "nv12",
+        .nb_channels = 2,
+        .color_type = FF_COLOR_YUV,
+        .pixel_type = FF_PIXEL_PLANAR,
+        .depth = 8,
+        .x_chroma_shift = 1, .y_chroma_shift = 1,
+    },
+    [PIX_FMT_NV21] = {
+        .name = "nv12",
+        .nb_channels = 2,
+        .color_type = FF_COLOR_YUV,
+        .pixel_type = FF_PIXEL_PLANAR,
+        .depth = 8,
+        .x_chroma_shift = 1, .y_chroma_shift = 1,
+    },
+
+    [PIX_FMT_BGR32_1] = {
+        .name = "bgr32_1",
+        .nb_channels = 4, .is_alpha = 1,
+        .color_type = FF_COLOR_RGB,
+        .pixel_type = FF_PIXEL_PACKED,
+        .depth = 8,
+        .x_chroma_shift = 0, .y_chroma_shift = 0,
+    },
+    [PIX_FMT_RGB32_1] = {
+        .name = "rgb32_1",
+        .nb_channels = 4, .is_alpha = 1,
+        .color_type = FF_COLOR_RGB,
+        .pixel_type = FF_PIXEL_PACKED,
+        .depth = 8,
+        .x_chroma_shift = 0, .y_chroma_shift = 0,
+    },
+};
+
+void avcodec_get_chroma_sub_sample(int pix_fmt, int *h_shift, int *v_shift)
+{
+    *h_shift = pix_fmt_info[pix_fmt].x_chroma_shift;
+    *v_shift = pix_fmt_info[pix_fmt].y_chroma_shift;
+}
+
+const char *avcodec_get_pix_fmt_name(int pix_fmt)
+{
+    if (pix_fmt < 0 || pix_fmt >= PIX_FMT_NB)
+        return "???";
+    else
+        return pix_fmt_info[pix_fmt].name;
+}
+
+enum PixelFormat avcodec_get_pix_fmt(const char* name)
+{
+    int i;
+
+    for (i=0; i < PIX_FMT_NB; i++)
+         if (!strcmp(pix_fmt_info[i].name, name))
+             break;
+    return i;
+}
+
+int avpicture_fill(AVPicture *picture, uint8_t *ptr,
+                   int pix_fmt, int width, int height)
+{
+    int size, w2, h2, size2;
+    const PixFmtInfo *pinfo;
+
+    if(avcodec_check_dimensions(NULL, width, height))
+        goto fail;
+
+    pinfo = &pix_fmt_info[pix_fmt];
+    size = width * height;
+    switch(pix_fmt) {
+    case PIX_FMT_YUV420P:
+    case PIX_FMT_YUV422P:
+    case PIX_FMT_YUV444P:
+    case PIX_FMT_YUV410P:
+    case PIX_FMT_YUV411P:
+    case PIX_FMT_YUVJ420P:
+    case PIX_FMT_YUVJ422P:
+    case PIX_FMT_YUVJ444P:
+        w2 = (width + (1 << pinfo->x_chroma_shift) - 1) >> pinfo->x_chroma_shift;
+        h2 = (height + (1 << pinfo->y_chroma_shift) - 1) >> pinfo->y_chroma_shift;
+        size2 = w2 * h2;
+        picture->data[0] = ptr;
+        picture->data[1] = picture->data[0] + size;
+        picture->data[2] = picture->data[1] + size2;
+        picture->linesize[0] = width;
+        picture->linesize[1] = w2;
+        picture->linesize[2] = w2;
+        return size + 2 * size2;
+    case PIX_FMT_NV12:
+    case PIX_FMT_NV21:
+        w2 = (width + (1 << pinfo->x_chroma_shift) - 1) >> pinfo->x_chroma_shift;
+        h2 = (height + (1 << pinfo->y_chroma_shift) - 1) >> pinfo->y_chroma_shift;
+        size2 = w2 * h2 * 2;
+        picture->data[0] = ptr;
+        picture->data[1] = picture->data[0] + size;
+        picture->data[2] = NULL;
+        picture->linesize[0] = width;
+        picture->linesize[1] = w2;
+        picture->linesize[2] = 0;
+        return size + 2 * size2;
+    case PIX_FMT_RGB24:
+    case PIX_FMT_BGR24:
+        picture->data[0] = ptr;
+        picture->data[1] = NULL;
+        picture->data[2] = NULL;
+        picture->linesize[0] = width * 3;
+        return size * 3;
+    case PIX_FMT_RGB32:
+    case PIX_FMT_BGR32:
+    case PIX_FMT_RGB32_1:
+    case PIX_FMT_BGR32_1:
+        picture->data[0] = ptr;
+        picture->data[1] = NULL;
+        picture->data[2] = NULL;
+        picture->linesize[0] = width * 4;
+        return size * 4;
+    case PIX_FMT_GRAY16BE:
+    case PIX_FMT_GRAY16LE:
+    case PIX_FMT_BGR555:
+    case PIX_FMT_BGR565:
+    case PIX_FMT_RGB555:
+    case PIX_FMT_RGB565:
+    case PIX_FMT_YUYV422:
+        picture->data[0] = ptr;
+        picture->data[1] = NULL;
+        picture->data[2] = NULL;
+        picture->linesize[0] = width * 2;
+        return size * 2;
+    case PIX_FMT_UYVY422:
+        picture->data[0] = ptr;
+        picture->data[1] = NULL;
+        picture->data[2] = NULL;
+        picture->linesize[0] = width * 2;
+        return size * 2;
+    case PIX_FMT_UYYVYY411:
+        picture->data[0] = ptr;
+        picture->data[1] = NULL;
+        picture->data[2] = NULL;
+        picture->linesize[0] = width + width/2;
+        return size + size/2;
+    case PIX_FMT_RGB8:
+    case PIX_FMT_BGR8:
+    case PIX_FMT_RGB4_BYTE:
+    case PIX_FMT_BGR4_BYTE:
+    case PIX_FMT_GRAY8:
+        picture->data[0] = ptr;
+        picture->data[1] = NULL;
+        picture->data[2] = NULL;
+        picture->linesize[0] = width;
+        return size;
+    case PIX_FMT_RGB4:
+    case PIX_FMT_BGR4:
+        picture->data[0] = ptr;
+        picture->data[1] = NULL;
+        picture->data[2] = NULL;
+        picture->linesize[0] = width / 2;
+        return size / 2;
+    case PIX_FMT_MONOWHITE:
+    case PIX_FMT_MONOBLACK:
+        picture->data[0] = ptr;
+        picture->data[1] = NULL;
+        picture->data[2] = NULL;
+        picture->linesize[0] = (width + 7) >> 3;
+        return picture->linesize[0] * height;
+    case PIX_FMT_PAL8:
+        size2 = (size + 3) & ~3;
+        picture->data[0] = ptr;
+        picture->data[1] = ptr + size2; /* palette is stored here as 256 32 bit words */
+        picture->data[2] = NULL;
+        picture->linesize[0] = width;
+        picture->linesize[1] = 4;
+        return size2 + 256 * 4;
+    default:
+fail:
+        picture->data[0] = NULL;
+        picture->data[1] = NULL;
+        picture->data[2] = NULL;
+        picture->data[3] = NULL;
+        return -1;
+    }
+}
+
+int avpicture_layout(const AVPicture* src, int pix_fmt, int width, int height,
+                     unsigned char *dest, int dest_size)
+{
+    const PixFmtInfo* pf = &pix_fmt_info[pix_fmt];
+    int i, j, w, h, data_planes;
+    const unsigned char* s;
+    int size = avpicture_get_size(pix_fmt, width, height);
+
+    if (size > dest_size || size < 0)
+        return -1;
+
+    if (pf->pixel_type == FF_PIXEL_PACKED || pf->pixel_type == FF_PIXEL_PALETTE) {
+        if (pix_fmt == PIX_FMT_YUYV422 ||
+            pix_fmt == PIX_FMT_UYVY422 ||
+            pix_fmt == PIX_FMT_BGR565 ||
+            pix_fmt == PIX_FMT_BGR555 ||
+            pix_fmt == PIX_FMT_RGB565 ||
+            pix_fmt == PIX_FMT_RGB555)
+            w = width * 2;
+        else if (pix_fmt == PIX_FMT_UYYVYY411)
+          w = width + width/2;
+        else if (pix_fmt == PIX_FMT_PAL8)
+          w = width;
+        else
+          w = width * (pf->depth * pf->nb_channels / 8);
+
+        data_planes = 1;
+        h = height;
+    } else {
+        data_planes = pf->nb_channels;
+        w = (width*pf->depth + 7)/8;
+        h = height;
+    }
+
+    for (i=0; i<data_planes; i++) {
+         if (i == 1) {
+             w = width >> pf->x_chroma_shift;
+             h = height >> pf->y_chroma_shift;
+         }
+         s = src->data[i];
+         for(j=0; j<h; j++) {
+             memcpy(dest, s, w);
+             dest += w;
+             s += src->linesize[i];
+         }
+    }
+
+    if (pf->pixel_type == FF_PIXEL_PALETTE)
+        memcpy((unsigned char *)(((size_t)dest + 3) & ~3), src->data[1], 256 * 4);
+
+    return size;
+}
+
+int avpicture_get_size(int pix_fmt, int width, int height)
+{
+    AVPicture dummy_pict;
+    return avpicture_fill(&dummy_pict, NULL, pix_fmt, width, height);
+}
+
+int avcodec_get_pix_fmt_loss(int dst_pix_fmt, int src_pix_fmt,
+                             int has_alpha)
+{
+    const PixFmtInfo *pf, *ps;
+    int loss;
+
+    ps = &pix_fmt_info[src_pix_fmt];
+    pf = &pix_fmt_info[dst_pix_fmt];
+
+    /* compute loss */
+    loss = 0;
+    pf = &pix_fmt_info[dst_pix_fmt];
+    if (pf->depth < ps->depth ||
+        (dst_pix_fmt == PIX_FMT_RGB555 && src_pix_fmt == PIX_FMT_RGB565))
+        loss |= FF_LOSS_DEPTH;
+    if (pf->x_chroma_shift > ps->x_chroma_shift ||
+        pf->y_chroma_shift > ps->y_chroma_shift)
+        loss |= FF_LOSS_RESOLUTION;
+    switch(pf->color_type) {
+    case FF_COLOR_RGB:
+        if (ps->color_type != FF_COLOR_RGB &&
+            ps->color_type != FF_COLOR_GRAY)
+            loss |= FF_LOSS_COLORSPACE;
+        break;
+    case FF_COLOR_GRAY:
+        if (ps->color_type != FF_COLOR_GRAY)
+            loss |= FF_LOSS_COLORSPACE;
+        break;
+    case FF_COLOR_YUV:
+        if (ps->color_type != FF_COLOR_YUV)
+            loss |= FF_LOSS_COLORSPACE;
+        break;
+    case FF_COLOR_YUV_JPEG:
+        if (ps->color_type != FF_COLOR_YUV_JPEG &&
+            ps->color_type != FF_COLOR_YUV &&
+            ps->color_type != FF_COLOR_GRAY)
+            loss |= FF_LOSS_COLORSPACE;
+        break;
+    default:
+        /* fail safe test */
+        if (ps->color_type != pf->color_type)
+            loss |= FF_LOSS_COLORSPACE;
+        break;
+    }
+    if (pf->color_type == FF_COLOR_GRAY &&
+        ps->color_type != FF_COLOR_GRAY)
+        loss |= FF_LOSS_CHROMA;
+    if (!pf->is_alpha && (ps->is_alpha && has_alpha))
+        loss |= FF_LOSS_ALPHA;
+    if (pf->pixel_type == FF_PIXEL_PALETTE &&
+        (ps->pixel_type != FF_PIXEL_PALETTE && ps->color_type != FF_COLOR_GRAY))
+        loss |= FF_LOSS_COLORQUANT;
+    return loss;
+}
+
+static int avg_bits_per_pixel(int pix_fmt)
+{
+    int bits;
+    const PixFmtInfo *pf;
+
+    pf = &pix_fmt_info[pix_fmt];
+    switch(pf->pixel_type) {
+    case FF_PIXEL_PACKED:
+        switch(pix_fmt) {
+        case PIX_FMT_YUYV422:
+        case PIX_FMT_UYVY422:
+        case PIX_FMT_RGB565:
+        case PIX_FMT_RGB555:
+        case PIX_FMT_BGR565:
+        case PIX_FMT_BGR555:
+            bits = 16;
+            break;
+        case PIX_FMT_UYYVYY411:
+            bits = 12;
+            break;
+        default:
+            bits = pf->depth * pf->nb_channels;
+            break;
+        }
+        break;
+    case FF_PIXEL_PLANAR:
+        if (pf->x_chroma_shift == 0 && pf->y_chroma_shift == 0) {
+            bits = pf->depth * pf->nb_channels;
+        } else {
+            bits = pf->depth + ((2 * pf->depth) >>
+                                (pf->x_chroma_shift + pf->y_chroma_shift));
+        }
+        break;
+    case FF_PIXEL_PALETTE:
+        bits = 8;
+        break;
+    default:
+        bits = -1;
+        break;
+    }
+    return bits;
+}
+
+static int avcodec_find_best_pix_fmt1(int pix_fmt_mask,
+                                      int src_pix_fmt,
+                                      int has_alpha,
+                                      int loss_mask)
+{
+    int dist, i, loss, min_dist, dst_pix_fmt;
+
+    /* find exact color match with smallest size */
+    dst_pix_fmt = -1;
+    min_dist = 0x7fffffff;
+    for(i = 0;i < PIX_FMT_NB; i++) {
+        if (pix_fmt_mask & (1 << i)) {
+            loss = avcodec_get_pix_fmt_loss(i, src_pix_fmt, has_alpha) & loss_mask;
+            if (loss == 0) {
+                dist = avg_bits_per_pixel(i);
+                if (dist < min_dist) {
+                    min_dist = dist;
+                    dst_pix_fmt = i;
+                }
+            }
+        }
+    }
+    return dst_pix_fmt;
+}
+
+int avcodec_find_best_pix_fmt(int pix_fmt_mask, int src_pix_fmt,
+                              int has_alpha, int *loss_ptr)
+{
+    int dst_pix_fmt, loss_mask, i;
+    static const int loss_mask_order[] = {
+        ~0, /* no loss first */
+        ~FF_LOSS_ALPHA,
+        ~FF_LOSS_RESOLUTION,
+        ~(FF_LOSS_COLORSPACE | FF_LOSS_RESOLUTION),
+        ~FF_LOSS_COLORQUANT,
+        ~FF_LOSS_DEPTH,
+        0,
+    };
+
+    /* try with successive loss */
+    i = 0;
+    for(;;) {
+        loss_mask = loss_mask_order[i++];
+        dst_pix_fmt = avcodec_find_best_pix_fmt1(pix_fmt_mask, src_pix_fmt,
+                                                 has_alpha, loss_mask);
+        if (dst_pix_fmt >= 0)
+            goto found;
+        if (loss_mask == 0)
+            break;
+    }
+    return -1;
+ found:
+    if (loss_ptr)
+        *loss_ptr = avcodec_get_pix_fmt_loss(dst_pix_fmt, src_pix_fmt, has_alpha);
+    return dst_pix_fmt;
+}
+
+void ff_img_copy_plane(uint8_t *dst, int dst_wrap,
+                           const uint8_t *src, int src_wrap,
+                           int width, int height)
+{
+    if((!dst) || (!src))
+        return;
+    for(;height > 0; height--) {
+        memcpy(dst, src, width);
+        dst += dst_wrap;
+        src += src_wrap;
+    }
+}
+
+void av_picture_copy(AVPicture *dst, const AVPicture *src,
+              int pix_fmt, int width, int height)
+{
+    int bwidth, bits, i;
+    const PixFmtInfo *pf = &pix_fmt_info[pix_fmt];
+
+    pf = &pix_fmt_info[pix_fmt];
+    switch(pf->pixel_type) {
+    case FF_PIXEL_PACKED:
+        switch(pix_fmt) {
+        case PIX_FMT_YUYV422:
+        case PIX_FMT_UYVY422:
+        case PIX_FMT_RGB565:
+        case PIX_FMT_RGB555:
+        case PIX_FMT_BGR565:
+        case PIX_FMT_BGR555:
+            bits = 16;
+            break;
+        case PIX_FMT_UYYVYY411:
+            bits = 12;
+            break;
+        default:
+            bits = pf->depth * pf->nb_channels;
+            break;
+        }
+        bwidth = (width * bits + 7) >> 3;
+        ff_img_copy_plane(dst->data[0], dst->linesize[0],
+                       src->data[0], src->linesize[0],
+                       bwidth, height);
+        break;
+    case FF_PIXEL_PLANAR:
+        for(i = 0; i < pf->nb_channels; i++) {
+            int w, h;
+            w = width;
+            h = height;
+            if (i == 1 || i == 2) {
+                w >>= pf->x_chroma_shift;
+                h >>= pf->y_chroma_shift;
+            }
+            bwidth = (w * pf->depth + 7) >> 3;
+            ff_img_copy_plane(dst->data[i], dst->linesize[i],
+                           src->data[i], src->linesize[i],
+                           bwidth, h);
+        }
+        break;
+    case FF_PIXEL_PALETTE:
+        ff_img_copy_plane(dst->data[0], dst->linesize[0],
+                       src->data[0], src->linesize[0],
+                       width, height);
+        /* copy the palette */
+        ff_img_copy_plane(dst->data[1], dst->linesize[1],
+                       src->data[1], src->linesize[1],
+                       4, 256);
+        break;
+    }
+}
+
+/* XXX: totally non optimized */
+
+static void yuyv422_to_yuv420p(AVPicture *dst, const AVPicture *src,
+                              int width, int height)
+{
+    const uint8_t *p, *p1;
+    uint8_t *lum, *cr, *cb, *lum1, *cr1, *cb1;
+    int w;
+
+    p1 = src->data[0];
+    lum1 = dst->data[0];
+    cb1 = dst->data[1];
+    cr1 = dst->data[2];
+
+    for(;height >= 1; height -= 2) {
+        p = p1;
+        lum = lum1;
+        cb = cb1;
+        cr = cr1;
+        for(w = width; w >= 2; w -= 2) {
+            lum[0] = p[0];
+            cb[0] = p[1];
+            lum[1] = p[2];
+            cr[0] = p[3];
+            p += 4;
+            lum += 2;
+            cb++;
+            cr++;
+        }
+        if (w) {
+            lum[0] = p[0];
+            cb[0] = p[1];
+            cr[0] = p[3];
+            cb++;
+            cr++;
+        }
+        p1 += src->linesize[0];
+        lum1 += dst->linesize[0];
+        if (height>1) {
+            p = p1;
+            lum = lum1;
+            for(w = width; w >= 2; w -= 2) {
+                lum[0] = p[0];
+                lum[1] = p[2];
+                p += 4;
+                lum += 2;
+            }
+            if (w) {
+                lum[0] = p[0];
+            }
+            p1 += src->linesize[0];
+            lum1 += dst->linesize[0];
+        }
+        cb1 += dst->linesize[1];
+        cr1 += dst->linesize[2];
+    }
+}
+
+static void uyvy422_to_yuv420p(AVPicture *dst, const AVPicture *src,
+                              int width, int height)
+{
+    const uint8_t *p, *p1;
+    uint8_t *lum, *cr, *cb, *lum1, *cr1, *cb1;
+    int w;
+
+    p1 = src->data[0];
+
+    lum1 = dst->data[0];
+    cb1 = dst->data[1];
+    cr1 = dst->data[2];
+
+    for(;height >= 1; height -= 2) {
+        p = p1;
+        lum = lum1;
+        cb = cb1;
+        cr = cr1;
+        for(w = width; w >= 2; w -= 2) {
+            lum[0] = p[1];
+            cb[0] = p[0];
+            lum[1] = p[3];
+            cr[0] = p[2];
+            p += 4;
+            lum += 2;
+            cb++;
+            cr++;
+        }
+        if (w) {
+            lum[0] = p[1];
+            cb[0] = p[0];
+            cr[0] = p[2];
+            cb++;
+            cr++;
+        }
+        p1 += src->linesize[0];
+        lum1 += dst->linesize[0];
+        if (height>1) {
+            p = p1;
+            lum = lum1;
+            for(w = width; w >= 2; w -= 2) {
+                lum[0] = p[1];
+                lum[1] = p[3];
+                p += 4;
+                lum += 2;
+            }
+            if (w) {
+                lum[0] = p[1];
+            }
+            p1 += src->linesize[0];
+            lum1 += dst->linesize[0];
+        }
+        cb1 += dst->linesize[1];
+        cr1 += dst->linesize[2];
+    }
+}
+
+
+static void uyvy422_to_yuv422p(AVPicture *dst, const AVPicture *src,
+                              int width, int height)
+{
+    const uint8_t *p, *p1;
+    uint8_t *lum, *cr, *cb, *lum1, *cr1, *cb1;
+    int w;
+
+    p1 = src->data[0];
+    lum1 = dst->data[0];
+    cb1 = dst->data[1];
+    cr1 = dst->data[2];
+    for(;height > 0; height--) {
+        p = p1;
+        lum = lum1;
+        cb = cb1;
+        cr = cr1;
+        for(w = width; w >= 2; w -= 2) {
+            lum[0] = p[1];
+            cb[0] = p[0];
+            lum[1] = p[3];
+            cr[0] = p[2];
+            p += 4;
+            lum += 2;
+            cb++;
+            cr++;
+        }
+        p1 += src->linesize[0];
+        lum1 += dst->linesize[0];
+        cb1 += dst->linesize[1];
+        cr1 += dst->linesize[2];
+    }
+}
+
+
+static void yuyv422_to_yuv422p(AVPicture *dst, const AVPicture *src,
+                              int width, int height)
+{
+    const uint8_t *p, *p1;
+    uint8_t *lum, *cr, *cb, *lum1, *cr1, *cb1;
+    int w;
+
+    p1 = src->data[0];
+    lum1 = dst->data[0];
+    cb1 = dst->data[1];
+    cr1 = dst->data[2];
+    for(;height > 0; height--) {
+        p = p1;
+        lum = lum1;
+        cb = cb1;
+        cr = cr1;
+        for(w = width; w >= 2; w -= 2) {
+            lum[0] = p[0];
+            cb[0] = p[1];
+            lum[1] = p[2];
+            cr[0] = p[3];
+            p += 4;
+            lum += 2;
+            cb++;
+            cr++;
+        }
+        p1 += src->linesize[0];
+        lum1 += dst->linesize[0];
+        cb1 += dst->linesize[1];
+        cr1 += dst->linesize[2];
+    }
+}
+
+static void yuv422p_to_yuyv422(AVPicture *dst, const AVPicture *src,
+                              int width, int height)
+{
+    uint8_t *p, *p1;
+    const uint8_t *lum, *cr, *cb, *lum1, *cr1, *cb1;
+    int w;
+
+    p1 = dst->data[0];
+    lum1 = src->data[0];
+    cb1 = src->data[1];
+    cr1 = src->data[2];
+    for(;height > 0; height--) {
+        p = p1;
+        lum = lum1;
+        cb = cb1;
+        cr = cr1;
+        for(w = width; w >= 2; w -= 2) {
+            p[0] = lum[0];
+            p[1] = cb[0];
+            p[2] = lum[1];
+            p[3] = cr[0];
+            p += 4;
+            lum += 2;
+            cb++;
+            cr++;
+        }
+        p1 += dst->linesize[0];
+        lum1 += src->linesize[0];
+        cb1 += src->linesize[1];
+        cr1 += src->linesize[2];
+    }
+}
+
+static void yuv422p_to_uyvy422(AVPicture *dst, const AVPicture *src,
+                              int width, int height)
+{
+    uint8_t *p, *p1;
+    const uint8_t *lum, *cr, *cb, *lum1, *cr1, *cb1;
+    int w;
+
+    p1 = dst->data[0];
+    lum1 = src->data[0];
+    cb1 = src->data[1];
+    cr1 = src->data[2];
+    for(;height > 0; height--) {
+        p = p1;
+        lum = lum1;
+        cb = cb1;
+        cr = cr1;
+        for(w = width; w >= 2; w -= 2) {
+            p[1] = lum[0];
+            p[0] = cb[0];
+            p[3] = lum[1];
+            p[2] = cr[0];
+            p += 4;
+            lum += 2;
+            cb++;
+            cr++;
+        }
+        p1 += dst->linesize[0];
+        lum1 += src->linesize[0];
+        cb1 += src->linesize[1];
+        cr1 += src->linesize[2];
+    }
+}
+
+static void uyyvyy411_to_yuv411p(AVPicture *dst, const AVPicture *src,
+                              int width, int height)
+{
+    const uint8_t *p, *p1;
+    uint8_t *lum, *cr, *cb, *lum1, *cr1, *cb1;
+    int w;
+
+    p1 = src->data[0];
+    lum1 = dst->data[0];
+    cb1 = dst->data[1];
+    cr1 = dst->data[2];
+    for(;height > 0; height--) {
+        p = p1;
+        lum = lum1;
+        cb = cb1;
+        cr = cr1;
+        for(w = width; w >= 4; w -= 4) {
+            cb[0] = p[0];
+            lum[0] = p[1];
+            lum[1] = p[2];
+            cr[0] = p[3];
+            lum[2] = p[4];
+            lum[3] = p[5];
+            p += 6;
+            lum += 4;
+            cb++;
+            cr++;
+        }
+        p1 += src->linesize[0];
+        lum1 += dst->linesize[0];
+        cb1 += dst->linesize[1];
+        cr1 += dst->linesize[2];
+    }
+}
+
+
+static void yuv420p_to_yuyv422(AVPicture *dst, const AVPicture *src,
+                              int width, int height)
+{
+    int w, h;
+    uint8_t *line1, *line2, *linesrc = dst->data[0];
+    uint8_t *lum1, *lum2, *lumsrc = src->data[0];
+    uint8_t *cb1, *cb2 = src->data[1];
+    uint8_t *cr1, *cr2 = src->data[2];
+
+    for(h = height / 2; h--;) {
+        line1 = linesrc;
+        line2 = linesrc + dst->linesize[0];
+
+        lum1 = lumsrc;
+        lum2 = lumsrc + src->linesize[0];
+
+        cb1 = cb2;
+        cr1 = cr2;
+
+        for(w = width / 2; w--;) {
+                *line1++ = *lum1++; *line2++ = *lum2++;
+                *line1++ =          *line2++ = *cb1++;
+                *line1++ = *lum1++; *line2++ = *lum2++;
+                *line1++ =          *line2++ = *cr1++;
+        }
+
+        linesrc += dst->linesize[0] * 2;
+        lumsrc += src->linesize[0] * 2;
+        cb2 += src->linesize[1];
+        cr2 += src->linesize[2];
+    }
+}
+
+static void yuv420p_to_uyvy422(AVPicture *dst, const AVPicture *src,
+                              int width, int height)
+{
+    int w, h;
+    uint8_t *line1, *line2, *linesrc = dst->data[0];
+    uint8_t *lum1, *lum2, *lumsrc = src->data[0];
+    uint8_t *cb1, *cb2 = src->data[1];
+    uint8_t *cr1, *cr2 = src->data[2];
+
+    for(h = height / 2; h--;) {
+        line1 = linesrc;
+        line2 = linesrc + dst->linesize[0];
+
+        lum1 = lumsrc;
+        lum2 = lumsrc + src->linesize[0];
+
+        cb1 = cb2;
+        cr1 = cr2;
+
+        for(w = width / 2; w--;) {
+                *line1++ =          *line2++ = *cb1++;
+                *line1++ = *lum1++; *line2++ = *lum2++;
+                *line1++ =          *line2++ = *cr1++;
+                *line1++ = *lum1++; *line2++ = *lum2++;
+        }
+
+        linesrc += dst->linesize[0] * 2;
+        lumsrc += src->linesize[0] * 2;
+        cb2 += src->linesize[1];
+        cr2 += src->linesize[2];
+    }
+}
+
+#define SCALEBITS 10
+#define ONE_HALF  (1 << (SCALEBITS - 1))
+#define FIX(x)    ((int) ((x) * (1<<SCALEBITS) + 0.5))
+
+#define YUV_TO_RGB1_CCIR(cb1, cr1)\
+{\
+    cb = (cb1) - 128;\
+    cr = (cr1) - 128;\
+    r_add = FIX(1.40200*255.0/224.0) * cr + ONE_HALF;\
+    g_add = - FIX(0.34414*255.0/224.0) * cb - FIX(0.71414*255.0/224.0) * cr + \
+            ONE_HALF;\
+    b_add = FIX(1.77200*255.0/224.0) * cb + ONE_HALF;\
+}
+
+#define YUV_TO_RGB2_CCIR(r, g, b, y1)\
+{\
+    y = ((y1) - 16) * FIX(255.0/219.0);\
+    r = cm[(y + r_add) >> SCALEBITS];\
+    g = cm[(y + g_add) >> SCALEBITS];\
+    b = cm[(y + b_add) >> SCALEBITS];\
+}
+
+#define YUV_TO_RGB1(cb1, cr1)\
+{\
+    cb = (cb1) - 128;\
+    cr = (cr1) - 128;\
+    r_add = FIX(1.40200) * cr + ONE_HALF;\
+    g_add = - FIX(0.34414) * cb - FIX(0.71414) * cr + ONE_HALF;\
+    b_add = FIX(1.77200) * cb + ONE_HALF;\
+}
+
+#define YUV_TO_RGB2(r, g, b, y1)\
+{\
+    y = (y1) << SCALEBITS;\
+    r = cm[(y + r_add) >> SCALEBITS];\
+    g = cm[(y + g_add) >> SCALEBITS];\
+    b = cm[(y + b_add) >> SCALEBITS];\
+}
+
+#define Y_CCIR_TO_JPEG(y)\
+ cm[((y) * FIX(255.0/219.0) + (ONE_HALF - 16 * FIX(255.0/219.0))) >> SCALEBITS]
+
+#define Y_JPEG_TO_CCIR(y)\
+ (((y) * FIX(219.0/255.0) + (ONE_HALF + (16 << SCALEBITS))) >> SCALEBITS)
+
+#define C_CCIR_TO_JPEG(y)\
+ cm[(((y) - 128) * FIX(127.0/112.0) + (ONE_HALF + (128 << SCALEBITS))) >> SCALEBITS]
+
+/* NOTE: the clamp is really necessary! */
+static inline int C_JPEG_TO_CCIR(int y) {
+    y = (((y - 128) * FIX(112.0/127.0) + (ONE_HALF + (128 << SCALEBITS))) >> SCALEBITS);
+    if (y < 16)
+        y = 16;
+    return y;
+}
+
+
+#define RGB_TO_Y(r, g, b) \
+((FIX(0.29900) * (r) + FIX(0.58700) * (g) + \
+  FIX(0.11400) * (b) + ONE_HALF) >> SCALEBITS)
+
+#define RGB_TO_U(r1, g1, b1, shift)\
+(((- FIX(0.16874) * r1 - FIX(0.33126) * g1 +         \
+     FIX(0.50000) * b1 + (ONE_HALF << shift) - 1) >> (SCALEBITS + shift)) + 128)
+
+#define RGB_TO_V(r1, g1, b1, shift)\
+(((FIX(0.50000) * r1 - FIX(0.41869) * g1 -           \
+   FIX(0.08131) * b1 + (ONE_HALF << shift) - 1) >> (SCALEBITS + shift)) + 128)
+
+#define RGB_TO_Y_CCIR(r, g, b) \
+((FIX(0.29900*219.0/255.0) * (r) + FIX(0.58700*219.0/255.0) * (g) + \
+  FIX(0.11400*219.0/255.0) * (b) + (ONE_HALF + (16 << SCALEBITS))) >> SCALEBITS)
+
+#define RGB_TO_U_CCIR(r1, g1, b1, shift)\
+(((- FIX(0.16874*224.0/255.0) * r1 - FIX(0.33126*224.0/255.0) * g1 +         \
+     FIX(0.50000*224.0/255.0) * b1 + (ONE_HALF << shift) - 1) >> (SCALEBITS + shift)) + 128)
+
+#define RGB_TO_V_CCIR(r1, g1, b1, shift)\
+(((FIX(0.50000*224.0/255.0) * r1 - FIX(0.41869*224.0/255.0) * g1 -           \
+   FIX(0.08131*224.0/255.0) * b1 + (ONE_HALF << shift) - 1) >> (SCALEBITS + shift)) + 128)
+
+static uint8_t y_ccir_to_jpeg[256];
+static uint8_t y_jpeg_to_ccir[256];
+static uint8_t c_ccir_to_jpeg[256];
+static uint8_t c_jpeg_to_ccir[256];
+
+/* init various conversion tables */
+static void img_convert_init(void)
+{
+    int i;
+    uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;
+
+    for(i = 0;i < 256; i++) {
+        y_ccir_to_jpeg[i] = Y_CCIR_TO_JPEG(i);
+        y_jpeg_to_ccir[i] = Y_JPEG_TO_CCIR(i);
+        c_ccir_to_jpeg[i] = C_CCIR_TO_JPEG(i);
+        c_jpeg_to_ccir[i] = C_JPEG_TO_CCIR(i);
+    }
+}
+
+/* apply to each pixel the given table */
+static void img_apply_table(uint8_t *dst, int dst_wrap,
+                            const uint8_t *src, int src_wrap,
+                            int width, int height, const uint8_t *table1)
+{
+    int n;
+    const uint8_t *s;
+    uint8_t *d;
+    const uint8_t *table;
+
+    table = table1;
+    for(;height > 0; height--) {
+        s = src;
+        d = dst;
+        n = width;
+        while (n >= 4) {
+            d[0] = table[s[0]];
+            d[1] = table[s[1]];
+            d[2] = table[s[2]];
+            d[3] = table[s[3]];
+            d += 4;
+            s += 4;
+            n -= 4;
+        }
+        while (n > 0) {
+            d[0] = table[s[0]];
+            d++;
+            s++;
+            n--;
+        }
+        dst += dst_wrap;
+        src += src_wrap;
+    }
+}
+
+/* XXX: use generic filter ? */
+/* XXX: in most cases, the sampling position is incorrect */
+
+/* 4x1 -> 1x1 */
+static void shrink41(uint8_t *dst, int dst_wrap,
+                     const uint8_t *src, int src_wrap,
+                     int width, int height)
+{
+    int w;
+    const uint8_t *s;
+    uint8_t *d;
+
+    for(;height > 0; height--) {
+        s = src;
+        d = dst;
+        for(w = width;w > 0; w--) {
+            d[0] = (s[0] + s[1] + s[2] + s[3] + 2) >> 2;
+            s += 4;
+            d++;
+        }
+        src += src_wrap;
+        dst += dst_wrap;
+    }
+}
+
+/* 2x1 -> 1x1 */
+static void shrink21(uint8_t *dst, int dst_wrap,
+                     const uint8_t *src, int src_wrap,
+                     int width, int height)
+{
+    int w;
+    const uint8_t *s;
+    uint8_t *d;
+
+    for(;height > 0; height--) {
+        s = src;
+        d = dst;
+        for(w = width;w > 0; w--) {
+            d[0] = (s[0] + s[1]) >> 1;
+            s += 2;
+            d++;
+        }
+        src += src_wrap;
+        dst += dst_wrap;
+    }
+}
+
+/* 1x2 -> 1x1 */
+static void shrink12(uint8_t *dst, int dst_wrap,
+                     const uint8_t *src, int src_wrap,
+                     int width, int height)
+{
+    int w;
+    uint8_t *d;
+    const uint8_t *s1, *s2;
+
+    for(;height > 0; height--) {
+        s1 = src;
+        s2 = s1 + src_wrap;
+        d = dst;
+        for(w = width;w >= 4; w-=4) {
+            d[0] = (s1[0] + s2[0]) >> 1;
+            d[1] = (s1[1] + s2[1]) >> 1;
+            d[2] = (s1[2] + s2[2]) >> 1;
+            d[3] = (s1[3] + s2[3]) >> 1;
+            s1 += 4;
+            s2 += 4;
+            d += 4;
+        }
+        for(;w > 0; w--) {
+            d[0] = (s1[0] + s2[0]) >> 1;
+            s1++;
+            s2++;
+            d++;
+        }
+        src += 2 * src_wrap;
+        dst += dst_wrap;
+    }
+}
+
+/* 2x2 -> 1x1 */
+void ff_shrink22(uint8_t *dst, int dst_wrap,
+                     const uint8_t *src, int src_wrap,
+                     int width, int height)
+{
+    int w;
+    const uint8_t *s1, *s2;
+    uint8_t *d;
+
+    for(;height > 0; height--) {
+        s1 = src;
+        s2 = s1 + src_wrap;
+        d = dst;
+        for(w = width;w >= 4; w-=4) {
+            d[0] = (s1[0] + s1[1] + s2[0] + s2[1] + 2) >> 2;
+            d[1] = (s1[2] + s1[3] + s2[2] + s2[3] + 2) >> 2;
+            d[2] = (s1[4] + s1[5] + s2[4] + s2[5] + 2) >> 2;
+            d[3] = (s1[6] + s1[7] + s2[6] + s2[7] + 2) >> 2;
+            s1 += 8;
+            s2 += 8;
+            d += 4;
+        }
+        for(;w > 0; w--) {
+            d[0] = (s1[0] + s1[1] + s2[0] + s2[1] + 2) >> 2;
+            s1 += 2;
+            s2 += 2;
+            d++;
+        }
+        src += 2 * src_wrap;
+        dst += dst_wrap;
+    }
+}
+
+/* 4x4 -> 1x1 */
+void ff_shrink44(uint8_t *dst, int dst_wrap,
+                     const uint8_t *src, int src_wrap,
+                     int width, int height)
+{
+    int w;
+    const uint8_t *s1, *s2, *s3, *s4;
+    uint8_t *d;
+
+    for(;height > 0; height--) {
+        s1 = src;
+        s2 = s1 + src_wrap;
+        s3 = s2 + src_wrap;
+        s4 = s3 + src_wrap;
+        d = dst;
+        for(w = width;w > 0; w--) {
+            d[0] = (s1[0] + s1[1] + s1[2] + s1[3] +
+                    s2[0] + s2[1] + s2[2] + s2[3] +
+                    s3[0] + s3[1] + s3[2] + s3[3] +
+                    s4[0] + s4[1] + s4[2] + s4[3] + 8) >> 4;
+            s1 += 4;
+            s2 += 4;
+            s3 += 4;
+            s4 += 4;
+            d++;
+        }
+        src += 4 * src_wrap;
+        dst += dst_wrap;
+    }
+}
+
+/* 8x8 -> 1x1 */
+void ff_shrink88(uint8_t *dst, int dst_wrap,
+                     const uint8_t *src, int src_wrap,
+                     int width, int height)
+{
+    int w, i;
+
+    for(;height > 0; height--) {
+        for(w = width;w > 0; w--) {
+            int tmp=0;
+            for(i=0; i<8; i++){
+                tmp += src[0] + src[1] + src[2] + src[3] + src[4] + src[5] + src[6] + src[7];
+                src += src_wrap;
+            }
+            *(dst++) = (tmp + 32)>>6;
+            src += 8 - 8*src_wrap;
+        }
+        src += 8*src_wrap - 8*width;
+        dst += dst_wrap - width;
+    }
+}
+
+static void grow21_line(uint8_t *dst, const uint8_t *src,
+                        int width)
+{
+    int w;
+    const uint8_t *s1;
+    uint8_t *d;
+
+    s1 = src;
+    d = dst;
+    for(w = width;w >= 4; w-=4) {
+        d[1] = d[0] = s1[0];
+        d[3] = d[2] = s1[1];
+        s1 += 2;
+        d += 4;
+    }
+    for(;w >= 2; w -= 2) {
+        d[1] = d[0] = s1[0];
+        s1 ++;
+        d += 2;
+    }
+    /* only needed if width is not a multiple of two */
+    /* XXX: veryfy that */
+    if (w) {
+        d[0] = s1[0];
+    }
+}
+
+static void grow41_line(uint8_t *dst, const uint8_t *src,
+                        int width)
+{
+    int w, v;
+    const uint8_t *s1;
+    uint8_t *d;
+
+    s1 = src;
+    d = dst;
+    for(w = width;w >= 4; w-=4) {
+        v = s1[0];
+        d[0] = v;
+        d[1] = v;
+        d[2] = v;
+        d[3] = v;
+        s1 ++;
+        d += 4;
+    }
+}
+
+/* 1x1 -> 2x1 */
+static void grow21(uint8_t *dst, int dst_wrap,
+                   const uint8_t *src, int src_wrap,
+                   int width, int height)
+{
+    for(;height > 0; height--) {
+        grow21_line(dst, src, width);
+        src += src_wrap;
+        dst += dst_wrap;
+    }
+}
+
+/* 1x1 -> 2x2 */
+static void grow22(uint8_t *dst, int dst_wrap,
+                   const uint8_t *src, int src_wrap,
+                   int width, int height)
+{
+    for(;height > 0; height--) {
+        grow21_line(dst, src, width);
+        if (height%2)
+            src += src_wrap;
+        dst += dst_wrap;
+    }
+}
+
+/* 1x1 -> 4x1 */
+static void grow41(uint8_t *dst, int dst_wrap,
+                   const uint8_t *src, int src_wrap,
+                   int width, int height)
+{
+    for(;height > 0; height--) {
+        grow41_line(dst, src, width);
+        src += src_wrap;
+        dst += dst_wrap;
+    }
+}
+
+/* 1x1 -> 4x4 */
+static void grow44(uint8_t *dst, int dst_wrap,
+                   const uint8_t *src, int src_wrap,
+                   int width, int height)
+{
+    for(;height > 0; height--) {
+        grow41_line(dst, src, width);
+        if ((height & 3) == 1)
+            src += src_wrap;
+        dst += dst_wrap;
+    }
+}
+
+/* 1x2 -> 2x1 */
+static void conv411(uint8_t *dst, int dst_wrap,
+                    const uint8_t *src, int src_wrap,
+                    int width, int height)
+{
+    int w, c;
+    const uint8_t *s1, *s2;
+    uint8_t *d;
+
+    width>>=1;
+
+    for(;height > 0; height--) {
+        s1 = src;
+        s2 = src + src_wrap;
+        d = dst;
+        for(w = width;w > 0; w--) {
+            c = (s1[0] + s2[0]) >> 1;
+            d[0] = c;
+            d[1] = c;
+            s1++;
+            s2++;
+            d += 2;
+        }
+        src += src_wrap * 2;
+        dst += dst_wrap;
+    }
+}
+
+/* XXX: add jpeg quantize code */
+
+#define TRANSP_INDEX (6*6*6)
+
+/* this is maybe slow, but allows for extensions */
+static inline unsigned char gif_clut_index(uint8_t r, uint8_t g, uint8_t b)
+{
+    return ((((r)/47)%6)*6*6+(((g)/47)%6)*6+(((b)/47)%6));
+}
+
+static void build_rgb_palette(uint8_t *palette, int has_alpha)
+{
+    uint32_t *pal;
+    static const uint8_t pal_value[6] = { 0x00, 0x33, 0x66, 0x99, 0xcc, 0xff };
+    int i, r, g, b;
+
+    pal = (uint32_t *)palette;
+    i = 0;
+    for(r = 0; r < 6; r++) {
+        for(g = 0; g < 6; g++) {
+            for(b = 0; b < 6; b++) {
+                pal[i++] = (0xff << 24) | (pal_value[r] << 16) |
+                    (pal_value[g] << 8) | pal_value[b];
+            }
+        }
+    }
+    if (has_alpha)
+        pal[i++] = 0;
+    while (i < 256)
+        pal[i++] = 0xff000000;
+}
+
+/* copy bit n to bits 0 ... n - 1 */
+static inline unsigned int bitcopy_n(unsigned int a, int n)
+{
+    int mask;
+    mask = (1 << n) - 1;
+    return (a & (0xff & ~mask)) | ((-((a >> n) & 1)) & mask);
+}
+
+/* rgb555 handling */
+
+#define RGB_NAME rgb555
+
+#define RGB_IN(r, g, b, s)\
+{\
+    unsigned int v = ((const uint16_t *)(s))[0];\
+    r = bitcopy_n(v >> (10 - 3), 3);\
+    g = bitcopy_n(v >> (5 - 3), 3);\
+    b = bitcopy_n(v << 3, 3);\
+}
+
+
+#define RGB_OUT(d, r, g, b)\
+{\
+    ((uint16_t *)(d))[0] = ((r >> 3) << 10) | ((g >> 3) << 5) | (b >> 3);\
+}
+
+#define BPP 2
+
+#include "imgconvert_template.h"
+
+/* rgb565 handling */
+
+#define RGB_NAME rgb565
+
+#define RGB_IN(r, g, b, s)\
+{\
+    unsigned int v = ((const uint16_t *)(s))[0];\
+    r = bitcopy_n(v >> (11 - 3), 3);\
+    g = bitcopy_n(v >> (5 - 2), 2);\
+    b = bitcopy_n(v << 3, 3);\
+}
+
+#define RGB_OUT(d, r, g, b)\
+{\
+    ((uint16_t *)(d))[0] = ((r >> 3) << 11) | ((g >> 2) << 5) | (b >> 3);\
+}
+
+#define BPP 2
+
+#include "imgconvert_template.h"
+
+/* bgr24 handling */
+
+#define RGB_NAME bgr24
+
+#define RGB_IN(r, g, b, s)\
+{\
+    b = (s)[0];\
+    g = (s)[1];\
+    r = (s)[2];\
+}
+
+#define RGB_OUT(d, r, g, b)\
+{\
+    (d)[0] = b;\
+    (d)[1] = g;\
+    (d)[2] = r;\
+}
+
+#define BPP 3
+
+#include "imgconvert_template.h"
+
+#undef RGB_IN
+#undef RGB_OUT
+#undef BPP
+
+/* rgb24 handling */
+
+#define RGB_NAME rgb24
+#define FMT_RGB24
+
+#define RGB_IN(r, g, b, s)\
+{\
+    r = (s)[0];\
+    g = (s)[1];\
+    b = (s)[2];\
+}
+
+#define RGB_OUT(d, r, g, b)\
+{\
+    (d)[0] = r;\
+    (d)[1] = g;\
+    (d)[2] = b;\
+}
+
+#define BPP 3
+
+#include "imgconvert_template.h"
+
+/* rgb32 handling */
+
+#define RGB_NAME rgb32
+#define FMT_RGB32
+
+#define RGB_IN(r, g, b, s)\
+{\
+    unsigned int v = ((const uint32_t *)(s))[0];\
+    r = (v >> 16) & 0xff;\
+    g = (v >> 8) & 0xff;\
+    b = v & 0xff;\
+}
+
+#define RGBA_IN(r, g, b, a, s)\
+{\
+    unsigned int v = ((const uint32_t *)(s))[0];\
+    a = (v >> 24) & 0xff;\
+    r = (v >> 16) & 0xff;\
+    g = (v >> 8) & 0xff;\
+    b = v & 0xff;\
+}
+
+#define RGBA_OUT(d, r, g, b, a)\
+{\
+    ((uint32_t *)(d))[0] = (a << 24) | (r << 16) | (g << 8) | b;\
+}
 
-#ifdef USE_FASTMEMCPY
-#include "fastmemcpy.h"
-#endif
-/* XXX: totally non optimized */
+#define BPP 4
+
+#include "imgconvert_template.h"
 
-static void yuv422_to_yuv420p(UINT8 *lum, UINT8 *cb, UINT8 *cr,
-                              UINT8 *src, int width, int height)
+static void mono_to_gray(AVPicture *dst, const AVPicture *src,
+                         int width, int height, int xor_mask)
 {
-    int x, y;
-    UINT8 *p = src;
+    const unsigned char *p;
+    unsigned char *q;
+    int v, dst_wrap, src_wrap;
+    int y, w;
 
-    for(y=0;y<height;y+=2) {
-        for(x=0;x<width;x+=2) {
-            lum[0] = p[0];
-            cb[0] = p[1];
-            lum[1] = p[2];
-            cr[0] = p[3];
-            p += 4;
-            lum += 2;
-            cb++;
-            cr++;
+    p = src->data[0];
+    src_wrap = src->linesize[0] - ((width + 7) >> 3);
+
+    q = dst->data[0];
+    dst_wrap = dst->linesize[0] - width;
+    for(y=0;y<height;y++) {
+        w = width;
+        while (w >= 8) {
+            v = *p++ ^ xor_mask;
+            q[0] = -(v >> 7);
+            q[1] = -((v >> 6) & 1);
+            q[2] = -((v >> 5) & 1);
+            q[3] = -((v >> 4) & 1);
+            q[4] = -((v >> 3) & 1);
+            q[5] = -((v >> 2) & 1);
+            q[6] = -((v >> 1) & 1);
+            q[7] = -((v >> 0) & 1);
+            w -= 8;
+            q += 8;
         }
-        for(x=0;x<width;x+=2) {
-            lum[0] = p[0];
-            lum[1] = p[2];
-            p += 4;
-            lum += 2;
+        if (w > 0) {
+            v = *p++ ^ xor_mask;
+            do {
+                q[0] = -((v >> 7) & 1);
+                q++;
+                v <<= 1;
+            } while (--w);
         }
+        p += src_wrap;
+        q += dst_wrap;
     }
 }
 
-#define SCALEBITS 8
-#define ONE_HALF  (1 << (SCALEBITS - 1))
-#define FIX(x)         ((int) ((x) * (1L<<SCALEBITS) + 0.5))
-
-static void rgb24_to_yuv420p(UINT8 *lum, UINT8 *cb, UINT8 *cr,
-                              UINT8 *src, int width, int height)
-{
-    int wrap, wrap3, x, y;
-    int r, g, b, r1, g1, b1;
-    UINT8 *p;
-
-    wrap = width;
-    wrap3 = width * 3;
-    p = src;
-    for(y=0;y<height;y+=2) {
-        for(x=0;x<width;x+=2) {
-            r = p[0];
-            g = p[1];
-            b = p[2];
-            r1 = r;
-            g1 = g;
-            b1 = b;
-            lum[0] = (FIX(0.29900) * r + FIX(0.58700) * g + 
-                      FIX(0.11400) * b + ONE_HALF) >> SCALEBITS;
-            r = p[3];
-            g = p[4];
-            b = p[5];
-            r1 += r;
-            g1 += g;
-            b1 += b;
-            lum[1] = (FIX(0.29900) * r + FIX(0.58700) * g + 
-                      FIX(0.11400) * b + ONE_HALF) >> SCALEBITS;
-            p += wrap3;
-            lum += wrap;
-
-            r = p[0];
-            g = p[1];
-            b = p[2];
-            r1 += r;
-            g1 += g;
-            b1 += b;
-            lum[0] = (FIX(0.29900) * r + FIX(0.58700) * g + 
-                      FIX(0.11400) * b + ONE_HALF) >> SCALEBITS;
-            r = p[3];
-            g = p[4];
-            b = p[5];
-            r1 += r;
-            g1 += g;
-            b1 += b;
-            lum[1] = (FIX(0.29900) * r + FIX(0.58700) * g + 
-                      FIX(0.11400) * b + ONE_HALF) >> SCALEBITS;
-            
-            cb[0] = ((- FIX(0.16874) * r1 - FIX(0.33126) * g1 + 
-                      FIX(0.50000) * b1 + 4 * ONE_HALF - 1) >> (SCALEBITS + 2)) + 128;
-            cr[0] = ((FIX(0.50000) * r1 - FIX(0.41869) * g1 - 
-                     FIX(0.08131) * b1 + 4 * ONE_HALF - 1) >> (SCALEBITS + 2)) + 128;
-
-            cb++;
-            cr++;
-            p += -wrap3 + 2 * 3;
-            lum += -wrap + 2;
-        }
-        p += wrap3;
-        lum += wrap;
-    }
-}
-
-static void bgr24_to_yuv420p(UINT8 *lum, UINT8 *cb, UINT8 *cr,
-                              UINT8 *src, int width, int height)
-{
-    int wrap, wrap3, x, y;
-    int r, g, b, r1, g1, b1;
-    UINT8 *p;
-
-    wrap = width;
-    wrap3 = width * 3;
-    p = src;
-    for(y=0;y<height;y+=2) {
-        for(x=0;x<width;x+=2) {
-            b = p[0];
-            g = p[1];
-            r = p[2];
-            r1 = r;
-            g1 = g;
-            b1 = b;
-            lum[0] = (FIX(0.29900) * r + FIX(0.58700) * g + 
-                      FIX(0.11400) * b + ONE_HALF) >> SCALEBITS;
-            b = p[3];
-            g = p[4];
-            r = p[5];
-            r1 += r;
-            g1 += g;
-            b1 += b;
-            lum[1] = (FIX(0.29900) * r + FIX(0.58700) * g + 
-                      FIX(0.11400) * b + ONE_HALF) >> SCALEBITS;
-            p += wrap3;
-            lum += wrap;
-
-            b = p[0];
-            g = p[1];
-            r = p[2];
-            r1 += r;
-            g1 += g;
-            b1 += b;
-            lum[0] = (FIX(0.29900) * r + FIX(0.58700) * g + 
-                      FIX(0.11400) * b + ONE_HALF) >> SCALEBITS;
-            b = p[3];
-            g = p[4];
-            r = p[5];
-            r1 += r;
-            g1 += g;
-            b1 += b;
-            lum[1] = (FIX(0.29900) * r + FIX(0.58700) * g + 
-                      FIX(0.11400) * b + ONE_HALF) >> SCALEBITS;
-            
-            cb[0] = ((- FIX(0.16874) * r1 - FIX(0.33126) * g1 + 
-                      FIX(0.50000) * b1 + 4 * ONE_HALF - 1) >> (SCALEBITS + 2)) + 128;
-            cr[0] = ((FIX(0.50000) * r1 - FIX(0.41869) * g1 - 
-                     FIX(0.08131) * b1 + 4 * ONE_HALF - 1) >> (SCALEBITS + 2)) + 128;
+static void monowhite_to_gray(AVPicture *dst, const AVPicture *src,
+                               int width, int height)
+{
+    mono_to_gray(dst, src, width, height, 0xff);
+}
 
-            cb++;
-            cr++;
-            p += -wrap3 + 2 * 3;
-            lum += -wrap + 2;
-        }
-        p += wrap3;
-        lum += wrap;
-    }
+static void monoblack_to_gray(AVPicture *dst, const AVPicture *src,
+                               int width, int height)
+{
+    mono_to_gray(dst, src, width, height, 0x00);
 }
 
-/* XXX: use generic filter ? */
-/* 1x2 -> 1x1 */
-static void shrink2(UINT8 *dst, int dst_wrap, 
-                    UINT8 *src, int src_wrap,
-                    int width, int height)
+static void gray_to_mono(AVPicture *dst, const AVPicture *src,
+                         int width, int height, int xor_mask)
 {
-    int w;
-    UINT8 *s1, *s2, *d;
+    int n;
+    const uint8_t *s;
+    uint8_t *d;
+    int j, b, v, n1, src_wrap, dst_wrap, y;
 
-    for(;height > 0; height--) {
-        s1 = src;
-        s2 = s1 + src_wrap;
-        d = dst;
-        for(w = width;w >= 4; w-=4) {
-            d[0] = (s1[0] + s2[0]) >> 1;
-            d[1] = (s1[1] + s2[1]) >> 1;
-            d[2] = (s1[2] + s2[2]) >> 1;
-            d[3] = (s1[3] + s2[3]) >> 1;
-            s1 += 4;
-            s2 += 4;
-            d += 4;
+    s = src->data[0];
+    src_wrap = src->linesize[0] - width;
+
+    d = dst->data[0];
+    dst_wrap = dst->linesize[0] - ((width + 7) >> 3);
+
+    for(y=0;y<height;y++) {
+        n = width;
+        while (n >= 8) {
+            v = 0;
+            for(j=0;j<8;j++) {
+                b = s[0];
+                s++;
+                v = (v << 1) | (b >> 7);
+            }
+            d[0] = v ^ xor_mask;
+            d++;
+            n -= 8;
         }
-        for(;w > 0; w--) {
-            d[0] = (s1[0] + s2[0]) >> 1;
-            s1++;
-            s2++;
+        if (n > 0) {
+            n1 = n;
+            v = 0;
+            while (n > 0) {
+                b = s[0];
+                s++;
+                v = (v << 1) | (b >> 7);
+                n--;
+            }
+            d[0] = (v << (8 - (n1 & 7))) ^ xor_mask;
             d++;
         }
-        src += 2 * src_wrap;
-        dst += dst_wrap;
+        s += src_wrap;
+        d += dst_wrap;
     }
 }
 
-/* 2x2 -> 1x1 */
-static void shrink22(UINT8 *dst, int dst_wrap, 
-                     UINT8 *src, int src_wrap,
-                     int width, int height)
+static void gray_to_monowhite(AVPicture *dst, const AVPicture *src,
+                              int width, int height)
 {
-    int w;
-    UINT8 *s1, *s2, *d;
+    gray_to_mono(dst, src, width, height, 0xff);
+}
 
-    for(;height > 0; height--) {
-        s1 = src;
-        s2 = s1 + src_wrap;
-        d = dst;
-        for(w = width;w >= 4; w-=4) {
-            d[0] = (s1[0] + s1[1] + s2[0] + s2[1] + 2) >> 1;
-            d[1] = (s1[2] + s1[3] + s2[2] + s2[3] + 2) >> 1;
-            d[2] = (s1[4] + s1[5] + s2[4] + s2[5] + 2) >> 1;
-            d[3] = (s1[6] + s1[7] + s2[6] + s2[7] + 2) >> 1;
-            s1 += 8;
-            s2 += 8;
-            d += 4;
+static void gray_to_monoblack(AVPicture *dst, const AVPicture *src,
+                              int width, int height)
+{
+    gray_to_mono(dst, src, width, height, 0x00);
+}
+
+static void gray_to_gray16(AVPicture *dst, const AVPicture *src,
+                              int width, int height)
+{
+    int x, y, src_wrap, dst_wrap;
+    uint8_t *s, *d;
+    s = src->data[0];
+    src_wrap = src->linesize[0] - width;
+    d = dst->data[0];
+    dst_wrap = dst->linesize[0] - width * 2;
+    for(y=0; y<height; y++){
+        for(x=0; x<width; x++){
+            *d++ = *s;
+            *d++ = *s++;
         }
-        for(;w > 0; w--) {
-            d[0] = (s1[0] + s1[1] + s2[0] + s2[1] + 2) >> 1;
-            s1 += 2;
-            s2 += 2;
-            d++;
+        s += src_wrap;
+        d += dst_wrap;
+    }
+}
+
+static void gray16_to_gray(AVPicture *dst, const AVPicture *src,
+                              int width, int height)
+{
+    int x, y, src_wrap, dst_wrap;
+    uint8_t *s, *d;
+    s = src->data[0];
+    src_wrap = src->linesize[0] - width * 2;
+    d = dst->data[0];
+    dst_wrap = dst->linesize[0] - width;
+    for(y=0; y<height; y++){
+        for(x=0; x<width; x++){
+            *d++ = *s;
+            s += 2;
         }
-        src += 2 * src_wrap;
-        dst += dst_wrap;
+        s += src_wrap;
+        d += dst_wrap;
     }
 }
 
-static void img_copy(UINT8 *dst, int dst_wrap, 
-                     UINT8 *src, int src_wrap,
-                     int width, int height)
+static void gray16be_to_gray(AVPicture *dst, const AVPicture *src,
+                              int width, int height)
 {
-    for(;height > 0; height--) {
-        memcpy(dst, src, width);
-        dst += dst_wrap;
-        src += src_wrap;
+    gray16_to_gray(dst, src, width, height);
+}
+
+static void gray16le_to_gray(AVPicture *dst, const AVPicture *src,
+                              int width, int height)
+{
+    gray16_to_gray(dst, src + 1, width, height);
+}
+
+static void gray16_to_gray16(AVPicture *dst, const AVPicture *src,
+                              int width, int height)
+{
+    int x, y, src_wrap, dst_wrap;
+    uint16_t *s, *d;
+    s = src->data[0];
+    src_wrap = (src->linesize[0] - width * 2)/2;
+    d = dst->data[0];
+    dst_wrap = (dst->linesize[0] - width * 2)/2;
+    for(y=0; y<height; y++){
+        for(x=0; x<width; x++){
+            *d++ = bswap_16(*s++);
+        }
+        s += src_wrap;
+        d += dst_wrap;
     }
 }
 
-#define SCALE_BITS 10
 
-#define C_Y  (76309 >> (16 - SCALE_BITS))
-#define C_RV (117504 >> (16 - SCALE_BITS))
-#define C_BU (138453 >> (16 - SCALE_BITS))
-#define C_GU (13954 >> (16 - SCALE_BITS))
-#define C_GV (34903 >> (16 - SCALE_BITS))
+typedef struct ConvertEntry {
+    void (*convert)(AVPicture *dst,
+                    const AVPicture *src, int width, int height);
+} ConvertEntry;
 
-#define RGBOUT(r, g, b, y1)\
-{\
-    y = (y1 - 16) * C_Y;\
-    r = cm[(y + r_add) >> SCALE_BITS];\
-    g = cm[(y + g_add) >> SCALE_BITS];\
-    b = cm[(y + b_add) >> SCALE_BITS];\
+/* Add each new convertion function in this table. In order to be able
+   to convert from any format to any format, the following constraints
+   must be satisfied:
+
+   - all FF_COLOR_RGB formats must convert to and from PIX_FMT_RGB24
+
+   - all FF_COLOR_GRAY formats must convert to and from PIX_FMT_GRAY8
+
+   - all FF_COLOR_RGB formats with alpha must convert to and from PIX_FMT_RGB32
+
+   - PIX_FMT_YUV444P and PIX_FMT_YUVJ444P must convert to and from
+     PIX_FMT_RGB24.
+
+   - PIX_FMT_422 must convert to and from PIX_FMT_422P.
+
+   The other conversion functions are just optimisations for common cases.
+*/
+static const ConvertEntry convert_table[PIX_FMT_NB][PIX_FMT_NB] = {
+    [PIX_FMT_YUV420P] = {
+        [PIX_FMT_YUYV422] = {
+            .convert = yuv420p_to_yuyv422,
+        },
+        [PIX_FMT_RGB555] = {
+            .convert = yuv420p_to_rgb555
+        },
+        [PIX_FMT_RGB565] = {
+            .convert = yuv420p_to_rgb565
+        },
+        [PIX_FMT_BGR24] = {
+            .convert = yuv420p_to_bgr24
+        },
+        [PIX_FMT_RGB24] = {
+            .convert = yuv420p_to_rgb24
+        },
+        [PIX_FMT_RGB32] = {
+            .convert = yuv420p_to_rgb32
+        },
+        [PIX_FMT_UYVY422] = {
+            .convert = yuv420p_to_uyvy422,
+        },
+    },
+    [PIX_FMT_YUV422P] = {
+        [PIX_FMT_YUYV422] = {
+            .convert = yuv422p_to_yuyv422,
+        },
+        [PIX_FMT_UYVY422] = {
+            .convert = yuv422p_to_uyvy422,
+        },
+    },
+    [PIX_FMT_YUV444P] = {
+        [PIX_FMT_RGB24] = {
+            .convert = yuv444p_to_rgb24
+        },
+    },
+    [PIX_FMT_YUVJ420P] = {
+        [PIX_FMT_RGB555] = {
+            .convert = yuvj420p_to_rgb555
+        },
+        [PIX_FMT_RGB565] = {
+            .convert = yuvj420p_to_rgb565
+        },
+        [PIX_FMT_BGR24] = {
+            .convert = yuvj420p_to_bgr24
+        },
+        [PIX_FMT_RGB24] = {
+            .convert = yuvj420p_to_rgb24
+        },
+        [PIX_FMT_RGB32] = {
+            .convert = yuvj420p_to_rgb32
+        },
+    },
+    [PIX_FMT_YUVJ444P] = {
+        [PIX_FMT_RGB24] = {
+            .convert = yuvj444p_to_rgb24
+        },
+    },
+    [PIX_FMT_YUYV422] = {
+        [PIX_FMT_YUV420P] = {
+            .convert = yuyv422_to_yuv420p,
+        },
+        [PIX_FMT_YUV422P] = {
+            .convert = yuyv422_to_yuv422p,
+        },
+    },
+    [PIX_FMT_UYVY422] = {
+        [PIX_FMT_YUV420P] = {
+            .convert = uyvy422_to_yuv420p,
+        },
+        [PIX_FMT_YUV422P] = {
+            .convert = uyvy422_to_yuv422p,
+        },
+    },
+    [PIX_FMT_RGB24] = {
+        [PIX_FMT_YUV420P] = {
+            .convert = rgb24_to_yuv420p
+        },
+        [PIX_FMT_RGB565] = {
+            .convert = rgb24_to_rgb565
+        },
+        [PIX_FMT_RGB555] = {
+            .convert = rgb24_to_rgb555
+        },
+        [PIX_FMT_RGB32] = {
+            .convert = rgb24_to_rgb32
+        },
+        [PIX_FMT_BGR24] = {
+            .convert = rgb24_to_bgr24
+        },
+        [PIX_FMT_GRAY8] = {
+            .convert = rgb24_to_gray
+        },
+        [PIX_FMT_PAL8] = {
+            .convert = rgb24_to_pal8
+        },
+        [PIX_FMT_YUV444P] = {
+            .convert = rgb24_to_yuv444p
+        },
+        [PIX_FMT_YUVJ420P] = {
+            .convert = rgb24_to_yuvj420p
+        },
+        [PIX_FMT_YUVJ444P] = {
+            .convert = rgb24_to_yuvj444p
+        },
+    },
+    [PIX_FMT_RGB32] = {
+        [PIX_FMT_RGB24] = {
+            .convert = rgb32_to_rgb24
+        },
+        [PIX_FMT_BGR24] = {
+            .convert = rgb32_to_bgr24
+        },
+        [PIX_FMT_RGB565] = {
+            .convert = rgb32_to_rgb565
+        },
+        [PIX_FMT_RGB555] = {
+            .convert = rgb32_to_rgb555
+        },
+        [PIX_FMT_PAL8] = {
+            .convert = rgb32_to_pal8
+        },
+        [PIX_FMT_YUV420P] = {
+            .convert = rgb32_to_yuv420p
+        },
+        [PIX_FMT_GRAY8] = {
+            .convert = rgb32_to_gray
+        },
+    },
+    [PIX_FMT_BGR24] = {
+        [PIX_FMT_RGB32] = {
+            .convert = bgr24_to_rgb32
+        },
+        [PIX_FMT_RGB24] = {
+            .convert = bgr24_to_rgb24
+        },
+        [PIX_FMT_YUV420P] = {
+            .convert = bgr24_to_yuv420p
+        },
+        [PIX_FMT_GRAY8] = {
+            .convert = bgr24_to_gray
+        },
+    },
+    [PIX_FMT_RGB555] = {
+        [PIX_FMT_RGB24] = {
+            .convert = rgb555_to_rgb24
+        },
+        [PIX_FMT_RGB32] = {
+            .convert = rgb555_to_rgb32
+        },
+        [PIX_FMT_YUV420P] = {
+            .convert = rgb555_to_yuv420p
+        },
+        [PIX_FMT_GRAY8] = {
+            .convert = rgb555_to_gray
+        },
+    },
+    [PIX_FMT_RGB565] = {
+        [PIX_FMT_RGB32] = {
+            .convert = rgb565_to_rgb32
+        },
+        [PIX_FMT_RGB24] = {
+            .convert = rgb565_to_rgb24
+        },
+        [PIX_FMT_YUV420P] = {
+            .convert = rgb565_to_yuv420p
+        },
+        [PIX_FMT_GRAY8] = {
+            .convert = rgb565_to_gray
+        },
+    },
+    [PIX_FMT_GRAY16BE] = {
+        [PIX_FMT_GRAY8] = {
+            .convert = gray16be_to_gray
+        },
+        [PIX_FMT_GRAY16LE] = {
+            .convert = gray16_to_gray16
+        },
+    },
+    [PIX_FMT_GRAY16LE] = {
+        [PIX_FMT_GRAY8] = {
+            .convert = gray16le_to_gray
+        },
+        [PIX_FMT_GRAY16BE] = {
+            .convert = gray16_to_gray16
+        },
+    },
+    [PIX_FMT_GRAY8] = {
+        [PIX_FMT_RGB555] = {
+            .convert = gray_to_rgb555
+        },
+        [PIX_FMT_RGB565] = {
+            .convert = gray_to_rgb565
+        },
+        [PIX_FMT_RGB24] = {
+            .convert = gray_to_rgb24
+        },
+        [PIX_FMT_BGR24] = {
+            .convert = gray_to_bgr24
+        },
+        [PIX_FMT_RGB32] = {
+            .convert = gray_to_rgb32
+        },
+        [PIX_FMT_MONOWHITE] = {
+            .convert = gray_to_monowhite
+        },
+        [PIX_FMT_MONOBLACK] = {
+            .convert = gray_to_monoblack
+        },
+        [PIX_FMT_GRAY16LE] = {
+            .convert = gray_to_gray16
+        },
+        [PIX_FMT_GRAY16BE] = {
+            .convert = gray_to_gray16
+        },
+    },
+    [PIX_FMT_MONOWHITE] = {
+        [PIX_FMT_GRAY8] = {
+            .convert = monowhite_to_gray
+        },
+    },
+    [PIX_FMT_MONOBLACK] = {
+        [PIX_FMT_GRAY8] = {
+            .convert = monoblack_to_gray
+        },
+    },
+    [PIX_FMT_PAL8] = {
+        [PIX_FMT_RGB555] = {
+            .convert = pal8_to_rgb555
+        },
+        [PIX_FMT_RGB565] = {
+            .convert = pal8_to_rgb565
+        },
+        [PIX_FMT_BGR24] = {
+            .convert = pal8_to_bgr24
+        },
+        [PIX_FMT_RGB24] = {
+            .convert = pal8_to_rgb24
+        },
+        [PIX_FMT_RGB32] = {
+            .convert = pal8_to_rgb32
+        },
+    },
+    [PIX_FMT_UYYVYY411] = {
+        [PIX_FMT_YUV411P] = {
+            .convert = uyyvyy411_to_yuv411p,
+        },
+    },
+
+};
+
+int avpicture_alloc(AVPicture *picture,
+                           int pix_fmt, int width, int height)
+{
+    int size;
+    void *ptr;
+
+    size = avpicture_get_size(pix_fmt, width, height);
+    if(size<0)
+        goto fail;
+    ptr = av_malloc(size);
+    if (!ptr)
+        goto fail;
+    avpicture_fill(picture, ptr, pix_fmt, width, height);
+    return 0;
+ fail:
+    memset(picture, 0, sizeof(AVPicture));
+    return -1;
+}
+
+void avpicture_free(AVPicture *picture)
+{
+    av_free(picture->data[0]);
 }
 
-/* XXX: no chroma interpolating is done */
-static void yuv420p_to_rgb24(AVPicture *dst, AVPicture *src, 
-                             int width, int height)
+/* return true if yuv planar */
+static inline int is_yuv_planar(const PixFmtInfo *ps)
 {
-    UINT8 *y1_ptr, *y2_ptr, *cb_ptr, *cr_ptr, *d, *d1, *d2;
-    int w, y, cb, cr, r_add, g_add, b_add, width2;
-    UINT8 *cm = cropTbl + MAX_NEG_CROP;
+    return (ps->color_type == FF_COLOR_YUV ||
+            ps->color_type == FF_COLOR_YUV_JPEG) &&
+        ps->pixel_type == FF_PIXEL_PLANAR;
+}
 
-    d = dst->data[0];
-    y1_ptr = src->data[0];
-    cb_ptr = src->data[1];
-    cr_ptr = src->data[2];
-    width2 = width >> 1;
-    for(;height > 0; height -= 2) {
-        d1 = d;
-        d2 = d + dst->linesize[0];
-        y2_ptr = y1_ptr + src->linesize[0];
-        for(w = width2; w > 0; w --) {
-            cb = cb_ptr[0] - 128;
-            cr = cr_ptr[0] - 128;
-            r_add = C_RV * cr + (1 << (SCALE_BITS - 1));
-            g_add = - C_GU * cb - C_GV * cr + (1 << (SCALE_BITS - 1));
-            b_add = C_BU * cb + (1 << (SCALE_BITS - 1));
-            
-            /* output 4 pixels */
-            RGBOUT(d1[0], d1[1], d1[2], y1_ptr[0]);
-            RGBOUT(d1[3], d1[4], d1[5], y1_ptr[1]);
-            RGBOUT(d2[0], d2[1], d2[2], y2_ptr[0]);
-            RGBOUT(d2[3], d2[4], d2[5], y2_ptr[1]);
-
-            d1 += 6;
-            d2 += 6;
-            y1_ptr += 2;
-            y2_ptr += 2;
-            cb_ptr++;
-            cr_ptr++;
-        }
-        d += 2 * dst->linesize[0];
-        y1_ptr += 2 * src->linesize[0] - width;
-        cb_ptr += src->linesize[1] - width2;
-        cr_ptr += src->linesize[2] - width2;
-    }
-}
-
-/* XXX: no chroma interpolating is done */
-static void yuv422p_to_rgb24(AVPicture *dst, AVPicture *src, 
-                             int width, int height)
-{
-    UINT8 *y1_ptr, *cb_ptr, *cr_ptr, *d, *d1;
-    int w, y, cb, cr, r_add, g_add, b_add, width2;
-    UINT8 *cm = cropTbl + MAX_NEG_CROP;
+int av_picture_crop(AVPicture *dst, const AVPicture *src,
+              int pix_fmt, int top_band, int left_band)
+{
+    int y_shift;
+    int x_shift;
 
-    d = dst->data[0];
-    y1_ptr = src->data[0];
-    cb_ptr = src->data[1];
-    cr_ptr = src->data[2];
-    width2 = width >> 1;
-    for(;height > 0; height --) {
-        d1 = d;
-        for(w = width2; w > 0; w --) {
-            cb = cb_ptr[0] - 128;
-            cr = cr_ptr[0] - 128;
-            r_add = C_RV * cr + (1 << (SCALE_BITS - 1));
-            g_add = - C_GU * cb - C_GV * cr + (1 << (SCALE_BITS - 1));
-            b_add = C_BU * cb + (1 << (SCALE_BITS - 1));
-            
-            /* output 2 pixels */
-            RGBOUT(d1[0], d1[1], d1[2], y1_ptr[0]);
-            RGBOUT(d1[3], d1[4], d1[5], y1_ptr[1]);
-
-            d1 += 6;
-            y1_ptr += 2;
-            cb_ptr++;
-            cr_ptr++;
-        }
-        d += dst->linesize[0];
-        y1_ptr += src->linesize[0] - width;
-        cb_ptr += src->linesize[1] - width2;
-        cr_ptr += src->linesize[2] - width2;
+    if (pix_fmt < 0 || pix_fmt >= PIX_FMT_NB || !is_yuv_planar(&pix_fmt_info[pix_fmt]))
+        return -1;
+
+    y_shift = pix_fmt_info[pix_fmt].y_chroma_shift;
+    x_shift = pix_fmt_info[pix_fmt].x_chroma_shift;
+
+    dst->data[0] = src->data[0] + (top_band * src->linesize[0]) + left_band;
+    dst->data[1] = src->data[1] + ((top_band >> y_shift) * src->linesize[1]) + (left_band >> x_shift);
+    dst->data[2] = src->data[2] + ((top_band >> y_shift) * src->linesize[2]) + (left_band >> x_shift);
+
+    dst->linesize[0] = src->linesize[0];
+    dst->linesize[1] = src->linesize[1];
+    dst->linesize[2] = src->linesize[2];
+    return 0;
+}
+
+int av_picture_pad(AVPicture *dst, const AVPicture *src, int height, int width,
+            int pix_fmt, int padtop, int padbottom, int padleft, int padright,
+            int *color)
+{
+    uint8_t *optr;
+    int y_shift;
+    int x_shift;
+    int yheight;
+    int i, y;
+
+    if (pix_fmt < 0 || pix_fmt >= PIX_FMT_NB ||
+        !is_yuv_planar(&pix_fmt_info[pix_fmt])) return -1;
+
+    for (i = 0; i < 3; i++) {
+        x_shift = i ? pix_fmt_info[pix_fmt].x_chroma_shift : 0;
+        y_shift = i ? pix_fmt_info[pix_fmt].y_chroma_shift : 0;
+
+        if (padtop || padleft) {
+            memset(dst->data[i], color[i],
+                dst->linesize[i] * (padtop >> y_shift) + (padleft >> x_shift));
+        }
+
+        if (padleft || padright) {
+            optr = dst->data[i] + dst->linesize[i] * (padtop >> y_shift) +
+                (dst->linesize[i] - (padright >> x_shift));
+            yheight = (height - 1 - (padtop + padbottom)) >> y_shift;
+            for (y = 0; y < yheight; y++) {
+                memset(optr, color[i], (padleft + padright) >> x_shift);
+                optr += dst->linesize[i];
+            }
+        }
+
+        if (src) { /* first line */
+            uint8_t *iptr = src->data[i];
+            optr = dst->data[i] + dst->linesize[i] * (padtop >> y_shift) +
+                    (padleft >> x_shift);
+            memcpy(optr, iptr, src->linesize[i]);
+            iptr += src->linesize[i];
+            optr = dst->data[i] + dst->linesize[i] * (padtop >> y_shift) +
+                (dst->linesize[i] - (padright >> x_shift));
+            yheight = (height - 1 - (padtop + padbottom)) >> y_shift;
+            for (y = 0; y < yheight; y++) {
+                memset(optr, color[i], (padleft + padright) >> x_shift);
+                memcpy(optr + ((padleft + padright) >> x_shift), iptr,
+                    src->linesize[i]);
+                iptr += src->linesize[i];
+                optr += dst->linesize[i];
+            }
+        }
+
+        if (padbottom || padright) {
+            optr = dst->data[i] + dst->linesize[i] *
+                ((height - padbottom) >> y_shift) - (padright >> x_shift);
+            memset(optr, color[i],dst->linesize[i] *
+                (padbottom >> y_shift) + (padright >> x_shift));
+        }
     }
+    return 0;
+}
+
+#if LIBAVCODEC_VERSION_INT < ((52<<16)+(0<<8)+0)
+void img_copy(AVPicture *dst, const AVPicture *src,
+              int pix_fmt, int width, int height)
+{
+    av_picture_copy(dst, src, pix_fmt, width, height);
+}
+
+int img_crop(AVPicture *dst, const AVPicture *src,
+              int pix_fmt, int top_band, int left_band)
+{
+    return av_picture_crop(dst, src, pix_fmt, top_band, left_band);
+}
+
+int img_pad(AVPicture *dst, const AVPicture *src, int height, int width,
+            int pix_fmt, int padtop, int padbottom, int padleft, int padright,
+            int *color)
+{
+    return av_picture_pad(dst, src, height, width, pix_fmt, padtop, padbottom, padleft, padright, color);
 }
+#endif
 
+#ifndef CONFIG_SWSCALER
 /* XXX: always use linesize. Return -1 if not supported */
 int img_convert(AVPicture *dst, int dst_pix_fmt,
-                AVPicture *src, int pix_fmt, 
-                int width, int height)
+                const AVPicture *src, int src_pix_fmt,
+                int src_width, int src_height)
 {
-    int i;
+    static int inited;
+    int i, ret, dst_width, dst_height, int_pix_fmt;
+    const PixFmtInfo *src_pix, *dst_pix;
+    const ConvertEntry *ce;
+    AVPicture tmp1, *tmp = &tmp1;
 
-    if (dst_pix_fmt == pix_fmt) {
-        switch(pix_fmt) {
-        case PIX_FMT_YUV420P:
-            for(i=0;i<3;i++) {
-                if (i == 1) {
-                    width >>= 1;
-                    height >>= 1;
-                }
-                img_copy(dst->data[i], dst->linesize[i],
-                         src->data[i], src->linesize[i],
-                         width, height);
-            }
-            break;
-        default:
-            return -1;
+    if (src_pix_fmt < 0 || src_pix_fmt >= PIX_FMT_NB ||
+        dst_pix_fmt < 0 || dst_pix_fmt >= PIX_FMT_NB)
+        return -1;
+    if (src_width <= 0 || src_height <= 0)
+        return 0;
+
+    if (!inited) {
+        inited = 1;
+        img_convert_init();
+    }
+
+    dst_width = src_width;
+    dst_height = src_height;
+
+    dst_pix = &pix_fmt_info[dst_pix_fmt];
+    src_pix = &pix_fmt_info[src_pix_fmt];
+    if (src_pix_fmt == dst_pix_fmt) {
+        /* no conversion needed: just copy */
+        av_picture_copy(dst, src, dst_pix_fmt, dst_width, dst_height);
+        return 0;
+    }
+
+    ce = &convert_table[src_pix_fmt][dst_pix_fmt];
+    if (ce->convert) {
+        /* specific conversion routine */
+        ce->convert(dst, src, dst_width, dst_height);
+        return 0;
+    }
+
+    /* gray to YUV */
+    if (is_yuv_planar(dst_pix) &&
+        src_pix_fmt == PIX_FMT_GRAY8) {
+        int w, h, y;
+        uint8_t *d;
+
+        if (dst_pix->color_type == FF_COLOR_YUV_JPEG) {
+            ff_img_copy_plane(dst->data[0], dst->linesize[0],
+                     src->data[0], src->linesize[0],
+                     dst_width, dst_height);
+        } else {
+            img_apply_table(dst->data[0], dst->linesize[0],
+                            src->data[0], src->linesize[0],
+                            dst_width, dst_height,
+                            y_jpeg_to_ccir);
         }
-    } else if (dst_pix_fmt == PIX_FMT_YUV420P) {
-        
-        switch(pix_fmt) {
-        case PIX_FMT_YUV420P:
-            for(i=0;i<3;i++) {
-                img_copy(dst->data[i], dst->linesize[i],
-                         src->data[i], src->linesize[i],
-                         width, height);
+        /* fill U and V with 128 */
+        w = dst_width;
+        h = dst_height;
+        w >>= dst_pix->x_chroma_shift;
+        h >>= dst_pix->y_chroma_shift;
+        for(i = 1; i <= 2; i++) {
+            d = dst->data[i];
+            for(y = 0; y< h; y++) {
+                memset(d, 128, w);
+                d += dst->linesize[i];
             }
-            break;
-        case PIX_FMT_YUV422P:
-            img_copy(dst->data[0], dst->linesize[0],
+        }
+        return 0;
+    }
+
+    /* YUV to gray */
+    if (is_yuv_planar(src_pix) &&
+        dst_pix_fmt == PIX_FMT_GRAY8) {
+        if (src_pix->color_type == FF_COLOR_YUV_JPEG) {
+            ff_img_copy_plane(dst->data[0], dst->linesize[0],
                      src->data[0], src->linesize[0],
-                     width, height);
-            width >>= 1;
-            height >>= 1;
-            for(i=1;i<3;i++) {
-                shrink2(dst->data[i], dst->linesize[i],
-                        src->data[i], src->linesize[i],
-                        width, height);
-            }
+                     dst_width, dst_height);
+        } else {
+            img_apply_table(dst->data[0], dst->linesize[0],
+                            src->data[0], src->linesize[0],
+                            dst_width, dst_height,
+                            y_ccir_to_jpeg);
+        }
+        return 0;
+    }
+
+    /* YUV to YUV planar */
+    if (is_yuv_planar(dst_pix) && is_yuv_planar(src_pix)) {
+        int x_shift, y_shift, w, h, xy_shift;
+        void (*resize_func)(uint8_t *dst, int dst_wrap,
+                            const uint8_t *src, int src_wrap,
+                            int width, int height);
+
+        /* compute chroma size of the smallest dimensions */
+        w = dst_width;
+        h = dst_height;
+        if (dst_pix->x_chroma_shift >= src_pix->x_chroma_shift)
+            w >>= dst_pix->x_chroma_shift;
+        else
+            w >>= src_pix->x_chroma_shift;
+        if (dst_pix->y_chroma_shift >= src_pix->y_chroma_shift)
+            h >>= dst_pix->y_chroma_shift;
+        else
+            h >>= src_pix->y_chroma_shift;
+
+        x_shift = (dst_pix->x_chroma_shift - src_pix->x_chroma_shift);
+        y_shift = (dst_pix->y_chroma_shift - src_pix->y_chroma_shift);
+        xy_shift = ((x_shift & 0xf) << 4) | (y_shift & 0xf);
+        /* there must be filters for conversion at least from and to
+           YUV444 format */
+        switch(xy_shift) {
+        case 0x00:
+            resize_func = ff_img_copy_plane;
             break;
-        case PIX_FMT_YUV444P:
-            img_copy(dst->data[0], dst->linesize[0],
-                     src->data[0], src->linesize[0],
-                     width, height);
-            width >>= 1;
-            height >>= 1;
-            for(i=1;i<3;i++) {
-                shrink22(dst->data[i], dst->linesize[i],
-                         src->data[i], src->linesize[i],
-                         width, height);
-            }
+        case 0x10:
+            resize_func = shrink21;
             break;
-        case PIX_FMT_YUV422:
-            yuv422_to_yuv420p(dst->data[0], dst->data[1], dst->data[2], 
-                              src->data[0], width, height);
+        case 0x20:
+            resize_func = shrink41;
             break;
-        case PIX_FMT_RGB24:
-            rgb24_to_yuv420p(dst->data[0], dst->data[1], dst->data[2], 
-                             src->data[0], width, height);
+        case 0x01:
+            resize_func = shrink12;
             break;
-        case PIX_FMT_BGR24:
-            bgr24_to_yuv420p(dst->data[0], dst->data[1], dst->data[2], 
-                             src->data[0], width, height);
+        case 0x11:
+            resize_func = ff_shrink22;
             break;
-        default:
-            return -1;
-        }
-    } else if (dst_pix_fmt == PIX_FMT_RGB24) {
-        switch(pix_fmt) {
-        case PIX_FMT_YUV420P:
-            yuv420p_to_rgb24(dst, src, width, height);
+        case 0x22:
+            resize_func = ff_shrink44;
+            break;
+        case 0xf0:
+            resize_func = grow21;
+            break;
+        case 0xe0:
+            resize_func = grow41;
+            break;
+        case 0xff:
+            resize_func = grow22;
             break;
-        case PIX_FMT_YUV422P:
-            yuv422p_to_rgb24(dst, src, width, height);
+        case 0xee:
+            resize_func = grow44;
+            break;
+        case 0xf1:
+            resize_func = conv411;
             break;
         default:
-            return -1;
+            /* currently not handled */
+            goto no_chroma_filter;
+        }
+
+        ff_img_copy_plane(dst->data[0], dst->linesize[0],
+                       src->data[0], src->linesize[0],
+                       dst_width, dst_height);
+
+        for(i = 1;i <= 2; i++)
+            resize_func(dst->data[i], dst->linesize[i],
+                        src->data[i], src->linesize[i],
+                        dst_width>>dst_pix->x_chroma_shift, dst_height>>dst_pix->y_chroma_shift);
+        /* if yuv color space conversion is needed, we do it here on
+           the destination image */
+        if (dst_pix->color_type != src_pix->color_type) {
+            const uint8_t *y_table, *c_table;
+            if (dst_pix->color_type == FF_COLOR_YUV) {
+                y_table = y_jpeg_to_ccir;
+                c_table = c_jpeg_to_ccir;
+            } else {
+                y_table = y_ccir_to_jpeg;
+                c_table = c_ccir_to_jpeg;
+            }
+            img_apply_table(dst->data[0], dst->linesize[0],
+                            dst->data[0], dst->linesize[0],
+                            dst_width, dst_height,
+                            y_table);
+
+            for(i = 1;i <= 2; i++)
+                img_apply_table(dst->data[i], dst->linesize[i],
+                                dst->data[i], dst->linesize[i],
+                                dst_width>>dst_pix->x_chroma_shift,
+                                dst_height>>dst_pix->y_chroma_shift,
+                                c_table);
         }
+        return 0;
+    }
+ no_chroma_filter:
+
+    /* try to use an intermediate format */
+    if (src_pix_fmt == PIX_FMT_YUYV422 ||
+        dst_pix_fmt == PIX_FMT_YUYV422) {
+        /* specific case: convert to YUV422P first */
+        int_pix_fmt = PIX_FMT_YUV422P;
+    } else if (src_pix_fmt == PIX_FMT_UYVY422 ||
+        dst_pix_fmt == PIX_FMT_UYVY422) {
+        /* specific case: convert to YUV422P first */
+        int_pix_fmt = PIX_FMT_YUV422P;
+    } else if (src_pix_fmt == PIX_FMT_UYYVYY411 ||
+        dst_pix_fmt == PIX_FMT_UYYVYY411) {
+        /* specific case: convert to YUV411P first */
+        int_pix_fmt = PIX_FMT_YUV411P;
+    } else if ((src_pix->color_type == FF_COLOR_GRAY &&
+                src_pix_fmt != PIX_FMT_GRAY8) ||
+               (dst_pix->color_type == FF_COLOR_GRAY &&
+                dst_pix_fmt != PIX_FMT_GRAY8)) {
+        /* gray8 is the normalized format */
+        int_pix_fmt = PIX_FMT_GRAY8;
+    } else if ((is_yuv_planar(src_pix) &&
+                src_pix_fmt != PIX_FMT_YUV444P &&
+                src_pix_fmt != PIX_FMT_YUVJ444P)) {
+        /* yuv444 is the normalized format */
+        if (src_pix->color_type == FF_COLOR_YUV_JPEG)
+            int_pix_fmt = PIX_FMT_YUVJ444P;
+        else
+            int_pix_fmt = PIX_FMT_YUV444P;
+    } else if ((is_yuv_planar(dst_pix) &&
+                dst_pix_fmt != PIX_FMT_YUV444P &&
+                dst_pix_fmt != PIX_FMT_YUVJ444P)) {
+        /* yuv444 is the normalized format */
+        if (dst_pix->color_type == FF_COLOR_YUV_JPEG)
+            int_pix_fmt = PIX_FMT_YUVJ444P;
+        else
+            int_pix_fmt = PIX_FMT_YUV444P;
     } else {
+        /* the two formats are rgb or gray8 or yuv[j]444p */
+        if (src_pix->is_alpha && dst_pix->is_alpha)
+            int_pix_fmt = PIX_FMT_RGB32;
+        else
+            int_pix_fmt = PIX_FMT_RGB24;
+    }
+    if (avpicture_alloc(tmp, int_pix_fmt, dst_width, dst_height) < 0)
         return -1;
+    ret = -1;
+    if (img_convert(tmp, int_pix_fmt,
+                    src, src_pix_fmt, src_width, src_height) < 0)
+        goto fail1;
+    if (img_convert(dst, dst_pix_fmt,
+                    tmp, int_pix_fmt, dst_width, dst_height) < 0)
+        goto fail1;
+    ret = 0;
+ fail1:
+    avpicture_free(tmp);
+    return ret;
+}
+#endif
+
+/* NOTE: we scan all the pixels to have an exact information */
+static int get_alpha_info_pal8(const AVPicture *src, int width, int height)
+{
+    const unsigned char *p;
+    int src_wrap, ret, x, y;
+    unsigned int a;
+    uint32_t *palette = (uint32_t *)src->data[1];
+
+    p = src->data[0];
+    src_wrap = src->linesize[0] - width;
+    ret = 0;
+    for(y=0;y<height;y++) {
+        for(x=0;x<width;x++) {
+            a = palette[p[0]] >> 24;
+            if (a == 0x00) {
+                ret |= FF_ALPHA_TRANSP;
+            } else if (a != 0xff) {
+                ret |= FF_ALPHA_SEMI_TRANSP;
+            }
+            p++;
+        }
+        p += src_wrap;
     }
-    return 0;
+    return ret;
+}
+
+int img_get_alpha_info(const AVPicture *src,
+                       int pix_fmt, int width, int height)
+{
+    const PixFmtInfo *pf = &pix_fmt_info[pix_fmt];
+    int ret;
+
+    pf = &pix_fmt_info[pix_fmt];
+    /* no alpha can be represented in format */
+    if (!pf->is_alpha)
+        return 0;
+    switch(pix_fmt) {
+    case PIX_FMT_RGB32:
+        ret = get_alpha_info_rgb32(src, width, height);
+        break;
+    case PIX_FMT_PAL8:
+        ret = get_alpha_info_pal8(src, width, height);
+        break;
+    default:
+        /* we do not know, so everything is indicated */
+        ret = FF_ALPHA_TRANSP | FF_ALPHA_SEMI_TRANSP;
+        break;
+    }
+    return ret;
 }
 
+#ifdef HAVE_MMX
+#define DEINT_INPLACE_LINE_LUM \
+                    movd_m2r(lum_m4[0],mm0);\
+                    movd_m2r(lum_m3[0],mm1);\
+                    movd_m2r(lum_m2[0],mm2);\
+                    movd_m2r(lum_m1[0],mm3);\
+                    movd_m2r(lum[0],mm4);\
+                    punpcklbw_r2r(mm7,mm0);\
+                    movd_r2m(mm2,lum_m4[0]);\
+                    punpcklbw_r2r(mm7,mm1);\
+                    punpcklbw_r2r(mm7,mm2);\
+                    punpcklbw_r2r(mm7,mm3);\
+                    punpcklbw_r2r(mm7,mm4);\
+                    paddw_r2r(mm3,mm1);\
+                    psllw_i2r(1,mm2);\
+                    paddw_r2r(mm4,mm0);\
+                    psllw_i2r(2,mm1);\
+                    paddw_r2r(mm6,mm2);\
+                    paddw_r2r(mm2,mm1);\
+                    psubusw_r2r(mm0,mm1);\
+                    psrlw_i2r(3,mm1);\
+                    packuswb_r2r(mm7,mm1);\
+                    movd_r2m(mm1,lum_m2[0]);
+
+#define DEINT_LINE_LUM \
+                    movd_m2r(lum_m4[0],mm0);\
+                    movd_m2r(lum_m3[0],mm1);\
+                    movd_m2r(lum_m2[0],mm2);\
+                    movd_m2r(lum_m1[0],mm3);\
+                    movd_m2r(lum[0],mm4);\
+                    punpcklbw_r2r(mm7,mm0);\
+                    punpcklbw_r2r(mm7,mm1);\
+                    punpcklbw_r2r(mm7,mm2);\
+                    punpcklbw_r2r(mm7,mm3);\
+                    punpcklbw_r2r(mm7,mm4);\
+                    paddw_r2r(mm3,mm1);\
+                    psllw_i2r(1,mm2);\
+                    paddw_r2r(mm4,mm0);\
+                    psllw_i2r(2,mm1);\
+                    paddw_r2r(mm6,mm2);\
+                    paddw_r2r(mm2,mm1);\
+                    psubusw_r2r(mm0,mm1);\
+                    psrlw_i2r(3,mm1);\
+                    packuswb_r2r(mm7,mm1);\
+                    movd_r2m(mm1,dst[0]);
+#endif
+
 /* filter parameters: [-1 4 2 4 -1] // 8 */
-static void deinterlace_line(UINT8 *dst, UINT8 *src, int src_wrap,
+static void deinterlace_line(uint8_t *dst,
+                             const uint8_t *lum_m4, const uint8_t *lum_m3,
+                             const uint8_t *lum_m2, const uint8_t *lum_m1,
+                             const uint8_t *lum,
                              int size)
 {
-    UINT8 *cm = cropTbl + MAX_NEG_CROP;
+#ifndef HAVE_MMX
+    uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;
     int sum;
-    UINT8 *s;
 
     for(;size > 0;size--) {
-        s = src;
-        sum = -s[0];
-        s += src_wrap;
-        sum += s[0] << 2;
-        s += src_wrap;
-        sum += s[0] << 1;
-        s += src_wrap;
-        sum += s[0] << 2;
-        s += src_wrap;
-        sum += -s[0];
+        sum = -lum_m4[0];
+        sum += lum_m3[0] << 2;
+        sum += lum_m2[0] << 1;
+        sum += lum_m1[0] << 2;
+        sum += -lum[0];
         dst[0] = cm[(sum + 4) >> 3];
+        lum_m4++;
+        lum_m3++;
+        lum_m2++;
+        lum_m1++;
+        lum++;
         dst++;
-        src++;
     }
+#else
+
+    {
+        mmx_t rounder;
+        rounder.uw[0]=4;
+        rounder.uw[1]=4;
+        rounder.uw[2]=4;
+        rounder.uw[3]=4;
+        pxor_r2r(mm7,mm7);
+        movq_m2r(rounder,mm6);
+    }
+    for (;size > 3; size-=4) {
+        DEINT_LINE_LUM
+        lum_m4+=4;
+        lum_m3+=4;
+        lum_m2+=4;
+        lum_m1+=4;
+        lum+=4;
+        dst+=4;
+    }
+#endif
+}
+static void deinterlace_line_inplace(uint8_t *lum_m4, uint8_t *lum_m3, uint8_t *lum_m2, uint8_t *lum_m1, uint8_t *lum,
+                             int size)
+{
+#ifndef HAVE_MMX
+    uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;
+    int sum;
+
+    for(;size > 0;size--) {
+        sum = -lum_m4[0];
+        sum += lum_m3[0] << 2;
+        sum += lum_m2[0] << 1;
+        lum_m4[0]=lum_m2[0];
+        sum += lum_m1[0] << 2;
+        sum += -lum[0];
+        lum_m2[0] = cm[(sum + 4) >> 3];
+        lum_m4++;
+        lum_m3++;
+        lum_m2++;
+        lum_m1++;
+        lum++;
+    }
+#else
+
+    {
+        mmx_t rounder;
+        rounder.uw[0]=4;
+        rounder.uw[1]=4;
+        rounder.uw[2]=4;
+        rounder.uw[3]=4;
+        pxor_r2r(mm7,mm7);
+        movq_m2r(rounder,mm6);
+    }
+    for (;size > 3; size-=4) {
+        DEINT_INPLACE_LINE_LUM
+        lum_m4+=4;
+        lum_m3+=4;
+        lum_m2+=4;
+        lum_m1+=4;
+        lum+=4;
+    }
+#endif
 }
 
 /* deinterlacing : 2 temporal taps, 3 spatial taps linear filter. The
    top field is copied as is, but the bottom field is deinterlaced
    against the top field. */
-static void deinterlace_bottom_field(UINT8 *dst, int dst_wrap,
-                                     UINT8 *src1, int src_wrap,
-                                     int width, int height)
+static void deinterlace_bottom_field(uint8_t *dst, int dst_wrap,
+                                    const uint8_t *src1, int src_wrap,
+                                    int width, int height)
 {
-    UINT8 *src, *ptr;
-    int y, y1, i;
-    UINT8 buf[5 * width];
+    const uint8_t *src_m2, *src_m1, *src_0, *src_p1, *src_p2;
+    int y;
 
-    src = src1;
-    for(y=0;y<height;y+=2) {
-        /* copy top field line */
-        memcpy(dst, src, width);
+    src_m2 = src1;
+    src_m1 = src1;
+    src_0=&src_m1[src_wrap];
+    src_p1=&src_0[src_wrap];
+    src_p2=&src_p1[src_wrap];
+    for(y=0;y<(height-2);y+=2) {
+        memcpy(dst,src_m1,width);
         dst += dst_wrap;
-        src += (1 - 2) * src_wrap;
-        y1 = y - 2;
-        if (y1 >= 0 && (y1 + 4) < height) {
-            /* fast case : no edges */
-            deinterlace_line(dst, src, src_wrap, width);
-        } else {
-            /* in order to use the same function, we use an intermediate buffer */
-            ptr = buf;
-            for(i=0;i<5;i++) {
-                if (y1 < 0)
-                    memcpy(ptr, src1, width);
-                else if (y1 >= height) 
-                    memcpy(ptr, src1 + (height - 1) * src_wrap, width);
-                else
-                    memcpy(ptr, src1 + y1 * src_wrap, width);
-                y1++;
-                ptr += width;
-            }
-            deinterlace_line(dst, buf, width, width);
-        }
+        deinterlace_line(dst,src_m2,src_m1,src_0,src_p1,src_p2,width);
+        src_m2 = src_0;
+        src_m1 = src_p1;
+        src_0 = src_p2;
+        src_p1 += 2*src_wrap;
+        src_p2 += 2*src_wrap;
         dst += dst_wrap;
-        src += (2 + 1) * src_wrap;
     }
+    memcpy(dst,src_m1,width);
+    dst += dst_wrap;
+    /* do last line */
+    deinterlace_line(dst,src_m2,src_m1,src_0,src_0,src_0,width);
 }
 
+static void deinterlace_bottom_field_inplace(uint8_t *src1, int src_wrap,
+                                             int width, int height)
+{
+    uint8_t *src_m1, *src_0, *src_p1, *src_p2;
+    int y;
+    uint8_t *buf;
+    buf = (uint8_t*)av_malloc(width);
+
+    src_m1 = src1;
+    memcpy(buf,src_m1,width);
+    src_0=&src_m1[src_wrap];
+    src_p1=&src_0[src_wrap];
+    src_p2=&src_p1[src_wrap];
+    for(y=0;y<(height-2);y+=2) {
+        deinterlace_line_inplace(buf,src_m1,src_0,src_p1,src_p2,width);
+        src_m1 = src_p1;
+        src_0 = src_p2;
+        src_p1 += 2*src_wrap;
+        src_p2 += 2*src_wrap;
+    }
+    /* do last line */
+    deinterlace_line_inplace(buf,src_m1,src_0,src_0,src_0,width);
+    av_free(buf);
+}
 
-/* deinterlace, return -1 if format not handled */
-int avpicture_deinterlace(AVPicture *dst, AVPicture *src,
+int avpicture_deinterlace(AVPicture *dst, const AVPicture *src,
                           int pix_fmt, int width, int height)
 {
     int i;
 
     if (pix_fmt != PIX_FMT_YUV420P &&
         pix_fmt != PIX_FMT_YUV422P &&
-        pix_fmt != PIX_FMT_YUV444P)
+        pix_fmt != PIX_FMT_YUV444P &&
+        pix_fmt != PIX_FMT_YUV411P)
         return -1;
-    if ((width & 1) != 0 || (height & 3) != 0)
+    if ((width & 3) != 0 || (height & 3) != 0)
         return -1;
-    
+
     for(i=0;i<3;i++) {
         if (i == 1) {
             switch(pix_fmt) {
@@ -537,13 +2829,26 @@ int avpicture_deinterlace(AVPicture *dst, AVPicture *src,
             case PIX_FMT_YUV422P:
                 width >>= 1;
                 break;
+            case PIX_FMT_YUV411P:
+                width >>= 2;
+                break;
             default:
                 break;
             }
         }
-        deinterlace_bottom_field(dst->data[i], dst->linesize[i],
-                                 src->data[i], src->linesize[i],
+        if (src == dst) {
+            deinterlace_bottom_field_inplace(dst->data[i], dst->linesize[i],
                                  width, height);
+        } else {
+            deinterlace_bottom_field(dst->data[i],dst->linesize[i],
+                                        src->data[i], src->linesize[i],
+                                        width, height);
+        }
     }
+#ifdef HAVE_MMX
+    emms();
+#endif
     return 0;
 }
+
+#undef FIX