2 * Copyright (c) 2013 Paul B Mahol
4 * This file is part of FFmpeg.
6 * FFmpeg is free software; you can redistribute it and/or
7 * modify it under the terms of the GNU Lesser General Public
8 * License as published by the Free Software Foundation; either
9 * version 2.1 of the License, or (at your option) any later version.
11 * FFmpeg is distributed in the hope that it will be useful,
12 * but WITHOUT ANY WARRANTY; without even the implied warranty of
13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
14 * Lesser General Public License for more details.
16 * You should have received a copy of the GNU Lesser General Public
17 * License along with FFmpeg; if not, write to the Free Software
18 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
23 * audio to video multimedia vectorscope filter
26 #include "libavutil/avassert.h"
27 #include "libavutil/channel_layout.h"
28 #include "libavutil/opt.h"
29 #include "libavutil/parseutils.h"
36 enum VectorScopeMode {
43 enum VectorScopeDraw {
49 enum VectorScopeScale {
57 typedef struct AudioVectorScopeContext {
68 unsigned prev_x, prev_y;
69 AVRational frame_rate;
70 } AudioVectorScopeContext;
72 #define OFFSET(x) offsetof(AudioVectorScopeContext, x)
73 #define FLAGS AV_OPT_FLAG_FILTERING_PARAM|AV_OPT_FLAG_VIDEO_PARAM
75 static const AVOption avectorscope_options[] = {
76 { "mode", "set mode", OFFSET(mode), AV_OPT_TYPE_INT, {.i64=LISSAJOUS}, 0, MODE_NB-1, FLAGS, "mode" },
77 { "m", "set mode", OFFSET(mode), AV_OPT_TYPE_INT, {.i64=LISSAJOUS}, 0, MODE_NB-1, FLAGS, "mode" },
78 { "lissajous", "", 0, AV_OPT_TYPE_CONST, {.i64=LISSAJOUS}, 0, 0, FLAGS, "mode" },
79 { "lissajous_xy", "", 0, AV_OPT_TYPE_CONST, {.i64=LISSAJOUS_XY}, 0, 0, FLAGS, "mode" },
80 { "polar", "", 0, AV_OPT_TYPE_CONST, {.i64=POLAR}, 0, 0, FLAGS, "mode" },
81 { "rate", "set video rate", OFFSET(frame_rate), AV_OPT_TYPE_VIDEO_RATE, {.str="25"}, 0, INT_MAX, FLAGS },
82 { "r", "set video rate", OFFSET(frame_rate), AV_OPT_TYPE_VIDEO_RATE, {.str="25"}, 0, INT_MAX, FLAGS },
83 { "size", "set video size", OFFSET(w), AV_OPT_TYPE_IMAGE_SIZE, {.str="400x400"}, 0, 0, FLAGS },
84 { "s", "set video size", OFFSET(w), AV_OPT_TYPE_IMAGE_SIZE, {.str="400x400"}, 0, 0, FLAGS },
85 { "rc", "set red contrast", OFFSET(contrast[0]), AV_OPT_TYPE_INT, {.i64=40}, 0, 255, FLAGS },
86 { "gc", "set green contrast", OFFSET(contrast[1]), AV_OPT_TYPE_INT, {.i64=160}, 0, 255, FLAGS },
87 { "bc", "set blue contrast", OFFSET(contrast[2]), AV_OPT_TYPE_INT, {.i64=80}, 0, 255, FLAGS },
88 { "ac", "set alpha contrast", OFFSET(contrast[3]), AV_OPT_TYPE_INT, {.i64=255}, 0, 255, FLAGS },
89 { "rf", "set red fade", OFFSET(fade[0]), AV_OPT_TYPE_INT, {.i64=15}, 0, 255, FLAGS },
90 { "gf", "set green fade", OFFSET(fade[1]), AV_OPT_TYPE_INT, {.i64=10}, 0, 255, FLAGS },
91 { "bf", "set blue fade", OFFSET(fade[2]), AV_OPT_TYPE_INT, {.i64=5}, 0, 255, FLAGS },
92 { "af", "set alpha fade", OFFSET(fade[3]), AV_OPT_TYPE_INT, {.i64=5}, 0, 255, FLAGS },
93 { "zoom", "set zoom factor", OFFSET(zoom), AV_OPT_TYPE_DOUBLE, {.dbl=1}, 0, 10, FLAGS },
94 { "draw", "set draw mode", OFFSET(draw), AV_OPT_TYPE_INT, {.i64=DOT}, 0, DRAW_NB-1, FLAGS, "draw" },
95 { "dot", "", 0, AV_OPT_TYPE_CONST, {.i64=DOT} , 0, 0, FLAGS, "draw" },
96 { "line", "", 0, AV_OPT_TYPE_CONST, {.i64=LINE}, 0, 0, FLAGS, "draw" },
97 { "scale", "set amplitude scale mode", OFFSET(scale), AV_OPT_TYPE_INT, {.i64=LIN}, 0, SCALE_NB-1, FLAGS, "scale" },
98 { "lin", "linear", 0, AV_OPT_TYPE_CONST, {.i64=LIN}, 0, 0, FLAGS, "scale" },
99 { "sqrt", "square root", 0, AV_OPT_TYPE_CONST, {.i64=SQRT}, 0, 0, FLAGS, "scale" },
100 { "cbrt", "cube root", 0, AV_OPT_TYPE_CONST, {.i64=CBRT}, 0, 0, FLAGS, "scale" },
101 { "log", "logarithmic", 0, AV_OPT_TYPE_CONST, {.i64=LOG}, 0, 0, FLAGS, "scale" },
105 AVFILTER_DEFINE_CLASS(avectorscope);
107 static void draw_dot(AudioVectorScopeContext *s, unsigned x, unsigned y)
109 const int linesize = s->outpicref->linesize[0];
113 if (y >= s->h || x >= s->w)
116 y = FFMIN(y, s->h - 1);
117 x = FFMIN(x, s->w - 1);
120 dst = &s->outpicref->data[0][y * linesize + x * 4];
121 dst[0] = FFMIN(dst[0] + s->contrast[0], 255);
122 dst[1] = FFMIN(dst[1] + s->contrast[1], 255);
123 dst[2] = FFMIN(dst[2] + s->contrast[2], 255);
124 dst[3] = FFMIN(dst[3] + s->contrast[3], 255);
127 static void draw_line(AudioVectorScopeContext *s, int x0, int y0, int x1, int y1)
129 int dx = FFABS(x1-x0), sx = x0 < x1 ? 1 : -1;
130 int dy = FFABS(y1-y0), sy = y0 < y1 ? 1 : -1;
131 int err = (dx>dy ? dx : -dy) / 2, e2;
136 if (x0 == x1 && y0 == y1)
153 static void fade(AudioVectorScopeContext *s)
155 const int linesize = s->outpicref->linesize[0];
158 if (s->fade[0] || s->fade[1] || s->fade[2]) {
159 uint8_t *d = s->outpicref->data[0];
160 for (i = 0; i < s->h; i++) {
161 for (j = 0; j < s->w*4; j+=4) {
162 d[j+0] = FFMAX(d[j+0] - s->fade[0], 0);
163 d[j+1] = FFMAX(d[j+1] - s->fade[1], 0);
164 d[j+2] = FFMAX(d[j+2] - s->fade[2], 0);
165 d[j+3] = FFMAX(d[j+3] - s->fade[3], 0);
172 static int query_formats(AVFilterContext *ctx)
174 AVFilterFormats *formats = NULL;
175 AVFilterChannelLayouts *layout = NULL;
176 AVFilterLink *inlink = ctx->inputs[0];
177 AVFilterLink *outlink = ctx->outputs[0];
178 static const enum AVSampleFormat sample_fmts[] = { AV_SAMPLE_FMT_S16, AV_SAMPLE_FMT_FLT, AV_SAMPLE_FMT_NONE };
179 static const enum AVPixelFormat pix_fmts[] = { AV_PIX_FMT_RGBA, AV_PIX_FMT_NONE };
182 formats = ff_make_format_list(sample_fmts);
183 if ((ret = ff_formats_ref (formats, &inlink->out_formats )) < 0 ||
184 (ret = ff_add_channel_layout (&layout, AV_CH_LAYOUT_STEREO )) < 0 ||
185 (ret = ff_channel_layouts_ref (layout , &inlink->out_channel_layouts)) < 0)
188 formats = ff_all_samplerates();
189 if ((ret = ff_formats_ref(formats, &inlink->out_samplerates)) < 0)
192 formats = ff_make_format_list(pix_fmts);
193 if ((ret = ff_formats_ref(formats, &outlink->in_formats)) < 0)
199 static int config_input(AVFilterLink *inlink)
201 AVFilterContext *ctx = inlink->dst;
202 AudioVectorScopeContext *s = ctx->priv;
205 nb_samples = FFMAX(1024, ((double)inlink->sample_rate / av_q2d(s->frame_rate)) + 0.5);
206 inlink->partial_buf_size =
207 inlink->min_samples =
208 inlink->max_samples = nb_samples;
213 static int config_output(AVFilterLink *outlink)
215 AudioVectorScopeContext *s = outlink->src->priv;
219 outlink->sample_aspect_ratio = (AVRational){1,1};
220 outlink->frame_rate = s->frame_rate;
222 s->prev_x = s->hw = s->w / 2;
223 s->prev_y = s->hh = s->mode == POLAR ? s->h - 1 : s->h / 2;
228 static int filter_frame(AVFilterLink *inlink, AVFrame *insamples)
230 AVFilterContext *ctx = inlink->dst;
231 AVFilterLink *outlink = ctx->outputs[0];
232 AudioVectorScopeContext *s = ctx->priv;
233 const int hw = s->hw;
234 const int hh = s->hh;
236 unsigned prev_x = s->prev_x, prev_y = s->prev_y;
237 double zoom = s->zoom;
240 if (!s->outpicref || s->outpicref->width != outlink->w ||
241 s->outpicref->height != outlink->h) {
242 av_frame_free(&s->outpicref);
243 s->outpicref = ff_get_video_buffer(outlink, outlink->w, outlink->h);
245 av_frame_free(&insamples);
246 return AVERROR(ENOMEM);
249 s->outpicref->sample_aspect_ratio = (AVRational){1,1};
250 for (i = 0; i < outlink->h; i++)
251 memset(s->outpicref->data[0] + i * s->outpicref->linesize[0], 0, outlink->w * 4);
253 s->outpicref->pts = insamples->pts;
260 switch (insamples->format) {
261 case AV_SAMPLE_FMT_S16: {
262 int16_t *samples = (int16_t *)insamples->data[0];
264 for (i = 0; i < insamples->nb_samples * 2; i++) {
265 float sample = samples[i] / (float)INT16_MAX;
266 max = FFMAX(FFABS(sample), max);
271 case AV_SAMPLE_FMT_FLT: {
272 float *samples = (float *)insamples->data[0];
274 for (i = 0; i < insamples->nb_samples * 2; i++) {
275 max = FFMAX(FFABS(samples[i]), max);
286 for (i = 0; i < insamples->nb_samples; i++) {
287 int16_t *samples = (int16_t *)insamples->data[0] + i * 2;
288 float *samplesf = (float *)insamples->data[0] + i * 2;
291 switch (insamples->format) {
292 case AV_SAMPLE_FMT_S16:
293 src[0] = samples[0] / (float)INT16_MAX;
294 src[1] = samples[1] / (float)INT16_MAX;
296 case AV_SAMPLE_FMT_FLT:
297 src[0] = samplesf[0];
298 src[1] = samplesf[1];
306 src[0] = FFSIGN(src[0]) * sqrtf(FFABS(src[0]));
307 src[1] = FFSIGN(src[1]) * sqrtf(FFABS(src[1]));
310 src[0] = FFSIGN(src[0]) * cbrtf(FFABS(src[0]));
311 src[1] = FFSIGN(src[1]) * cbrtf(FFABS(src[1]));
314 src[0] = FFSIGN(src[0]) * logf(1 + FFABS(src[0])) / logf(2);
315 src[1] = FFSIGN(src[1]) * logf(1 + FFABS(src[1])) / logf(2);
319 if (s->mode == LISSAJOUS) {
320 x = ((src[1] - src[0]) * zoom / 2 + 1) * hw;
321 y = (1.0 - (src[0] + src[1]) * zoom / 2) * hh;
322 } else if (s->mode == LISSAJOUS_XY) {
323 x = (src[1] * zoom + 1) * hw;
324 y = (src[0] * zoom + 1) * hh;
326 float sx, sy, cx, cy;
330 cx = sx * sqrtf(1 - 0.5 * sy * sy);
331 cy = sy * sqrtf(1 - 0.5 * sx * sx);
332 x = hw + hw * FFSIGN(cx + cy) * (cx - cy) * .7;
333 y = s->h - s->h * fabsf(cx + cy) * .7;
336 if (s->draw == DOT) {
339 draw_line(s, x, y, prev_x, prev_y);
345 s->prev_x = x, s->prev_y = y;
346 av_frame_free(&insamples);
348 return ff_filter_frame(outlink, av_frame_clone(s->outpicref));
351 static av_cold void uninit(AVFilterContext *ctx)
353 AudioVectorScopeContext *s = ctx->priv;
355 av_frame_free(&s->outpicref);
358 static const AVFilterPad audiovectorscope_inputs[] = {
361 .type = AVMEDIA_TYPE_AUDIO,
362 .config_props = config_input,
363 .filter_frame = filter_frame,
368 static const AVFilterPad audiovectorscope_outputs[] = {
371 .type = AVMEDIA_TYPE_VIDEO,
372 .config_props = config_output,
377 AVFilter ff_avf_avectorscope = {
378 .name = "avectorscope",
379 .description = NULL_IF_CONFIG_SMALL("Convert input audio to vectorscope video output."),
381 .query_formats = query_formats,
382 .priv_size = sizeof(AudioVectorScopeContext),
383 .inputs = audiovectorscope_inputs,
384 .outputs = audiovectorscope_outputs,
385 .priv_class = &avectorscope_class,