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 nb_frames;
39  int planes;
40  float percentile;
41 
42  int tmedian;
43  int radius;
44  int index;
45  int depth;
46  int max;
47  int nb_planes;
48  int linesize[4];
49  int width[4];
50  int height[4];
51 
54 
55  int (*median_frames)(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs);
57 
59 {
60  static const enum AVPixelFormat pixel_fmts[] = {
88  };
90  if (!formats)
91  return AVERROR(ENOMEM);
92  return ff_set_common_formats(ctx, formats);
93 }
94 
96 {
97  XMedianContext *s = ctx->priv;
98  int ret;
99 
100  s->tmedian = !strcmp(ctx->filter->name, "tmedian");
101 
102  if (!s->tmedian) {
103  s->radius = s->nb_inputs / 2;
104  } else {
105  s->nb_inputs = s->radius * 2 + 1;
106  }
107 
108  if (s->nb_inputs & 1)
109  s->index = s->radius * 2.f * s->percentile;
110  else
111  s->index = av_clip(s->radius * 2.f * s->percentile, 1, s->nb_inputs - 1);
112  s->frames = av_calloc(s->nb_inputs, sizeof(*s->frames));
113  if (!s->frames)
114  return AVERROR(ENOMEM);
115 
116  for (int i = 0; i < s->nb_inputs && !s->tmedian; i++) {
117  AVFilterPad pad = { 0 };
118 
119  pad.type = AVMEDIA_TYPE_VIDEO;
120  pad.name = av_asprintf("input%d", i);
121  if (!pad.name)
122  return AVERROR(ENOMEM);
123 
124  if ((ret = ff_insert_inpad(ctx, i, &pad)) < 0) {
125  av_freep(&pad.name);
126  return ret;
127  }
128  }
129 
130  return 0;
131 }
132 
133 typedef struct ThreadData {
134  AVFrame **in, *out;
135 } ThreadData;
136 
137 static int comparei(const void *p1, const void *p2)
138 {
139  int left = *(const int *)p1;
140  int right = *(const int *)p2;
141  return FFDIFFSIGN(left, right);
142 }
143 
144 static int median_frames16(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs)
145 {
146  XMedianContext *s = ctx->priv;
147  ThreadData *td = arg;
148  AVFrame **in = td->in;
149  AVFrame *out = td->out;
150  const int nb_inputs = s->nb_inputs;
151  const int radius = s->radius;
152  const int index = s->index;
153  int values[256];
154 
155  for (int p = 0; p < s->nb_planes; p++) {
156  const int slice_start = (s->height[p] * jobnr) / nb_jobs;
157  const int slice_end = (s->height[p] * (jobnr+1)) / nb_jobs;
158  uint16_t *dst = (uint16_t *)(out->data[p] + slice_start * out->linesize[p]);
159 
160  if (!((1 << p) & s->planes)) {
161  av_image_copy_plane((uint8_t *)dst, out->linesize[p],
162  in[radius]->data[p] + slice_start * in[radius]->linesize[p],
163  in[radius]->linesize[p],
164  s->linesize[p], slice_end - slice_start);
165  continue;
166  }
167 
168  for (int y = slice_start; y < slice_end; y++) {
169  for (int x = 0; x < s->width[p]; x++) {
170  for (int i = 0; i < nb_inputs; i++) {
171  const uint16_t *src = (const uint16_t *)(in[i]->data[p] + y * in[i]->linesize[p]);
172  values[i] = src[x];
173  }
174 
175  AV_QSORT(values, nb_inputs, int, comparei);
176  if (nb_inputs & 1)
177  dst[x] = values[index];
178  else
179  dst[x] = (values[index] + values[index - 1]) >> 1;
180  }
181 
182  dst += out->linesize[p] / 2;
183  }
184  }
185 
186  return 0;
187 }
188 
189 static int median_frames8(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs)
190 {
191  XMedianContext *s = ctx->priv;
192  ThreadData *td = arg;
193  AVFrame **in = td->in;
194  AVFrame *out = td->out;
195  const int nb_inputs = s->nb_inputs;
196  const int radius = s->radius;
197  const int index = s->index;
198  int values[256];
199 
200  for (int p = 0; p < s->nb_planes; p++) {
201  const int slice_start = (s->height[p] * jobnr) / nb_jobs;
202  const int slice_end = (s->height[p] * (jobnr+1)) / nb_jobs;
203  uint8_t *dst = out->data[p] + slice_start * out->linesize[p];
204 
205  if (!((1 << p) & s->planes)) {
206  av_image_copy_plane(dst, out->linesize[p],
207  in[radius]->data[p] + slice_start * in[radius]->linesize[p],
208  in[radius]->linesize[p],
209  s->linesize[p], slice_end - slice_start);
210  continue;
211  }
212 
213  for (int y = slice_start; y < slice_end; y++) {
214  for (int x = 0; x < s->width[p]; x++) {
215  for (int i = 0; i < nb_inputs; i++)
216  values[i] = in[i]->data[p][y * in[i]->linesize[p] + x];
217 
218  AV_QSORT(values, nb_inputs, int, comparei);
219  if (nb_inputs & 1)
220  dst[x] = values[index];
221  else
222  dst[x] = (values[index] + values[index - 1]) >> 1;
223  }
224 
225  dst += out->linesize[p];
226  }
227  }
228 
229  return 0;
230 }
231 
233 {
234  AVFilterContext *ctx = fs->parent;
235  AVFilterLink *outlink = ctx->outputs[0];
236  XMedianContext *s = fs->opaque;
237  AVFrame **in = s->frames;
238  AVFrame *out;
239  ThreadData td;
240  int i, ret;
241 
242  for (i = 0; i < s->nb_inputs; i++) {
243  if ((ret = ff_framesync_get_frame(&s->fs, i, &in[i], 0)) < 0)
244  return ret;
245  }
246 
247  out = ff_get_video_buffer(outlink, outlink->w, outlink->h);
248  if (!out)
249  return AVERROR(ENOMEM);
250  out->pts = av_rescale_q(s->fs.pts, s->fs.time_base, outlink->time_base);
251 
252  td.in = in;
253  td.out = out;
254  ctx->internal->execute(ctx, s->median_frames, &td, NULL, FFMIN(s->height[1], ff_filter_get_nb_threads(ctx)));
255 
256  return ff_filter_frame(outlink, out);
257 }
258 
259 static int config_output(AVFilterLink *outlink)
260 {
261  AVFilterContext *ctx = outlink->src;
262  XMedianContext *s = ctx->priv;
263  AVRational frame_rate = ctx->inputs[0]->frame_rate;
264  AVRational sar = ctx->inputs[0]->sample_aspect_ratio;
265  AVFilterLink *inlink = ctx->inputs[0];
266  int height = ctx->inputs[0]->h;
267  int width = ctx->inputs[0]->w;
268  FFFrameSyncIn *in;
269  int i, ret;
270 
271  for (int i = 1; i < s->nb_inputs && !s->tmedian; i++) {
272  if (ctx->inputs[i]->h != height || ctx->inputs[i]->w != width) {
273  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);
274  return AVERROR(EINVAL);
275  }
276  }
277 
278  s->desc = av_pix_fmt_desc_get(outlink->format);
279  if (!s->desc)
280  return AVERROR_BUG;
282  s->depth = s->desc->comp[0].depth;
283  s->max = (1 << s->depth) - 1;
284 
285  if (s->depth <= 8)
287  else
289 
290  if ((ret = av_image_fill_linesizes(s->linesize, inlink->format, inlink->w)) < 0)
291  return ret;
292 
293  s->width[1] = s->width[2] = AV_CEIL_RSHIFT(inlink->w, s->desc->log2_chroma_w);
294  s->width[0] = s->width[3] = inlink->w;
295  s->height[1] = s->height[2] = AV_CEIL_RSHIFT(inlink->h, s->desc->log2_chroma_h);
296  s->height[0] = s->height[3] = inlink->h;
297 
298  if (s->tmedian)
299  return 0;
300 
301  outlink->w = width;
302  outlink->h = height;
303  outlink->frame_rate = frame_rate;
304  outlink->sample_aspect_ratio = sar;
305 
306  if ((ret = ff_framesync_init(&s->fs, ctx, s->nb_inputs)) < 0)
307  return ret;
308 
309  in = s->fs.in;
310  s->fs.opaque = s;
312 
313  for (i = 0; i < s->nb_inputs; i++) {
314  AVFilterLink *inlink = ctx->inputs[i];
315 
316  in[i].time_base = inlink->time_base;
317  in[i].sync = 1;
318  in[i].before = EXT_STOP;
319  in[i].after = EXT_STOP;
320  }
321 
322  ret = ff_framesync_configure(&s->fs);
323  outlink->time_base = s->fs.time_base;
324 
325  return ret;
326 }
327 
329 {
330  XMedianContext *s = ctx->priv;
331 
332  ff_framesync_uninit(&s->fs);
333 
334  for (int i = 0; i < ctx->nb_inputs && !s->tmedian; i++)
335  av_freep(&ctx->input_pads[i].name);
336  for (int i = 0; i < s->nb_frames && s->frames && s->tmedian; i++)
337  av_frame_free(&s->frames[i]);
338  av_freep(&s->frames);
339 }
340 
342 {
343  XMedianContext *s = ctx->priv;
344  return ff_framesync_activate(&s->fs);
345 }
346 
347 #define OFFSET(x) offsetof(XMedianContext, x)
348 #define FLAGS AV_OPT_FLAG_VIDEO_PARAM | AV_OPT_FLAG_FILTERING_PARAM
349 
350 static const AVOption xmedian_options[] = {
351  { "inputs", "set number of inputs", OFFSET(nb_inputs), AV_OPT_TYPE_INT, {.i64=3}, 3, 255, .flags = FLAGS },
352  { "planes", "set planes to filter", OFFSET(planes), AV_OPT_TYPE_INT, {.i64=15}, 0, 15, .flags = FLAGS },
353  { "percentile", "set percentile", OFFSET(percentile),AV_OPT_TYPE_FLOAT,{.dbl=0.5}, 0, 1, .flags = FLAGS },
354  { NULL },
355 };
356 
357 static const AVFilterPad outputs[] = {
358  {
359  .name = "default",
360  .type = AVMEDIA_TYPE_VIDEO,
361  .config_props = config_output,
362  },
363  { NULL }
364 };
365 
366 #if CONFIG_XMEDIAN_FILTER
367 AVFILTER_DEFINE_CLASS(xmedian);
368 
370  .name = "xmedian",
371  .description = NULL_IF_CONFIG_SMALL("Pick median pixels from several video inputs."),
372  .priv_size = sizeof(XMedianContext),
373  .priv_class = &xmedian_class,
375  .outputs = outputs,
376  .init = init,
377  .uninit = uninit,
378  .activate = activate,
380 };
381 
382 #endif /* CONFIG_XMEDIAN_FILTER */
383 
384 #if CONFIG_TMEDIAN_FILTER
385 static int tmedian_filter_frame(AVFilterLink *inlink, AVFrame *in)
386 {
387  AVFilterContext *ctx = inlink->dst;
388  AVFilterLink *outlink = ctx->outputs[0];
389  XMedianContext *s = ctx->priv;
390  ThreadData td;
391  AVFrame *out;
392 
393  if (s->nb_frames < s->nb_inputs) {
394  s->frames[s->nb_frames] = in;
395  s->nb_frames++;
396  if (s->nb_frames < s->nb_inputs)
397  return 0;
398  } else {
399  av_frame_free(&s->frames[0]);
400  memmove(&s->frames[0], &s->frames[1], sizeof(*s->frames) * (s->nb_inputs - 1));
401  s->frames[s->nb_inputs - 1] = in;
402  }
403 
404  if (ctx->is_disabled) {
405  out = av_frame_clone(s->frames[0]);
406  if (!out)
407  return AVERROR(ENOMEM);
408  return ff_filter_frame(outlink, out);
409  }
410 
411  out = ff_get_video_buffer(outlink, outlink->w, outlink->h);
412  if (!out)
413  return AVERROR(ENOMEM);
414  out->pts = s->frames[0]->pts;
415 
416  td.out = out;
417  td.in = s->frames;
418  ctx->internal->execute(ctx, s->median_frames, &td, NULL, FFMIN(s->height[0], ff_filter_get_nb_threads(ctx)));
419 
420  return ff_filter_frame(outlink, out);
421 }
422 
423 static const AVOption tmedian_options[] = {
424  { "radius", "set median filter radius", OFFSET(radius), AV_OPT_TYPE_INT, {.i64=1}, 1, 127, .flags = FLAGS },
425  { "planes", "set planes to filter", OFFSET(planes), AV_OPT_TYPE_INT, {.i64=15}, 0, 15, .flags = FLAGS },
426  { "percentile", "set percentile", OFFSET(percentile), AV_OPT_TYPE_FLOAT, {.dbl=0.5}, 0, 1, .flags = FLAGS },
427  { NULL },
428 };
429 
430 static const AVFilterPad tmedian_inputs[] = {
431  {
432  .name = "default",
433  .type = AVMEDIA_TYPE_VIDEO,
434  .filter_frame = tmedian_filter_frame,
435  },
436  { NULL }
437 };
438 
439 static const AVFilterPad tmedian_outputs[] = {
440  {
441  .name = "default",
442  .type = AVMEDIA_TYPE_VIDEO,
443  .config_props = config_output,
444  },
445  { NULL }
446 };
447 
449 
451  .name = "tmedian",
452  .description = NULL_IF_CONFIG_SMALL("Pick median pixels from successive frames."),
453  .priv_size = sizeof(XMedianContext),
454  .priv_class = &tmedian_class,
456  .inputs = tmedian_inputs,
457  .outputs = tmedian_outputs,
458  .init = init,
459  .uninit = uninit,
461 };
462 
463 #endif /* CONFIG_TMEDIAN_FILTER */
#define NULL
Definition: coverity.c:32
#define AV_PIX_FMT_YUVA422P16
Definition: pixfmt.h:442
AVFrame * out
Definition: af_adeclick.c:494
#define AV_PIX_FMT_YUV440P10
Definition: pixfmt.h:401
#define AV_PIX_FMT_YUVA422P9
Definition: pixfmt.h:434
const AVPixFmtDescriptor * av_pix_fmt_desc_get(enum AVPixelFormat pix_fmt)
Definition: pixdesc.c:2573
This structure describes decoded (raw) audio or video data.
Definition: frame.h:314
#define OFFSET(x)
Definition: vf_xmedian.c:347
AVOption.
Definition: opt.h:248
ptrdiff_t const GLvoid * data
Definition: opengl_enc.c:100
#define AV_PIX_FMT_YUVA420P10
Definition: pixfmt.h:436
#define AV_PIX_FMT_YUV444P14
Definition: pixfmt.h:409
#define AV_PIX_FMT_GBRAP10
Definition: pixfmt.h:419
#define AV_PIX_FMT_YUVA422P10
Definition: pixfmt.h:437
planar YUV 4:4:4, 24bpp, (1 Cr & Cb sample per 1x1 Y samples)
Definition: pixfmt.h:71
misc image utilities
int av_pix_fmt_count_planes(enum AVPixelFormat pix_fmt)
Definition: pixdesc.c:2613
Main libavfilter public API header.
FFFrameSync fs
Definition: vf_xmedian.c:53
AVFilter ff_vf_tmedian
#define AVFILTER_FLAG_DYNAMIC_INPUTS
The number of the filter inputs is not determined just by AVFilter.inputs.
Definition: avfilter.h:106
planar GBR 4:4:4 24bpp
Definition: pixfmt.h:168
#define AV_PIX_FMT_GBRP10
Definition: pixfmt.h:415
#define AV_PIX_FMT_GRAY9
Definition: pixfmt.h:379
static int comparei(const void *p1, const void *p2)
Definition: vf_xmedian.c:137
enum AVMediaType type
AVFilterPad type.
Definition: internal.h:65
#define AV_PIX_FMT_YUV420P12
Definition: pixfmt.h:403
int ff_framesync_configure(FFFrameSync *fs)
Configure a frame sync structure.
Definition: framesync.c:124
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:259
int is_disabled
the enabled state from the last expression evaluation
Definition: avfilter.h:388
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:189
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:245
AVFilterFormats * ff_make_format_list(const int *fmts)
Create a list of supported formats.
Definition: formats.c:287
#define AV_PIX_FMT_GRAY10
Definition: pixfmt.h:380
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:381
AVFilterLink ** inputs
array of pointers to input links
Definition: avfilter.h:349
int height[4]
Definition: vf_xmedian.c:50
int ff_filter_frame(AVFilterLink *link, AVFrame *frame)
Send a frame of data to the next filter.
Definition: avfilter.c:1091
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:88
AVOptions.
static int median_frames16(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs)
Definition: vf_xmedian.c:144
int64_t pts
Presentation timestamp in time_base units (time when frame should be shown to user).
Definition: frame.h:407
FFFrameSyncIn * in
Pointer to array of inputs.
Definition: framesync.h:203
#define AV_PIX_FMT_YUVA420P9
Definition: pixfmt.h:433
#define AV_PIX_FMT_GBRP9
Definition: pixfmt.h:414
int width[4]
Definition: vf_xmedian.c:49
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:412
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:404
#define AV_PIX_FMT_YUVA420P16
Definition: pixfmt.h:441
#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:350
#define src
Definition: vp8dsp.c:254
AVFilterPad * input_pads
array of input pads
Definition: avfilter.h:348
planar YUV 4:2:2 24bpp, (1 Cr & Cb sample per 2x1 Y & A samples)
Definition: pixfmt.h:176
#define AV_LOG_ERROR
Something went wrong and cannot losslessly be recovered.
Definition: log.h:194
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:588
#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:290
Frame sync structure.
Definition: framesync.h:146
void av_frame_free(AVFrame **frame)
Free the frame and any dynamically allocated objects in it, e.g.
Definition: frame.c:203
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification. ...
Definition: internal.h:153
void * priv
private data for use by the filter
Definition: avfilter.h:356
#define AVFILTER_FLAG_SLICE_THREADS
The filter supports multithreading by splitting frames into multiple parts and processing them concur...
Definition: avfilter.h:117
#define AV_PIX_FMT_YUVA444P16
Definition: pixfmt.h:443
const char * arg
Definition: jacosubdec.c:66
#define AV_PIX_FMT_GBRAP12
Definition: pixfmt.h:420
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:341
int linesize[4]
Definition: vf_xmedian.c:48
#define AV_PIX_FMT_YUV444P10
Definition: pixfmt.h:402
int(* on_event)(struct FFFrameSync *fs)
Callback called when a frame event is ready.
Definition: framesync.h:172
AVFilter ff_vf_xmedian
#define AV_PIX_FMT_GBRAP16
Definition: pixfmt.h:421
static av_cold void uninit(AVFilterContext *ctx)
Definition: vf_xmedian.c:328
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:397
static int activate(AVFilterContext *ctx)
Definition: vf_xmedian.c:341
#define AV_PIX_FMT_GBRP16
Definition: pixfmt.h:418
int ff_filter_get_nb_threads(AVFilterContext *ctx)
Get number of threads for current filter instance.
Definition: avfilter.c:800
unsigned nb_inputs
number of input pads
Definition: avfilter.h:350
#define AV_PIX_FMT_GRAY16
Definition: pixfmt.h:383
#define FFMIN(a, b)
Definition: common.h:96
static int query_formats(AVFilterContext *ctx)
Definition: vf_xmedian.c:58
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:440
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:438
void * opaque
Opaque pointer, not used by the API.
Definition: framesync.h:177
#define AV_PIX_FMT_YUV444P9
Definition: pixfmt.h:398
static av_cold int init(AVFilterContext *ctx)
Definition: vf_xmedian.c:95
#define AV_PIX_FMT_GBRP14
Definition: pixfmt.h:417
AVFrame * av_frame_clone(const AVFrame *src)
Create a new frame that references the same data as src.
Definition: frame.c:540
#define AV_PIX_FMT_YUV420P16
Definition: pixfmt.h:410
#define AV_PIX_FMT_YUV420P14
Definition: pixfmt.h:407
Used for passing data between threads.
Definition: dsddec.c:67
int linesize[AV_NUM_DATA_POINTERS]
For video, size in bytes of each picture line.
Definition: frame.h:345
int ff_framesync_init(FFFrameSync *fs, AVFilterContext *parent, unsigned nb_in)
Initialize a frame sync structure.
Definition: framesync.c:84
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:382
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:55
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 inputs
#define AVERROR_BUG
Internal bug, also see AVERROR_BUG2.
Definition: error.h:50
AVFrame ** frames
Definition: vf_xmedian.c:52
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:399
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:145
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:149
#define AV_PIX_FMT_YUV440P12
Definition: pixfmt.h:405
#define AV_PIX_FMT_YUV420P9
Definition: pixfmt.h:396
#define AVFILTER_FLAG_SUPPORT_TIMELINE_INTERNAL
Same as AVFILTER_FLAG_SUPPORT_TIMELINE_GENERIC, except that the filter will have its filter_frame() c...
Definition: avfilter.h:134
AVFilterLink ** outputs
array of pointers to output links
Definition: avfilter.h:353
static int process_frame(FFFrameSync *fs)
Definition: vf_xmedian.c:232
static const AVFilterPad outputs[]
Definition: vf_xmedian.c:357
#define AV_PIX_FMT_YUV422P14
Definition: pixfmt.h:408
#define AV_PIX_FMT_GBRP12
Definition: pixfmt.h:416
#define flags(name, subs,...)
Definition: cbs_av1.c:560
AVFilterInternal * internal
An opaque struct for libavfilter internal use.
Definition: avfilter.h:381
#define AV_PIX_FMT_YUV422P10
Definition: pixfmt.h:400
#define AV_PIX_FMT_YUV444P12
Definition: pixfmt.h:406
uint8_t * data[AV_NUM_DATA_POINTERS]
pointer to the picture/channel planes.
Definition: frame.h:328
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:40
#define AV_PIX_FMT_YUVA444P9
Definition: pixfmt.h:435
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:136
static int slice_end(AVCodecContext *avctx, AVFrame *pict)
Handle slice ends.
Definition: mpeg12dec.c:2035
Completely stop all streams with this one.
Definition: framesync.h:65
#define AVFILTER_DEFINE_CLASS(fname)
Definition: internal.h:288
A list of supported formats for one end of a filter link.
Definition: formats.h:65
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:341
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_adenorm.c:223
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:373
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:253
int depth
Number of bits in the component.
Definition: pixdesc.h:58
#define FLAGS
Definition: vf_xmedian.c:348
AVPixelFormat
Pixel format.
Definition: pixfmt.h:64
const AVPixFmtDescriptor * desc
Definition: vf_xmedian.c:36
const AVFilter * filter
the AVFilter of which this is an instance
Definition: avfilter.h:344
#define AV_PIX_FMT_YUV422P16
Definition: pixfmt.h:411
#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
int i
Definition: input.c:407
#define AV_PIX_FMT_YUVA422P12
Definition: pixfmt.h:439
#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:240