1 <!-- Author's note; this was intended to become a home page at some point,
2 but I'm not interested enough in grokking HTML right now, so it became
3 the README instead. Most of it should be valid Markdown. -->
8 Movit is the Modern Video Toolkit, notwithstanding that anything that's
9 called “modern” usually isn't, and it's really not a toolkit.
11 Movit aims to be a _high-quality_, _high-performance_, _open-source_
12 library for video filters. It is currently in alpha stage.
15 TL;DR, please give me download link and system demands
16 ======================================================
20 * A C++98 compiler. GCC will do. (I haven't tried Windows, but it
21 works fine on Linux and OS X, and Movit is not very POSIX-bound.)
23 * A GPU capable of running GLSL fragment shaders,
24 process floating-point textures, and a few other things. If your machine
25 is less than five years old _and you have the appropriate drivers_,
27 * The [Eigen 3] and [Google Test] libraries. (The library itself
28 depends only on the former, but you probably want to run the unit tests.)
29 * The [GLEW] library, for dealing with OpenGL extensions on various
32 Movit has been tested with Intel GPUs with the Mesa drivers
33 (you'll probably need at least Mesa 8.0), Radeon 3850 and GeForce GTX 550
34 on Linux with the manufacturer's drivers, and with GeForce 8800 on OS X.
35 Again, most likely, GPU compatibility shouldn't be a big issue. See below
36 for performance estimates.
39 Still TL;DR, please give me the list of filters
40 ===============================================
42 Blur, diffusion, glow, lift/gamma/gain (color correction), mirror,
43 mix (add two inputs), overlay (the Porter-Duff “over” operation),
44 scale (bilinear and Lanczos), sharpen (both by unsharp mask and by
45 Wiener filters), saturation (or desaturation), vignette, and white balance.
47 Yes, that's a short list. But they all look great, are fast and don't give
48 you any nasty surprises. (I'd love to include denoise, deinterlace and
49 framerate up-/downconversion to the list, but doing them well are
50 all research-grade problems, and Movit is currently not there.)
53 TL;DR, but I am interested in a programming example instead
54 ===========================================================
56 Assuming you have an OpenGL context already set up:
59 using namespace movit;
60 EffectChain chain(1280, 720);
62 ImageFormat inout_format;
63 inout_format.color_space = COLORSPACE_sRGB;
64 inout_format.gamma_curve = GAMMA_sRGB;
65 FlatInput *input = new FlatInput(inout_format, FORMAT_BGRA_POSTMULTIPLIED_ALPHA, GL_UNSIGNED_BYTE, 1280, 720));
66 chain.add_input(input);
68 Effect *saturation_effect = chain.add_effect(new SaturationEffect());
69 saturation_effect->set_float("saturation", 0.7f);
71 Effect *lift_gamma_gain_effect = chain.add_effect(new LiftGammaGainEffect());
72 const float gain[] = { 0.8f, 1.0f, 1.0f };
73 lift_gamma_gain_effect->set_vec3("gain", &gain);
75 chain.add_output(inout_format, OUTPUT_ALPHA_FORMAT_POSTMULTIPLIED);
79 // Do whatever you need here to decode the next frame into <pixels>.
80 input->set_pixel_data(pixels);
81 chain.render_to_screen();
86 OK, I can read a bit. What do you mean by “modern”?
87 ===================================================
89 Backwards compatibility is fine and all, but sometimes we can do better
90 by observing that the world has moved on. In particular:
92 * It's 2014, so people want to edit HD video.
93 * It's 2014, so everybody has a GPU.
94 * It's 2014, so everybody has a working C++ compiler.
95 (Even Microsoft fixed theirs around 2003!)
97 While from a programming standpoint I'd love to say that it's 2014
98 and interlacing does no longer exist, but that's not true (and interlacing,
99 hated as it might be, is actually a useful and underrated technique for
100 bandwidth reduction in broadcast video). Movit will eventually provide
101 limited support for working with interlaced video, but currently does not.
104 What do you mean by “high-performance”?
105 =======================================
107 Today, you can hardly get a _cellphone_ without a multi-core, SIMD-capable
108 CPU, and a GPU. Yet, almost all open-source pixel processing I've seen
109 is written using straight-up single-threaded, scalar C! Clearly there is
110 room for improvement here, and that improvement is sorely needed.
111 We want to edit 1080p video, not watch slideshows.
113 Movit has chosen to run all pixel processing on the GPU, using GLSL—OpenCL is
114 way too young, and CUDA is single-vendor (and also surprisingly hard to
115 get good performance from for anything nontrivial). While “run on the GPU”
116 does not equal “infinite speed” (I am fairly certain that for many common
117 filters, I can beat the Intel-based GPU in my laptop with multithreaded SSE
118 code on the CPU—especially as moving the data to and from the GPU has a cost that is not
119 to be taken lightly), GPU programming is probably the _simplest_ way of writing
120 highly parallel code, and it also frees the CPU to do other things like video
123 Exactly what speeds you can expect is of course highly dependent on
124 your GPU and the exact filter chain you are running. As a rule of thumb,
125 you can run a reasonable filter chain (a lift/gamma/gain operation,
126 a bit of diffusion, maybe a vignette) at 720p in around 30 fps on a two-year-old
127 Intel laptop. If you have a somewhat newer Intel card, you can do 1080p
128 video without much problems. And on a mid-range nVidia card of today
129 (GTX 550 Ti), you can probably process 4K movies directly.
132 What do you mean by “high-quality”?
133 ===================================
135 Movit aims to be high-quality in two important aspects, namely _code quality_
136 and _output quality_. (Unfortunately, documentation quality is not on the
143 Movit works internally in linear floating-point all the way, strongly
144 reducing interim round-off and clipping errors. Furthermore, Movit is
145 (weakly) colorspace-aware. Why do colorspaces matter? Well, here's a video frame from a typical
146 camera, which records in Rec. 709 (the typical HDTV color space), and here's the
147 same frame misinterpreted as Rec. 601 (the typical SDTV color space):
149 [insert picture here]
151 The difference might be subtle, but would you like that color cast?
152 Maybe you could correct for it manually, but what if it happened on output
153 instead of on input? And I can promise you that once we move to more
154 wide-gamut color spaces, like the one in Rec. 2020 (used for UHDTV), the
155 difference will be anything but subtle. As of [why working in linear
156 light matters](http://www.4p8.com/eric.brasseur/gamma.html),
157 others have explained it better than I can; note also that this makes Movit
158 future-proof when the world moves towards 10- and 12-bit color precision
159 (although the latter requires Movit to change from 16-bit to 32-bit floating
160 point, it is a simple switch). The extra power from the GPU makes all of this
161 simple, so do we not need to make too many concessions for the sake of speed.
163 Movit does not currently do ICC profiles or advanced gamut mapping;
164 if you have out-of-gamut colors, they will clip. Sorry.
167 OK, and high-quality code?
168 ==========================
170 Image processing code can be surprisingly subtle; it's easy to write
171 code that looks right, but that makes subtle artifacts that explode
172 when processed further in a later step. (Or code that simply never
173 worked, just that nobody cared to look at the output when a given
174 parameter was set. I've seen that, too.)
176 Movit tries to counteract this by three different strategies:
178 * First, _look at the output_. Does it look good? Really?
179 Even if you zoom in on the results? Don't settle for “meh, I'm
180 sure that's the best it can get”.
181 * Second, _keep things simple_. Movit does not aim for including
182 every possible video effect under the sun (there are [others out there]
183 that want that); the [YAGNI] principle is applied quite strongly throughout
184 the code. It's much better to write less code but actually
185 understand what it does; whenever I can replace some magic matrix
186 or obscure formula from the web with a clean calculation and a descriptive
187 comment on top, it makes me a bit happier. (Most of the time,
188 it turns out that I had used the matrix or formula in a wrong
189 way anyway. My degree is in multimedia signal processing, but it
190 does not mean I have a deep understanding of everything people do
192 * Third, _have unit tests_. Tests are boring, but they are unforgiving
193 (much more unforgiving than your eye), and they keep stuff from breaking
194 afterwards. Almost every single test I wrote has uncovered bugs in Movit,
195 so they have already paid for themselves.
197 There is, of course, always room for improvement. I'm sure you can find
198 things that are stupid, little-thought-out, or buggy. If so, please let me
202 What do you mean by “open-source”?
203 ==================================
205 Movit is licensed under the [GNU GPL](http://www.gnu.org/licenses/gpl.html),
206 either version 2 or (at your option) any later version.