FFmpeg
vf_xmedian.c
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2019 Paul B Mahol
3  *
4  * This file is part of FFmpeg.
5  *
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.
10  *
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.
15  *
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
19  */
20 
21 #include "libavutil/avstring.h"
22 #include "libavutil/imgutils.h"
23 #include "libavutil/intreadwrite.h"
24 #include "libavutil/opt.h"
25 #include "libavutil/pixdesc.h"
26 #include "libavutil/qsort.h"
27 
28 #include "avfilter.h"
29 #include "formats.h"
30 #include "internal.h"
31 #include "framesync.h"
32 #include "video.h"
33 
34 typedef struct XMedianContext {
35  const AVClass *class;
37  int nb_inputs;
38  int planes;
39  float percentile;
40 
41  int radius;
42  int index;
43  int depth;
44  int max;
45  int nb_planes;
46  int linesize[4];
47  int width[4];
48  int height[4];
49 
52 
53  int (*median_frames)(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs);
55 
57 {
58  static const enum AVPixelFormat pixel_fmts[] = {
86  };
88  if (!formats)
89  return AVERROR(ENOMEM);
90  return ff_set_common_formats(ctx, formats);
91 }
92 
94 {
95  XMedianContext *s = ctx->priv;
96  int ret;
97 
98  s->radius = s->nb_inputs / 2;
99  if (s->nb_inputs & 1)
100  s->index = s->radius * 2.f * s->percentile;
101  else
102  s->index = av_clip(s->radius * 2.f * s->percentile, 1, s->nb_inputs - 1);
103  s->frames = av_calloc(s->nb_inputs, sizeof(*s->frames));
104  if (!s->frames)
105  return AVERROR(ENOMEM);
106 
107  for (int i = 0; i < s->nb_inputs; i++) {
108  AVFilterPad pad = { 0 };
109 
110  pad.type = AVMEDIA_TYPE_VIDEO;
111  pad.name = av_asprintf("input%d", i);
112  if (!pad.name)
113  return AVERROR(ENOMEM);
114 
115  if ((ret = ff_insert_inpad(ctx, i, &pad)) < 0) {
116  av_freep(&pad.name);
117  return ret;
118  }
119  }
120 
121  return 0;
122 }
123 
124 typedef struct ThreadData {
125  AVFrame **in, *out;
126 } ThreadData;
127 
128 static int comparei(const void *p1, const void *p2)
129 {
130  int left = *(const int *)p1;
131  int right = *(const int *)p2;
132  return FFDIFFSIGN(left, right);
133 }
134 
135 static int median_frames16(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs)
136 {
137  XMedianContext *s = ctx->priv;
138  ThreadData *td = arg;
139  AVFrame **in = td->in;
140  AVFrame *out = td->out;
141  const int nb_inputs = s->nb_inputs;
142  const int radius = s->radius;
143  const int index = s->index;
144  int values[256];
145 
146  for (int p = 0; p < s->nb_planes; p++) {
147  const int slice_start = (s->height[p] * jobnr) / nb_jobs;
148  const int slice_end = (s->height[p] * (jobnr+1)) / nb_jobs;
149  uint16_t *dst = (uint16_t *)(out->data[p] + slice_start * out->linesize[p]);
150 
151  if (!((1 << p) & s->planes)) {
152  av_image_copy_plane((uint8_t *)dst, out->linesize[p],
153  in[radius]->data[p] + slice_start * in[radius]->linesize[p],
154  in[radius]->linesize[p],
155  s->linesize[p], slice_end - slice_start);
156  continue;
157  }
158 
159  for (int y = slice_start; y < slice_end; y++) {
160  for (int x = 0; x < s->width[p]; x++) {
161  for (int i = 0; i < nb_inputs; i++) {
162  const uint16_t *src = (const uint16_t *)(in[i]->data[p] + y * in[i]->linesize[p]);
163  values[i] = src[x];
164  }
165 
166  AV_QSORT(values, nb_inputs, int, comparei);
167  if (nb_inputs & 1)
168  dst[x] = values[index];
169  else
170  dst[x] = (values[index] + values[index - 1]) >> 1;
171  }
172 
173  dst += out->linesize[p] / 2;
174  }
175  }
176 
177  return 0;
178 }
179 
180 static int median_frames8(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs)
181 {
182  XMedianContext *s = ctx->priv;
183  ThreadData *td = arg;
184  AVFrame **in = td->in;
185  AVFrame *out = td->out;
186  const int nb_inputs = s->nb_inputs;
187  const int radius = s->radius;
188  const int index = s->index;
189  int values[256];
190 
191  for (int p = 0; p < s->nb_planes; p++) {
192  const int slice_start = (s->height[p] * jobnr) / nb_jobs;
193  const int slice_end = (s->height[p] * (jobnr+1)) / nb_jobs;
194  uint8_t *dst = out->data[p] + slice_start * out->linesize[p];
195 
196  if (!((1 << p) & s->planes)) {
197  av_image_copy_plane(dst, out->linesize[p],
198  in[radius]->data[p] + slice_start * in[radius]->linesize[p],
199  in[radius]->linesize[p],
200  s->linesize[p], slice_end - slice_start);
201  continue;
202  }
203 
204  for (int y = slice_start; y < slice_end; y++) {
205  for (int x = 0; x < s->width[p]; x++) {
206  for (int i = 0; i < nb_inputs; i++)
207  values[i] = in[i]->data[p][y * in[i]->linesize[p] + x];
208 
209  AV_QSORT(values, nb_inputs, int, comparei);
210  if (nb_inputs & 1)
211  dst[x] = values[index];
212  else
213  dst[x] = (values[index] + values[index - 1]) >> 1;
214  }
215 
216  dst += out->linesize[p];
217  }
218  }
219 
220  return 0;
221 }
222 
224 {
225  AVFilterContext *ctx = fs->parent;
226  AVFilterLink *outlink = ctx->outputs[0];
227  XMedianContext *s = fs->opaque;
228  AVFrame **in = s->frames;
229  AVFrame *out;
230  ThreadData td;
231  int i, ret;
232 
233  for (i = 0; i < s->nb_inputs; i++) {
234  if ((ret = ff_framesync_get_frame(&s->fs, i, &in[i], 0)) < 0)
235  return ret;
236  }
237 
238  out = ff_get_video_buffer(outlink, outlink->w, outlink->h);
239  if (!out)
240  return AVERROR(ENOMEM);
241  out->pts = av_rescale_q(s->fs.pts, s->fs.time_base, outlink->time_base);
242 
243  td.in = in;
244  td.out = out;
245  ctx->internal->execute(ctx, s->median_frames, &td, NULL, FFMIN(s->height[1], ff_filter_get_nb_threads(ctx)));
246 
247  return ff_filter_frame(outlink, out);
248 }
249 
250 static int config_output(AVFilterLink *outlink)
251 {
252  AVFilterContext *ctx = outlink->src;
253  XMedianContext *s = ctx->priv;
254  AVRational frame_rate = ctx->inputs[0]->frame_rate;
255  AVRational sar = ctx->inputs[0]->sample_aspect_ratio;
256  AVFilterLink *inlink = ctx->inputs[0];
257  int height = ctx->inputs[0]->h;
258  int width = ctx->inputs[0]->w;
259  FFFrameSyncIn *in;
260  int i, ret;
261 
262  for (int i = 1; i < s->nb_inputs; i++) {
263  if (ctx->inputs[i]->h != height || ctx->inputs[i]->w != width) {
264  av_log(ctx, AV_LOG_ERROR, "Input %d size (%dx%d) does not match input %d size (%dx%d).\n", i, ctx->inputs[i]->w, ctx->inputs[i]->h, 0, width, height);
265  return AVERROR(EINVAL);
266  }
267  }
268 
269  s->desc = av_pix_fmt_desc_get(outlink->format);
270  if (!s->desc)
271  return AVERROR_BUG;
273  s->depth = s->desc->comp[0].depth;
274  s->max = (1 << s->depth) - 1;
275 
276  if (s->depth <= 8)
278  else
280 
281  if ((ret = av_image_fill_linesizes(s->linesize, inlink->format, inlink->w)) < 0)
282  return ret;
283 
284  s->width[1] = s->width[2] = AV_CEIL_RSHIFT(inlink->w, s->desc->log2_chroma_w);
285  s->width[0] = s->width[3] = inlink->w;
286  s->height[1] = s->height[2] = AV_CEIL_RSHIFT(inlink->h, s->desc->log2_chroma_h);
287  s->height[0] = s->height[3] = inlink->h;
288 
289  outlink->w = width;
290  outlink->h = height;
291  outlink->frame_rate = frame_rate;
292  outlink->sample_aspect_ratio = sar;
293 
294  if ((ret = ff_framesync_init(&s->fs, ctx, s->nb_inputs)) < 0)
295  return ret;
296 
297  in = s->fs.in;
298  s->fs.opaque = s;
300 
301  for (i = 0; i < s->nb_inputs; i++) {
302  AVFilterLink *inlink = ctx->inputs[i];
303 
304  in[i].time_base = inlink->time_base;
305  in[i].sync = 1;
306  in[i].before = EXT_STOP;
307  in[i].after = EXT_STOP;
308  }
309 
310  ret = ff_framesync_configure(&s->fs);
311  outlink->time_base = s->fs.time_base;
312 
313  return ret;
314 }
315 
317 {
318  XMedianContext *s = ctx->priv;
319 
320  ff_framesync_uninit(&s->fs);
321  av_freep(&s->frames);
322 
323  for (int i = 0; i < ctx->nb_inputs; i++)
324  av_freep(&ctx->input_pads[i].name);
325 }
326 
328 {
329  XMedianContext *s = ctx->priv;
330  return ff_framesync_activate(&s->fs);
331 }
332 
333 #define OFFSET(x) offsetof(XMedianContext, x)
334 #define FLAGS AV_OPT_FLAG_VIDEO_PARAM | AV_OPT_FLAG_FILTERING_PARAM
335 
336 static const AVOption xmedian_options[] = {
337  { "inputs", "set number of inputs", OFFSET(nb_inputs), AV_OPT_TYPE_INT, {.i64=3}, 3, 255, .flags = FLAGS },
338  { "planes", "set planes to filter", OFFSET(planes), AV_OPT_TYPE_INT, {.i64=15}, 0, 15, .flags = FLAGS },
339  { "percentile", "set percentile", OFFSET(percentile),AV_OPT_TYPE_FLOAT,{.dbl=0.5}, 0, 1, .flags = FLAGS },
340  { NULL },
341 };
342 
343 static const AVFilterPad outputs[] = {
344  {
345  .name = "default",
346  .type = AVMEDIA_TYPE_VIDEO,
347  .config_props = config_output,
348  },
349  { NULL }
350 };
351 
352 AVFILTER_DEFINE_CLASS(xmedian);
353 
355  .name = "xmedian",
356  .description = NULL_IF_CONFIG_SMALL("Pick median pixels from several video inputs."),
357  .priv_size = sizeof(XMedianContext),
358  .priv_class = &xmedian_class,
360  .outputs = outputs,
361  .init = init,
362  .uninit = uninit,
363  .activate = activate,
365 };
#define NULL
Definition: coverity.c:32
#define AV_PIX_FMT_YUVA422P16
Definition: pixfmt.h:440
AVFrame * out
Definition: af_adeclick.c:494
#define AV_PIX_FMT_YUV440P10
Definition: pixfmt.h:399
#define AV_PIX_FMT_YUVA422P9
Definition: pixfmt.h:432
const AVPixFmtDescriptor * av_pix_fmt_desc_get(enum AVPixelFormat pix_fmt)
Definition: pixdesc.c:2549
This structure describes decoded (raw) audio or video data.
Definition: frame.h:295
#define OFFSET(x)
Definition: vf_xmedian.c:333
AVOption.
Definition: opt.h:246
ptrdiff_t const GLvoid * data
Definition: opengl_enc.c:100
#define AV_PIX_FMT_YUVA420P10
Definition: pixfmt.h:434
#define AV_PIX_FMT_YUV444P14
Definition: pixfmt.h:407
#define AV_PIX_FMT_GBRAP10
Definition: pixfmt.h:417
#define AV_PIX_FMT_YUVA422P10
Definition: pixfmt.h:435
planar YUV 4:4:4, 24bpp, (1 Cr & Cb sample per 1x1 Y samples)
Definition: pixfmt.h:71
misc image utilities
AVFILTER_DEFINE_CLASS(xmedian)
int av_pix_fmt_count_planes(enum AVPixelFormat pix_fmt)
Definition: pixdesc.c:2589
Main libavfilter public API header.
FFFrameSync fs
Definition: vf_xmedian.c:51
#define AVFILTER_FLAG_DYNAMIC_INPUTS
The number of the filter inputs is not determined just by AVFilter.inputs.
Definition: avfilter.h:105
planar GBR 4:4:4 24bpp
Definition: pixfmt.h:168
#define AV_PIX_FMT_GBRP10
Definition: pixfmt.h:413
#define AV_PIX_FMT_GRAY9
Definition: pixfmt.h:377
static int comparei(const void *p1, const void *p2)
Definition: vf_xmedian.c:128
enum AVMediaType type
AVFilterPad type.
Definition: internal.h:65
#define AV_PIX_FMT_YUV420P12
Definition: pixfmt.h:401
int ff_framesync_configure(FFFrameSync *fs)
Configure a frame sync structure.
Definition: framesync.c:117
AVFrame * ff_get_video_buffer(AVFilterLink *link, int w, int h)
Request a picture buffer with a specific set of permissions.
Definition: video.c:99
static int config_output(AVFilterLink *outlink)
Definition: vf_xmedian.c:250
#define src
Definition: vp8dsp.c:254
int64_t pts
Timestamp of the current event.
Definition: framesync.h:167
uint8_t log2_chroma_w
Amount to shift the luma width right to find the chroma width.
Definition: pixdesc.h:92
static int median_frames8(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs)
Definition: vf_xmedian.c:180
enum FFFrameSyncExtMode before
Extrapolation mode for timestamps before the first frame.
Definition: framesync.h:86
void * av_calloc(size_t nmemb, size_t size)
Non-inlined equivalent of av_mallocz_array().
Definition: mem.c:244
AVFilterFormats * ff_make_format_list(const int *fmts)
Create a list of supported formats.
Definition: formats.c:283
#define AV_PIX_FMT_GRAY10
Definition: pixfmt.h:378
const char * name
Pad name.
Definition: internal.h:60
AVFilterContext * parent
Parent filter context.
Definition: framesync.h:152
#define AV_PIX_FMT_GRAY12
Definition: pixfmt.h:379
AVFilterLink ** inputs
array of pointers to input links
Definition: avfilter.h:346
int height[4]
Definition: vf_xmedian.c:48
int ff_filter_frame(AVFilterLink *link, AVFrame *frame)
Send a frame of data to the next filter.
Definition: avfilter.c:1075
planar YUV 4:2:0, 20bpp, (1 Cr & Cb sample per 2x2 Y & A samples)
Definition: pixfmt.h:101
AVComponentDescriptor comp[4]
Parameters that describe how pixels are packed.
Definition: pixdesc.h:117
uint8_t
#define av_cold
Definition: attributes.h:82
AVOptions.
static int median_frames16(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs)
Definition: vf_xmedian.c:135
int64_t pts
Presentation timestamp in time_base units (time when frame should be shown to user).
Definition: frame.h:388
FFFrameSyncIn * in
Pointer to array of inputs.
Definition: framesync.h:203
#define AV_PIX_FMT_YUVA420P9
Definition: pixfmt.h:431
#define AV_PIX_FMT_GBRP9
Definition: pixfmt.h:412
int width[4]
Definition: vf_xmedian.c:47
planar YUV 4:4:0 full scale (JPEG), deprecated in favor of AV_PIX_FMT_YUV440P and setting color_range...
Definition: pixfmt.h:100
planar YUV 4:2:2, 16bpp, full scale (JPEG), deprecated in favor of AV_PIX_FMT_YUV422P and setting col...
Definition: pixfmt.h:79
#define AV_PIX_FMT_YUV444P16
Definition: pixfmt.h:410
enum FFFrameSyncExtMode after
Extrapolation mode for timestamps after the last frame.
Definition: framesync.h:91
Input stream structure.
Definition: framesync.h:81
#define AV_PIX_FMT_YUV422P12
Definition: pixfmt.h:402
#define AV_PIX_FMT_YUVA420P16
Definition: pixfmt.h:439
#define av_log(a,...)
A filter pad used for either input or output.
Definition: internal.h:54
int64_t av_rescale_q(int64_t a, AVRational bq, AVRational cq)
Rescale a 64-bit integer by 2 rational numbers.
Definition: mathematics.c:142
static const AVOption xmedian_options[]
Definition: vf_xmedian.c:336
AVFilterPad * input_pads
array of input pads
Definition: avfilter.h:345
planar YUV 4:2:2 24bpp, (1 Cr & Cb sample per 2x1 Y & A samples)
Definition: pixfmt.h:176
#define i(width, name, range_min, range_max)
Definition: cbs_h2645.c:259
#define AV_LOG_ERROR
Something went wrong and cannot losslessly be recovered.
Definition: log.h:176
int ff_set_common_formats(AVFilterContext *ctx, AVFilterFormats *formats)
A helper for query_formats() which sets all links to the same list of formats.
Definition: formats.c:569
#define td
Definition: regdef.h:70
uint8_t log2_chroma_h
Amount to shift the luma height right to find the chroma height.
Definition: pixdesc.h:101
void ff_framesync_uninit(FFFrameSync *fs)
Free all memory currently allocated.
Definition: framesync.c:293
Frame sync structure.
Definition: framesync.h:146
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification. ...
Definition: internal.h:186
AVFilter ff_vf_xmedian
Definition: vf_xmedian.c:354
void * priv
private data for use by the filter
Definition: avfilter.h:353
#define AVFILTER_FLAG_SLICE_THREADS
The filter supports multithreading by splitting frames into multiple parts and processing them concur...
Definition: avfilter.h:116
#define AV_PIX_FMT_YUVA444P16
Definition: pixfmt.h:441
const char * arg
Definition: jacosubdec.c:66
#define AV_PIX_FMT_GBRAP12
Definition: pixfmt.h:418
AVRational time_base
Time base for the incoming frames.
Definition: framesync.h:96
int ff_framesync_activate(FFFrameSync *fs)
Examine the frames in the filter&#39;s input and try to produce output.
Definition: framesync.c:344
int linesize[4]
Definition: vf_xmedian.c:46
#define AV_PIX_FMT_YUV444P10
Definition: pixfmt.h:400
int(* on_event)(struct FFFrameSync *fs)
Callback called when a frame event is ready.
Definition: framesync.h:172
#define AV_PIX_FMT_GBRAP16
Definition: pixfmt.h:419
static av_cold void uninit(AVFilterContext *ctx)
Definition: vf_xmedian.c:316
planar YUV 4:2:2, 16bpp, (1 Cr & Cb sample per 2x1 Y samples)
Definition: pixfmt.h:70
char * av_asprintf(const char *fmt,...)
Definition: avstring.c:113
#define FFDIFFSIGN(x, y)
Comparator.
Definition: common.h:92
#define AV_PIX_FMT_YUV422P9
Definition: pixfmt.h:395
static int activate(AVFilterContext *ctx)
Definition: vf_xmedian.c:327
#define AV_PIX_FMT_GBRP16
Definition: pixfmt.h:416
int ff_filter_get_nb_threads(AVFilterContext *ctx)
Get number of threads for current filter instance.
Definition: avfilter.c:784
unsigned nb_inputs
number of input pads
Definition: avfilter.h:347
#define AV_PIX_FMT_GRAY16
Definition: pixfmt.h:381
#define FFMIN(a, b)
Definition: common.h:96
static int query_formats(AVFilterContext *ctx)
Definition: vf_xmedian.c:56
planar YUV 4:2:0, 12bpp, full scale (JPEG), deprecated in favor of AV_PIX_FMT_YUV420P and setting col...
Definition: pixfmt.h:78
#define AV_PIX_FMT_YUVA444P12
Definition: pixfmt.h:438
these buffered frames must be flushed immediately if a new input produces new the filter must not call request_frame to get more It must just process the frame or queue it The task of requesting more frames is left to the filter s request_frame method or the application If a filter has several the filter must be ready for frames arriving randomly on any input any filter with several inputs will most likely require some kind of queuing mechanism It is perfectly acceptable to have a limited queue and to drop frames when the inputs are too unbalanced request_frame For filters that do not use the this method is called when a frame is wanted on an output For a it should directly call filter_frame on the corresponding output For a if there are queued frames already one of these frames should be pushed If the filter should request a frame on one of its repeatedly until at least one frame has been pushed Return values
AVFormatContext * ctx
Definition: movenc.c:48
AVRational time_base
Time base for the output events.
Definition: framesync.h:162
#define s(width, name)
Definition: cbs_vp9.c:257
#define AV_PIX_FMT_YUVA444P10
Definition: pixfmt.h:436
void * opaque
Opaque pointer, not used by the API.
Definition: framesync.h:177
#define AV_PIX_FMT_YUV444P9
Definition: pixfmt.h:396
static av_cold int init(AVFilterContext *ctx)
Definition: vf_xmedian.c:93
#define AV_PIX_FMT_GBRP14
Definition: pixfmt.h:415
#define AV_PIX_FMT_YUV420P16
Definition: pixfmt.h:408
#define AV_PIX_FMT_YUV420P14
Definition: pixfmt.h:405
Used for passing data between threads.
Definition: dsddec.c:64
int linesize[AV_NUM_DATA_POINTERS]
For video, size in bytes of each picture line.
Definition: frame.h:326
int ff_framesync_init(FFFrameSync *fs, AVFilterContext *parent, unsigned nb_in)
Initialize a frame sync structure.
Definition: framesync.c:77
planar YUV 4:4:4 32bpp, (1 Cr & Cb sample per 1x1 Y & A samples)
Definition: pixfmt.h:177
Descriptor that unambiguously describes how the bits of a pixel are stored in the up to 4 data planes...
Definition: pixdesc.h:81
#define AV_PIX_FMT_GRAY14
Definition: pixfmt.h:380
Tag MUST be and< 10hcoeff half pel interpolation filter coefficients, hcoeff[0] are the 2 middle coefficients[1] are the next outer ones and so on, resulting in a filter like:...eff[2], hcoeff[1], hcoeff[0], hcoeff[0], hcoeff[1], hcoeff[2]...the sign of the coefficients is not explicitly stored but alternates after each coeff and coeff[0] is positive, so...,+,-,+,-,+,+,-,+,-,+,...hcoeff[0] is not explicitly stored but found by subtracting the sum of all stored coefficients with signs from 32 hcoeff[0]=32-hcoeff[1]-hcoeff[2]-...a good choice for hcoeff and htaps is htaps=6 hcoeff={40,-10, 2}an alternative which requires more computations at both encoder and decoder side and may or may not be better is htaps=8 hcoeff={42,-14, 6,-2}ref_frames minimum of the number of available reference frames and max_ref_frames for example the first frame after a key frame always has ref_frames=1spatial_decomposition_type wavelet type 0 is a 9/7 symmetric compact integer wavelet 1 is a 5/3 symmetric compact integer wavelet others are reserved stored as delta from last, last is reset to 0 if always_reset||keyframeqlog quality(logarithmic quantizer scale) stored as delta from last, last is reset to 0 if always_reset||keyframemv_scale stored as delta from last, last is reset to 0 if always_reset||keyframe FIXME check that everything works fine if this changes between framesqbias dequantization bias stored as delta from last, last is reset to 0 if always_reset||keyframeblock_max_depth maximum depth of the block tree stored as delta from last, last is reset to 0 if always_reset||keyframequant_table quantization tableHighlevel bitstream structure:==============================--------------------------------------------|Header|--------------------------------------------|------------------------------------|||Block0||||split?||||yes no||||.........intra?||||:Block01:yes no||||:Block02:.................||||:Block03::y DC::ref index:||||:Block04::cb DC::motion x:||||.........:cr DC::motion y:||||.................|||------------------------------------||------------------------------------|||Block1|||...|--------------------------------------------|------------------------------------|||Y subbands||Cb subbands||Cr subbands||||------||------||------|||||LL0||HL0||||LL0||HL0||||LL0||HL0|||||------||------||------||||------||------||------|||||LH0||HH0||||LH0||HH0||||LH0||HH0|||||------||------||------||||------||------||------|||||HL1||LH1||||HL1||LH1||||HL1||LH1|||||------||------||------||||------||------||------|||||HH1||HL2||||HH1||HL2||||HH1||HL2|||||...||...||...|||------------------------------------|--------------------------------------------Decoding process:=================------------|||Subbands|------------||||------------|Intra DC||||LL0 subband prediction------------|\Dequantization-------------------\||Reference frames|\IDWT|--------------|Motion\|||Frame 0||Frame 1||Compensation.OBMC v-------|--------------|--------------.\------> Frame n output Frame Frame<----------------------------------/|...|-------------------Range Coder:============Binary Range Coder:-------------------The implemented range coder is an adapted version based upon"Range encoding: an algorithm for removing redundancy from a digitised message."by G.N.N.Martin.The symbols encoded by the Snow range coder are bits(0|1).The associated probabilities are not fix but change depending on the symbol mix seen so far.bit seen|new state---------+-----------------------------------------------0|256-state_transition_table[256-old_state];1|state_transition_table[old_state];state_transition_table={0, 0, 0, 0, 0, 0, 0, 0, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 114, 115, 116, 117, 118, 119, 120, 121, 122, 123, 124, 125, 126, 127, 128, 129, 130, 131, 132, 133, 133, 134, 135, 136, 137, 138, 139, 140, 141, 142, 143, 144, 145, 146, 147, 148, 149, 150, 151, 152, 152, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 171, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 190, 191, 192, 194, 194, 195, 196, 197, 198, 199, 200, 201, 202, 202, 204, 205, 206, 207, 208, 209, 209, 210, 211, 212, 213, 215, 215, 216, 217, 218, 219, 220, 220, 222, 223, 224, 225, 226, 227, 227, 229, 229, 230, 231, 232, 234, 234, 235, 236, 237, 238, 239, 240, 241, 242, 243, 244, 245, 246, 247, 248, 248, 0, 0, 0, 0, 0, 0, 0};FIXME Range Coding of integers:-------------------------FIXME Neighboring Blocks:===================left and top are set to the respective blocks unless they are outside of the image in which case they are set to the Null block top-left is set to the top left block unless it is outside of the image in which case it is set to the left block if this block has no larger parent block or it is at the left side of its parent block and the top right block is not outside of the image then the top right block is used for top-right else the top-left block is used Null block y, cb, cr are 128 level, ref, mx and my are 0 Motion Vector Prediction:=========================1.the motion vectors of all the neighboring blocks are scaled to compensate for the difference of reference frames scaled_mv=(mv *(256 *(current_reference+1)/(mv.reference+1))+128)> the median of the scaled left
Definition: snow.txt:206
int(* median_frames)(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs)
Definition: vf_xmedian.c:53
#define AVERROR_BUG
Internal bug, also see AVERROR_BUG2.
Definition: error.h:50
AVFrame ** frames
Definition: vf_xmedian.c:50
uint8_t pi<< 24) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_U8, uint8_t,(*(const uint8_t *) pi-0x80)*(1.0f/(1<< 7))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_U8, uint8_t,(*(const uint8_t *) pi-0x80)*(1.0/(1<< 7))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_S16, int16_t,(*(const int16_t *) pi >> 8)+0x80) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_S16, int16_t,*(const int16_t *) pi *(1.0f/(1<< 15))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_S16, int16_t,*(const int16_t *) pi *(1.0/(1<< 15))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_S32, int32_t,(*(const int32_t *) pi >> 24)+0x80) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_S32, int32_t,*(const int32_t *) pi *(1.0f/(1U<< 31))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_S32, int32_t,*(const int32_t *) pi *(1.0/(1U<< 31))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_FLT, float, av_clip_uint8(lrintf(*(const float *) pi *(1<< 7))+0x80)) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S16, int16_t, AV_SAMPLE_FMT_FLT, float, av_clip_int16(lrintf(*(const float *) pi *(1<< 15)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S32, int32_t, AV_SAMPLE_FMT_FLT, float, av_clipl_int32(llrintf(*(const float *) pi *(1U<< 31)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_DBL, double, av_clip_uint8(lrint(*(const double *) pi *(1<< 7))+0x80)) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S16, int16_t, AV_SAMPLE_FMT_DBL, double, av_clip_int16(lrint(*(const double *) pi *(1<< 15)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S32, int32_t, AV_SAMPLE_FMT_DBL, double, av_clipl_int32(llrint(*(const double *) pi *(1U<< 31))))#define SET_CONV_FUNC_GROUP(ofmt, ifmt) static void set_generic_function(AudioConvert *ac){}void ff_audio_convert_free(AudioConvert **ac){if(!*ac) return;ff_dither_free(&(*ac) ->dc);av_freep(ac);}AudioConvert *ff_audio_convert_alloc(AVAudioResampleContext *avr, enum AVSampleFormat out_fmt, enum AVSampleFormat in_fmt, int channels, int sample_rate, int apply_map){AudioConvert *ac;int in_planar, out_planar;ac=av_mallocz(sizeof(*ac));if(!ac) return NULL;ac->avr=avr;ac->out_fmt=out_fmt;ac->in_fmt=in_fmt;ac->channels=channels;ac->apply_map=apply_map;if(avr->dither_method!=AV_RESAMPLE_DITHER_NONE &&av_get_packed_sample_fmt(out_fmt)==AV_SAMPLE_FMT_S16 &&av_get_bytes_per_sample(in_fmt) > 2){ac->dc=ff_dither_alloc(avr, out_fmt, in_fmt, channels, sample_rate, apply_map);if(!ac->dc){av_free(ac);return NULL;}return ac;}in_planar=ff_sample_fmt_is_planar(in_fmt, channels);out_planar=ff_sample_fmt_is_planar(out_fmt, channels);if(in_planar==out_planar){ac->func_type=CONV_FUNC_TYPE_FLAT;ac->planes=in_planar?ac->channels:1;}else if(in_planar) ac->func_type=CONV_FUNC_TYPE_INTERLEAVE;else ac->func_type=CONV_FUNC_TYPE_DEINTERLEAVE;set_generic_function(ac);if(ARCH_AARCH64) ff_audio_convert_init_aarch64(ac);if(ARCH_ARM) ff_audio_convert_init_arm(ac);if(ARCH_X86) ff_audio_convert_init_x86(ac);return ac;}int ff_audio_convert(AudioConvert *ac, AudioData *out, AudioData *in){int use_generic=1;int len=in->nb_samples;int p;if(ac->dc){av_log(ac->avr, AV_LOG_TRACE,"%d samples - audio_convert: %s to %s (dithered)\n", len, av_get_sample_fmt_name(ac->in_fmt), av_get_sample_fmt_name(ac->out_fmt));return ff_convert_dither(ac-> in
#define AV_PIX_FMT_YUV420P10
Definition: pixfmt.h:397
planar YUV 4:1:0, 9bpp, (1 Cr & Cb sample per 4x4 Y samples)
Definition: pixfmt.h:72
unsigned sync
Synchronization level: frames on input at the highest sync level will generate output frame events...
Definition: framesync.h:139
Describe the class of an AVClass context structure.
Definition: log.h:67
Filter definition.
Definition: avfilter.h:144
int av_image_fill_linesizes(int linesizes[4], enum AVPixelFormat pix_fmt, int width)
Fill plane linesizes for an image with pixel format pix_fmt and width width.
Definition: imgutils.c:89
Rational number (pair of numerator and denominator).
Definition: rational.h:58
const char * name
Filter name.
Definition: avfilter.h:148
#define AV_PIX_FMT_YUV440P12
Definition: pixfmt.h:403
#define AV_PIX_FMT_YUV420P9
Definition: pixfmt.h:394
AVFilterLink ** outputs
array of pointers to output links
Definition: avfilter.h:350
static int process_frame(FFFrameSync *fs)
Definition: vf_xmedian.c:223
static const AVFilterPad outputs[]
Definition: vf_xmedian.c:343
#define AV_PIX_FMT_YUV422P14
Definition: pixfmt.h:406
#define AV_PIX_FMT_GBRP12
Definition: pixfmt.h:414
#define flags(name, subs,...)
Definition: cbs_av1.c:564
AVFilterInternal * internal
An opaque struct for libavfilter internal use.
Definition: avfilter.h:378
#define AV_PIX_FMT_YUV422P10
Definition: pixfmt.h:398
#define AV_PIX_FMT_YUV444P12
Definition: pixfmt.h:404
uint8_t * data[AV_NUM_DATA_POINTERS]
pointer to the picture/channel planes.
Definition: frame.h:309
The exact code depends on how similar the blocks are and how related they are to the and needs to apply these operations to the correct inlink or outlink if there are several Macros are available to factor that when no extra processing is inlink
int
planar YUV 4:2:0, 12bpp, (1 Cr & Cb sample per 2x2 Y samples)
Definition: pixfmt.h:66
Y , 8bpp.
Definition: pixfmt.h:74
planar GBRA 4:4:4:4 32bpp
Definition: pixfmt.h:215
float percentile
Definition: vf_xmedian.c:39
#define AV_PIX_FMT_YUVA444P9
Definition: pixfmt.h:433
planar YUV 4:4:4, 24bpp, full scale (JPEG), deprecated in favor of AV_PIX_FMT_YUV444P and setting col...
Definition: pixfmt.h:80
planar YUV 4:1:1, 12bpp, (1 Cr & Cb sample per 4x1 Y samples)
Definition: pixfmt.h:73
avfilter_execute_func * execute
Definition: internal.h:144
static int slice_end(AVCodecContext *avctx, AVFrame *pict)
Handle slice ends.
Definition: mpeg12dec.c:2043
Completely stop all streams with this one.
Definition: framesync.h:65
A list of supported formats for one end of a filter link.
Definition: formats.h:64
planar YUV 4:1:1, 12bpp, (1 Cr & Cb sample per 4x1 Y samples) full scale (JPEG), deprecated in favor ...
Definition: pixfmt.h:258
An instance of a filter.
Definition: avfilter.h:338
FILE * out
Definition: movenc.c:54
#define av_freep(p)
planar YUV 4:4:0 (1 Cr & Cb sample per 1x2 Y samples)
Definition: pixfmt.h:99
AVFrame * in
Definition: af_afftdn.c:1083
formats
Definition: signature.h:48
void av_image_copy_plane(uint8_t *dst, int dst_linesize, const uint8_t *src, int src_linesize, int bytewidth, int height)
Copy image plane from src to dst.
Definition: imgutils.c:338
internal API functions
Filter the word “frame” indicates either a video frame or a group of audio as stored in an AVFrame structure Format for each input and each output the list of supported formats For video that means pixel format For audio that means channel sample they are references to shared objects When the negotiation mechanism computes the intersection of the formats supported at each end of a all references to both lists are replaced with a reference to the intersection And when a single format is eventually chosen for a link amongst the remaining all references to the list are updated That means that if a filter requires that its input and output have the same format amongst a supported all it has to do is use a reference to the same list of formats query_formats can leave some formats unset and return AVERROR(EAGAIN) to cause the negotiation mechanism toagain later.That can be used by filters with complex requirements to use the format negotiated on one link to set the formats supported on another.Frame references ownership and permissions
int ff_framesync_get_frame(FFFrameSync *fs, unsigned in, AVFrame **rframe, unsigned get)
Get the current frame in an input.
Definition: framesync.c:256
int depth
Number of bits in the component.
Definition: pixdesc.h:58
#define FLAGS
Definition: vf_xmedian.c:334
AVPixelFormat
Pixel format.
Definition: pixfmt.h:64
const AVPixFmtDescriptor * desc
Definition: vf_xmedian.c:36
#define AV_PIX_FMT_YUV422P16
Definition: pixfmt.h:409
#define AV_QSORT(p, num, type, cmp)
Quicksort This sort is fast, and fully inplace but not stable and it is possible to construct input t...
Definition: qsort.h:33
#define AV_PIX_FMT_YUVA422P12
Definition: pixfmt.h:437
#define AV_CEIL_RSHIFT(a, b)
Definition: common.h:58
static int ff_insert_inpad(AVFilterContext *f, unsigned index, AVFilterPad *p)
Insert a new input pad for the filter.
Definition: internal.h:266