FFmpeg
vf_lumakey.c
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2017 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/opt.h"
22 #include "libavutil/imgutils.h"
23 #include "avfilter.h"
24 #include "formats.h"
25 #include "internal.h"
26 #include "video.h"
27 
28 typedef struct LumakeyContext {
29  const AVClass *class;
30 
31  double threshold;
32  double tolerance;
33  double softness;
34 
35  int white;
36  int black;
37  int so;
38  int max;
39 
40  int (*do_lumakey_slice)(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs);
42 
43 static int do_lumakey_slice8(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs)
44 {
45  LumakeyContext *s = ctx->priv;
46  AVFrame *frame = arg;
47  const int slice_start = (frame->height * jobnr) / nb_jobs;
48  const int slice_end = (frame->height * (jobnr + 1)) / nb_jobs;
49  uint8_t *alpha = frame->data[3] + slice_start * frame->linesize[3];
50  const uint8_t *luma = frame->data[0] + slice_start * frame->linesize[0];
51  const int so = s->so;
52  const int w = s->white;
53  const int b = s->black;
54  int x, y;
55 
56  for (y = slice_start; y < slice_end; y++) {
57  for (x = 0; x < frame->width; x++) {
58  if (luma[x] >= b && luma[x] <= w) {
59  alpha[x] = 0;
60  } else if (luma[x] > b - so && luma[x] < w + so) {
61  if (luma[x] < b) {
62  alpha[x] = 255 - (luma[x] - b + so) * 255 / so;
63  } else {
64  alpha[x] = (luma[x] - w) * 255 / so;
65  }
66  }
67  }
68  luma += frame->linesize[0];
69  alpha += frame->linesize[3];
70  }
71 
72  return 0;
73 }
74 
75 static int do_lumakey_slice16(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs)
76 {
77  LumakeyContext *s = ctx->priv;
78  AVFrame *frame = arg;
79  const int slice_start = (frame->height * jobnr) / nb_jobs;
80  const int slice_end = (frame->height * (jobnr + 1)) / nb_jobs;
81  uint16_t *alpha = (uint16_t *)(frame->data[3] + slice_start * frame->linesize[3]);
82  const uint16_t *luma = (const uint16_t *)(frame->data[0] + slice_start * frame->linesize[0]);
83  const int so = s->so;
84  const int w = s->white;
85  const int b = s->black;
86  const int m = s->max;
87  int x, y;
88 
89  for (y = slice_start; y < slice_end; y++) {
90  for (x = 0; x < frame->width; x++) {
91  if (luma[x] >= b && luma[x] <= w) {
92  alpha[x] = 0;
93  } else if (luma[x] > b - so && luma[x] < w + so) {
94  if (luma[x] < b) {
95  alpha[x] = m - (luma[x] - b + so) * m / so;
96  } else {
97  alpha[x] = (luma[x] - w) * m / so;
98  }
99  }
100  }
101  luma += frame->linesize[0] / 2;
102  alpha += frame->linesize[3] / 2;
103  }
104 
105  return 0;
106 }
107 
109 {
111  AVFilterContext *ctx = inlink->dst;
112  LumakeyContext *s = ctx->priv;
113  int depth;
114 
115  depth = desc->comp[0].depth;
116  if (depth == 8) {
117  s->white = av_clip_uint8((s->threshold + s->tolerance) * 255);
118  s->black = av_clip_uint8((s->threshold - s->tolerance) * 255);
120  s->so = s->softness * 255;
121  } else {
122  s->max = (1 << depth) - 1;
123  s->white = av_clip((s->threshold + s->tolerance) * s->max, 0, s->max);
124  s->black = av_clip((s->threshold - s->tolerance) * s->max, 0, s->max);
126  s->so = s->softness * s->max;
127  }
128 
129  return 0;
130 }
131 
133 {
134  AVFilterContext *ctx = link->dst;
135  LumakeyContext *s = ctx->priv;
136  int ret;
137 
138  if (ret = av_frame_make_writable(frame))
139  return ret;
140 
141  if (ret = ctx->internal->execute(ctx, s->do_lumakey_slice, frame, NULL, FFMIN(frame->height, ff_filter_get_nb_threads(ctx))))
142  return ret;
143 
144  return ff_filter_frame(ctx->outputs[0], frame);
145 }
146 
148 {
149  static const enum AVPixelFormat pixel_fmts[] = {
156  };
158 
159  formats = ff_make_format_list(pixel_fmts);
160  if (!formats)
161  return AVERROR(ENOMEM);
162 
163  return ff_set_common_formats(ctx, formats);
164 }
165 
166 static int process_command(AVFilterContext *ctx, const char *cmd, const char *args,
167  char *res, int res_len, int flags)
168 {
169  int ret;
170 
171  ret = ff_filter_process_command(ctx, cmd, args, res, res_len, flags);
172  if (ret < 0)
173  return ret;
174 
175  return config_input(ctx->inputs[0]);
176 }
177 
178 static const AVFilterPad lumakey_inputs[] = {
179  {
180  .name = "default",
181  .type = AVMEDIA_TYPE_VIDEO,
182  .filter_frame = filter_frame,
183  .config_props = config_input,
184  },
185  { NULL }
186 };
187 
188 static const AVFilterPad lumakey_outputs[] = {
189  {
190  .name = "default",
191  .type = AVMEDIA_TYPE_VIDEO,
192  },
193  { NULL }
194 };
195 
196 #define OFFSET(x) offsetof(LumakeyContext, x)
197 #define FLAGS AV_OPT_FLAG_FILTERING_PARAM|AV_OPT_FLAG_VIDEO_PARAM|AV_OPT_FLAG_RUNTIME_PARAM
198 
199 static const AVOption lumakey_options[] = {
200  { "threshold", "set the threshold value", OFFSET(threshold), AV_OPT_TYPE_DOUBLE, {.dbl=0}, 0, 1, FLAGS },
201  { "tolerance", "set the tolerance value", OFFSET(tolerance), AV_OPT_TYPE_DOUBLE, {.dbl=0.01}, 0, 1, FLAGS },
202  { "softness", "set the softness value", OFFSET(softness), AV_OPT_TYPE_DOUBLE, {.dbl=0}, 0, 1, FLAGS },
203  { NULL }
204 };
205 
206 AVFILTER_DEFINE_CLASS(lumakey);
207 
209  .name = "lumakey",
210  .description = NULL_IF_CONFIG_SMALL("Turns a certain luma into transparency."),
211  .priv_size = sizeof(LumakeyContext),
212  .priv_class = &lumakey_class,
214  .inputs = lumakey_inputs,
215  .outputs = lumakey_outputs,
218 };
#define NULL
Definition: coverity.c:32
static int do_lumakey_slice8(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs)
Definition: vf_lumakey.c:43
#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
static const AVFilterPad lumakey_outputs[]
Definition: vf_lumakey.c:188
static int do_lumakey_slice16(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs)
Definition: vf_lumakey.c:75
#define AV_PIX_FMT_YUVA422P10
Definition: pixfmt.h:425
misc image utilities
Main libavfilter public API header.
const char * desc
Definition: nvenc.c:68
int(* do_lumakey_slice)(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs)
Definition: vf_lumakey.c:40
static const AVOption lumakey_options[]
Definition: vf_lumakey.c:199
AVFilterFormats * ff_make_format_list(const int *fmts)
Create a list of supported formats.
Definition: formats.c:283
double threshold
Definition: vf_lumakey.c:31
#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
AVFilterLink ** inputs
array of pointers to input links
Definition: avfilter.h:346
int ff_filter_frame(AVFilterLink *link, AVFrame *frame)
Send a frame of data to the next filter.
Definition: avfilter.c:1093
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.
#define AV_PIX_FMT_YUVA420P9
Definition: pixfmt.h:421
#define AV_PIX_FMT_YUVA420P16
Definition: pixfmt.h:429
AVFILTER_DEFINE_CLASS(lumakey)
A filter pad used for either input or output.
Definition: internal.h:54
planar YUV 4:2:2 24bpp, (1 Cr & Cb sample per 2x1 Y & A samples)
Definition: pixfmt.h:176
int width
Definition: frame.h:353
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 FLAGS
Definition: vf_lumakey.c:197
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification. ...
Definition: internal.h:186
int ff_filter_process_command(AVFilterContext *ctx, const char *cmd, const char *arg, char *res, int res_len, int flags)
Generic processing of user supplied commands that are set in the same way as the filter options...
Definition: avfilter.c:887
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
AVFilter ff_vf_lumakey
Definition: vf_lumakey.c:208
#define AV_PIX_FMT_YUVA444P16
Definition: pixfmt.h:431
const char * arg
Definition: jacosubdec.c:66
#define b
Definition: input.c:41
int ff_filter_get_nb_threads(AVFilterContext *ctx)
Get number of threads for current filter instance.
Definition: avfilter.c:802
#define FFMIN(a, b)
Definition: common.h:96
#define AV_PIX_FMT_YUVA444P12
Definition: pixfmt.h:428
uint8_t w
Definition: llviddspenc.c:38
AVFormatContext * ctx
Definition: movenc.c:48
double softness
Definition: vf_lumakey.c:33
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
static const AVFilterPad outputs[]
Definition: af_acontrast.c:203
static int process_command(AVFilterContext *ctx, const char *cmd, const char *args, char *res, int res_len, int flags)
Definition: vf_lumakey.c:166
static const AVFilterPad lumakey_inputs[]
Definition: vf_lumakey.c:178
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
static const int16_t alpha[]
Definition: ilbcdata.h:55
#define OFFSET(x)
Definition: vf_lumakey.c:196
static av_cold int query_formats(AVFilterContext *ctx)
Definition: vf_lumakey.c:147
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
Describe the class of an AVClass context structure.
Definition: log.h:67
Filter definition.
Definition: avfilter.h:144
const char * name
Filter name.
Definition: avfilter.h:148
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 link
AVFilterLink ** outputs
array of pointers to output links
Definition: avfilter.h:350
int av_frame_make_writable(AVFrame *frame)
Ensure that the frame data is writable, avoiding data copy if possible.
Definition: frame.c:611
#define flags(name, subs,...)
Definition: cbs_av1.c:561
AVFilterInternal * internal
An opaque struct for libavfilter internal use.
Definition: avfilter.h:378
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
#define AV_PIX_FMT_YUVA444P9
Definition: pixfmt.h:423
avfilter_execute_func * execute
Definition: internal.h:155
static int slice_end(AVCodecContext *avctx, AVFrame *pict)
Handle slice ends.
Definition: mpeg12dec.c:2036
static int filter_frame(AVFilterLink *link, AVFrame *frame)
Definition: vf_lumakey.c:132
A list of supported formats for one end of a filter link.
Definition: formats.h:64
An instance of a filter.
Definition: avfilter.h:338
int height
Definition: frame.h:353
static int config_input(AVFilterLink *inlink)
Definition: vf_lumakey.c:108
formats
Definition: signature.h:48
internal API functions
double tolerance
Definition: vf_lumakey.c:32
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
#define AV_PIX_FMT_YUVA422P12
Definition: pixfmt.h:427