FFmpeg
vf_maskfun.c
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2018 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/imgutils.h"
22 #include "libavutil/pixdesc.h"
23 #include "libavutil/opt.h"
24 #include "avfilter.h"
25 #include "formats.h"
26 #include "internal.h"
27 #include "video.h"
28 
29 typedef struct MaskFunContext {
30  const AVClass *class;
31 
32  int low, high;
33  int planes;
34  int fill;
35  int sum;
36 
37  int linesize[4];
38  int width[4], height[4];
39  int nb_planes;
40  int depth;
41  int max;
42  uint64_t max_sum;
43 
46  int (*maskfun)(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs);
48 
49 #define OFFSET(x) offsetof(MaskFunContext, x)
50 #define VF AV_OPT_FLAG_FILTERING_PARAM|AV_OPT_FLAG_VIDEO_PARAM
51 
52 static const AVOption maskfun_options[] = {
53  { "low", "set low threshold", OFFSET(low), AV_OPT_TYPE_INT, {.i64=10}, 0, UINT16_MAX, VF },
54  { "high", "set high threshold", OFFSET(high), AV_OPT_TYPE_INT, {.i64=10}, 0, UINT16_MAX, VF },
55  { "planes", "set planes", OFFSET(planes), AV_OPT_TYPE_INT, {.i64=0xF}, 0, 0xF, VF },
56  { "fill", "set fill value", OFFSET(fill), AV_OPT_TYPE_INT, {.i64=0}, 0, UINT16_MAX, VF },
57  { "sum", "set sum value", OFFSET(sum), AV_OPT_TYPE_INT, {.i64=10}, 0, UINT16_MAX, VF },
58  { NULL }
59 };
60 
62 
64 {
65  static const enum AVPixelFormat pix_fmts[] = {
84  };
85 
86  return ff_set_common_formats(ctx, ff_make_format_list(pix_fmts));
87 }
88 
90 {
91  AVFilterContext *ctx = inlink->dst;
92  MaskFunContext *s = ctx->priv;
93  AVFilterLink *outlink = ctx->outputs[0];
94 
95  if (s->getsum(ctx, frame)) {
97 
98  if (!out) {
99  av_frame_free(&frame);
100  return AVERROR(ENOMEM);
101  }
102  out->pts = frame->pts;
103  av_frame_free(&frame);
104 
105  return ff_filter_frame(outlink, out);
106  }
107 
108  ctx->internal->execute(ctx, s->maskfun, frame, NULL,
109  FFMIN(s->height[1], ff_filter_get_nb_threads(ctx)));
110 
111  return ff_filter_frame(outlink, frame);
112 }
113 
114 #define GETSUM(name, type, div) \
115 static int getsum##name(AVFilterContext *ctx, AVFrame *out) \
116 { \
117  MaskFunContext *s = ctx->priv; \
118  uint64_t sum = 0; \
119  int p; \
120  \
121  for (p = 0; p < s->nb_planes; p++) { \
122  const int linesize = out->linesize[p] / div; \
123  const int w = s->width[p]; \
124  const int h = s->height[p]; \
125  type *dst = (type *)out->data[p]; \
126  \
127  if (!((1 << p) & s->planes)) \
128  continue; \
129  \
130  for (int y = 0; y < h; y++) { \
131  for (int x = 0; x < w; x++) \
132  sum += dst[x]; \
133  if (sum >= s->max_sum) \
134  return 1; \
135  dst += linesize; \
136  } \
137  } \
138  \
139  return 0; \
140 }
141 
142 GETSUM(8, uint8_t, 1)
143 GETSUM(16, uint16_t, 2)
144 
145 #define MASKFUN(name, type, div) \
146 static int maskfun##name(AVFilterContext *ctx, void *arg, \
147  int jobnr, int nb_jobs) \
148 { \
149  MaskFunContext *s = ctx->priv; \
150  AVFrame *out = arg; \
151  const int low = s->low; \
152  const int high = s->high; \
153  const int max = s->max; \
154  int p; \
155  \
156  for (p = 0; p < s->nb_planes; p++) { \
157  const int linesize = out->linesize[p] / div; \
158  const int w = s->width[p]; \
159  const int h = s->height[p]; \
160  const int slice_start = (h * jobnr) / nb_jobs; \
161  const int slice_end = (h * (jobnr+1)) / nb_jobs; \
162  type *dst = (type *)out->data[p] + slice_start * linesize; \
163  \
164  if (!((1 << p) & s->planes)) \
165  continue; \
166  \
167  for (int y = slice_start; y < slice_end; y++) { \
168  for (int x = 0; x < w; x++) { \
169  if (dst[x] <= low) \
170  dst[x] = 0; \
171  else if (dst[x] > high) \
172  dst[x] = max; \
173  } \
174  \
175  dst += linesize; \
176  } \
177  } \
178  \
179  return 0; \
180 }
181 
182 MASKFUN(8, uint8_t, 1)
183 MASKFUN(16, uint16_t, 2)
184 
186 {
187  AVFilterContext *ctx = inlink->dst;
188  MaskFunContext *s = ctx->priv;
189  const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(inlink->format);
190  int vsub, hsub, ret;
191 
192  s->nb_planes = av_pix_fmt_count_planes(inlink->format);
193 
194  if ((ret = av_image_fill_linesizes(s->linesize, inlink->format, inlink->w)) < 0)
195  return ret;
196 
197  hsub = desc->log2_chroma_w;
198  vsub = desc->log2_chroma_h;
199  s->height[1] = s->height[2] = AV_CEIL_RSHIFT(inlink->h, vsub);
200  s->height[0] = s->height[3] = inlink->h;
201  s->width[1] = s->width[2] = AV_CEIL_RSHIFT(inlink->w, hsub);
202  s->width[0] = s->width[3] = inlink->w;
203 
204  s->depth = desc->comp[0].depth;
205  s->max = (1 << s->depth) - 1;
206  s->fill = FFMIN(s->fill, s->max);
207 
208  if (s->depth == 8) {
209  s->maskfun = maskfun8;
210  s->getsum = getsum8;
211  } else {
212  s->maskfun = maskfun16;
213  s->getsum = getsum16;
214  }
215 
216  s->empty = ff_get_video_buffer(inlink, inlink->w, inlink->h);
217  if (!s->empty)
218  return AVERROR(ENOMEM);
219 
220  if (s->depth == 8) {
221  for (int p = 0; p < s->nb_planes; p++) {
222  uint8_t *dst = s->empty->data[p];
223 
224  for (int y = 0; y < s->height[p]; y++) {
225  memset(dst, s->fill, s->width[p]);
226  dst += s->empty->linesize[p];
227  }
228  }
229  } else {
230  for (int p = 0; p < s->nb_planes; p++) {
231  uint16_t *dst = (uint16_t *)s->empty->data[p];
232 
233  for (int y = 0; y < s->height[p]; y++) {
234  for (int x = 0; x < s->width[p]; x++)
235  dst[x] = s->fill;
236  dst += s->empty->linesize[p] / 2;
237  }
238  }
239  }
240 
241  s->max_sum = 0;
242  for (int p = 0; p < s->nb_planes; p++) {
243  if (!((1 << p) & s->planes))
244  continue;
245  s->max_sum += (uint64_t)s->sum * s->width[p] * s->height[p];
246  }
247 
248  return 0;
249 }
250 
251 static const AVFilterPad maskfun_inputs[] = {
252  {
253  .name = "default",
254  .type = AVMEDIA_TYPE_VIDEO,
255  .filter_frame = filter_frame,
256  .config_props = config_input,
257  .needs_writable = 1,
258  },
259  { NULL }
260 };
261 
262 static const AVFilterPad maskfun_outputs[] = {
263  {
264  .name = "default",
265  .type = AVMEDIA_TYPE_VIDEO,
266  },
267  { NULL }
268 };
269 
271  .name = "maskfun",
272  .description = NULL_IF_CONFIG_SMALL("Create Mask."),
273  .priv_size = sizeof(MaskFunContext),
275  .inputs = maskfun_inputs,
276  .outputs = maskfun_outputs,
277  .priv_class = &maskfun_class,
279 };
#define NULL
Definition: coverity.c:32
#define AV_PIX_FMT_YUVA422P16
Definition: pixfmt.h:430
#define AV_PIX_FMT_YUVA422P9
Definition: pixfmt.h:422
const AVPixFmtDescriptor * av_pix_fmt_desc_get(enum AVPixelFormat pix_fmt)
Definition: pixdesc.c:2522
This structure describes decoded (raw) audio or video data.
Definition: frame.h:295
AVOption.
Definition: opt.h:246
#define AV_PIX_FMT_YUVA420P10
Definition: pixfmt.h:424
#define AV_PIX_FMT_YUV444P14
Definition: pixfmt.h:397
#define AV_PIX_FMT_GBRAP10
Definition: pixfmt.h:407
#define AV_PIX_FMT_YUVA422P10
Definition: pixfmt.h:425
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:2562
Main libavfilter public API header.
const char * desc
Definition: nvenc.c:68
int(* maskfun)(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs)
Definition: vf_maskfun.c:46
planar GBR 4:4:4 24bpp
Definition: pixfmt.h:168
#define AV_PIX_FMT_GBRP10
Definition: pixfmt.h:403
#define AV_PIX_FMT_GRAY9
Definition: pixfmt.h:367
#define AV_PIX_FMT_YUV420P12
Definition: pixfmt.h:391
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
uint8_t log2_chroma_w
Amount to shift the luma width right to find the chroma width.
Definition: pixdesc.h:92
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:368
AVFrame * empty
Definition: vf_maskfun.c:44
#define AVFILTER_FLAG_SUPPORT_TIMELINE_GENERIC
Some filters support a generic "enable" expression option that can be used to enable or disable a fil...
Definition: avfilter.h:125
const char * name
Pad name.
Definition: internal.h:60
#define AV_PIX_FMT_GRAY12
Definition: pixfmt.h:369
int ff_filter_frame(AVFilterLink *link, AVFrame *frame)
Send a frame of data to the next filter.
Definition: avfilter.c:1080
static const AVFilterPad maskfun_inputs[]
Definition: vf_maskfun.c:251
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
int width[4]
Definition: vf_maskfun.c:38
AVOptions.
int64_t pts
Presentation timestamp in time_base units (time when frame should be shown to user).
Definition: frame.h:388
#define AV_PIX_FMT_YUVA420P9
Definition: pixfmt.h:421
#define AV_PIX_FMT_GBRP9
Definition: pixfmt.h:402
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:400
#define AV_PIX_FMT_YUV422P12
Definition: pixfmt.h:392
#define AV_PIX_FMT_YUVA420P16
Definition: pixfmt.h:429
A filter pad used for either input or output.
Definition: internal.h:54
uint64_t max_sum
Definition: vf_maskfun.c:42
planar YUV 4:2:2 24bpp, (1 Cr & Cb sample per 2x1 Y & A samples)
Definition: pixfmt.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:568
uint8_t log2_chroma_h
Amount to shift the luma height right to find the chroma height.
Definition: pixdesc.h:101
int height[4]
Definition: vf_maskfun.c:38
void av_frame_free(AVFrame **frame)
Free the frame and any dynamically allocated objects in it, e.g.
Definition: frame.c:202
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification. ...
Definition: internal.h:186
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:431
const char * arg
Definition: jacosubdec.c:66
#define AV_PIX_FMT_GBRAP12
Definition: pixfmt.h:408
#define AV_PIX_FMT_YUV444P10
Definition: pixfmt.h:390
#define AV_PIX_FMT_GBRAP16
Definition: pixfmt.h:409
planar YUV 4:2:2, 16bpp, (1 Cr & Cb sample per 2x1 Y samples)
Definition: pixfmt.h:70
static int query_formats(AVFilterContext *ctx)
Definition: vf_maskfun.c:63
#define AV_PIX_FMT_YUV422P9
Definition: pixfmt.h:385
int linesize[4]
Definition: vf_maskfun.c:37
#define AV_PIX_FMT_GBRP16
Definition: pixfmt.h:406
int ff_filter_get_nb_threads(AVFilterContext *ctx)
Get number of threads for current filter instance.
Definition: avfilter.c:802
#define AV_PIX_FMT_GRAY16
Definition: pixfmt.h:371
#define FFMIN(a, b)
Definition: common.h:96
planar YUV 4:2:0, 12bpp, full scale (JPEG), deprecated in favor of AV_PIX_FMT_YUV420P and setting col...
Definition: pixfmt.h:78
AVFormatContext * ctx
Definition: movenc.c:48
AVFilter ff_vf_maskfun
Definition: vf_maskfun.c:270
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 or at least make progress towards producing a frame
#define s(width, name)
Definition: cbs_vp9.c:257
#define AV_PIX_FMT_YUVA444P10
Definition: pixfmt.h:426
#define AV_PIX_FMT_YUV444P9
Definition: pixfmt.h:386
#define AV_PIX_FMT_GBRP14
Definition: pixfmt.h:405
AVFrame * av_frame_clone(const AVFrame *src)
Create a new frame that references the same data as src.
Definition: frame.c:540
static const AVFilterPad outputs[]
Definition: af_acontrast.c:203
#define AV_PIX_FMT_YUV420P16
Definition: pixfmt.h:398
#define AV_PIX_FMT_YUV420P14
Definition: pixfmt.h:395
int linesize[AV_NUM_DATA_POINTERS]
For video, size in bytes of each picture line.
Definition: frame.h:326
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:370
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 VF
Definition: vf_maskfun.c:50
#define AV_PIX_FMT_YUV420P10
Definition: pixfmt.h:387
planar YUV 4:1:0, 9bpp, (1 Cr & Cb sample per 4x4 Y samples)
Definition: pixfmt.h:72
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
#define GETSUM(name, type, div)
Definition: vf_maskfun.c:114
#define MASKFUN(name, type, div)
Definition: vf_maskfun.c:145
AVFILTER_DEFINE_CLASS(maskfun)
const char * name
Filter name.
Definition: avfilter.h:148
#define AV_PIX_FMT_YUV440P12
Definition: pixfmt.h:393
#define OFFSET(x)
Definition: vf_maskfun.c:49
#define AV_PIX_FMT_YUV420P9
Definition: pixfmt.h:384
AVFilterLink ** outputs
array of pointers to output links
Definition: avfilter.h:350
static enum AVPixelFormat pix_fmts[]
Definition: libkvazaar.c:275
#define AV_PIX_FMT_YUV422P14
Definition: pixfmt.h:396
#define AV_PIX_FMT_GBRP12
Definition: pixfmt.h:404
#define flags(name, subs,...)
Definition: cbs_av1.c:561
AVFilterInternal * internal
An opaque struct for libavfilter internal use.
Definition: avfilter.h:378
#define AV_PIX_FMT_YUV422P10
Definition: pixfmt.h:388
#define AV_PIX_FMT_YUV444P12
Definition: pixfmt.h:394
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
static int config_input(AVFilterLink *inlink)
Definition: vf_maskfun.c:185
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
#define AV_PIX_FMT_YUVA444P9
Definition: pixfmt.h:423
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:155
int(* getsum)(AVFilterContext *ctx, AVFrame *out)
Definition: vf_maskfun.c:45
static const AVOption maskfun_options[]
Definition: vf_maskfun.c:52
static int filter_frame(AVFilterLink *inlink, AVFrame *frame)
Definition: vf_maskfun.c:89
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
planar YUV 4:4:0 (1 Cr & Cb sample per 1x2 Y samples)
Definition: pixfmt.h:99
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 depth
Number of bits in the component.
Definition: pixdesc.h:58
AVPixelFormat
Pixel format.
Definition: pixfmt.h:64
static const AVFilterPad maskfun_outputs[]
Definition: vf_maskfun.c:262
#define AV_PIX_FMT_YUV422P16
Definition: pixfmt.h:399
for(j=16;j >0;--j)
#define AV_CEIL_RSHIFT(a, b)
Definition: common.h:58