FFmpeg
vf_bbox.c
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2012 Stefano Sabatini
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 /**
22  * @file
23  * bounding box detection filter
24  */
25 
26 #include "libavutil/opt.h"
27 #include "libavutil/pixdesc.h"
28 #include "libavutil/timestamp.h"
29 #include "avfilter.h"
30 #include "bbox.h"
31 #include "internal.h"
32 
33 typedef struct BBoxContext {
34  const AVClass *class;
35  int min_val;
36  int depth;
37 } BBoxContext;
38 
39 #define OFFSET(x) offsetof(BBoxContext, x)
40 #define FLAGS AV_OPT_FLAG_VIDEO_PARAM|AV_OPT_FLAG_FILTERING_PARAM|AV_OPT_FLAG_RUNTIME_PARAM
41 
42 static const AVOption bbox_options[] = {
43  { "min_val", "set minimum luminance value for bounding box", OFFSET(min_val), AV_OPT_TYPE_INT, { .i64 = 16 }, 0, UINT16_MAX, FLAGS },
44  { NULL }
45 };
46 
48 
49 static const enum AVPixelFormat pix_fmts[] = {
71 };
72 
73 #define SET_META(key, value) \
74  av_dict_set_int(metadata, key, value, 0);
75 
77 {
78  AVFilterContext *ctx = inlink->dst;
79  BBoxContext *bbox = ctx->priv;
80  FFBoundingBox box;
81  int has_bbox, w, h;
82 
83  has_bbox =
85  frame->data[0], frame->linesize[0],
86  inlink->w, inlink->h, bbox->min_val, bbox->depth);
87  w = box.x2 - box.x1 + 1;
88  h = box.y2 - box.y1 + 1;
89 
91  "n:%"PRId64" pts:%s pts_time:%s", inlink->frame_count_out,
92  av_ts2str(frame->pts), av_ts2timestr(frame->pts, &inlink->time_base));
93 
94  if (has_bbox) {
95  AVDictionary **metadata = &frame->metadata;
96 
97  SET_META("lavfi.bbox.x1", box.x1)
98  SET_META("lavfi.bbox.x2", box.x2)
99  SET_META("lavfi.bbox.y1", box.y1)
100  SET_META("lavfi.bbox.y2", box.y2)
101  SET_META("lavfi.bbox.w", w)
102  SET_META("lavfi.bbox.h", h)
103 
105  " x1:%d x2:%d y1:%d y2:%d w:%d h:%d"
106  " crop=%d:%d:%d:%d drawbox=%d:%d:%d:%d",
107  box.x1, box.x2, box.y1, box.y2, w, h,
108  w, h, box.x1, box.y1, /* crop params */
109  box.x1, box.y1, w, h); /* drawbox params */
110  }
111  av_log(ctx, AV_LOG_INFO, "\n");
112 
113  return ff_filter_frame(inlink->dst->outputs[0], frame);
114 }
115 
116 static int config_output(AVFilterLink *outlink)
117 {
118  AVFilterContext *ctx = outlink->src;
119  BBoxContext *s = ctx->priv;
120  const AVPixFmtDescriptor *desc;
121 
122  desc = av_pix_fmt_desc_get(outlink->format);
123  if (!desc)
124  return AVERROR_BUG;
125  s->depth = desc->comp[0].depth;
126 
127  return 0;
128 }
129 
130 static const AVFilterPad bbox_inputs[] = {
131  {
132  .name = "default",
133  .type = AVMEDIA_TYPE_VIDEO,
134  .filter_frame = filter_frame,
135  },
136 };
137 
138 static const AVFilterPad bbox_outputs[] = {
139  {
140  .name = "default",
141  .type = AVMEDIA_TYPE_VIDEO,
142  .config_props = config_output,
143  },
144 };
145 
147  .name = "bbox",
148  .description = NULL_IF_CONFIG_SMALL("Compute bounding box for each frame."),
149  .priv_size = sizeof(BBoxContext),
150  .priv_class = &bbox_class,
155  .process_command = ff_filter_process_command,
156 };
AV_PIX_FMT_YUVA422P16
#define AV_PIX_FMT_YUVA422P16
Definition: pixfmt.h:526
AVPixelFormat
AVPixelFormat
Pixel format.
Definition: pixfmt.h:71
opt.h
bbox.h
filter_frame
static int filter_frame(AVFilterLink *inlink, AVFrame *frame)
Definition: vf_bbox.c:76
ff_filter_frame
int ff_filter_frame(AVFilterLink *link, AVFrame *frame)
Send a frame of data to the next filter.
Definition: avfilter.c:1007
av_pix_fmt_desc_get
const AVPixFmtDescriptor * av_pix_fmt_desc_get(enum AVPixelFormat pix_fmt)
Definition: pixdesc.c:2968
FILTER_PIXFMTS_ARRAY
#define FILTER_PIXFMTS_ARRAY(array)
Definition: internal.h:162
inlink
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
Definition: filter_design.txt:212
AV_PIX_FMT_YUVA422P9
#define AV_PIX_FMT_YUVA422P9
Definition: pixfmt.h:518
AVFrame
This structure describes decoded (raw) audio or video data.
Definition: frame.h:340
pixdesc.h
AVFrame::pts
int64_t pts
Presentation timestamp in time_base units (time when frame should be shown to user).
Definition: frame.h:452
ff_calculate_bounding_box
int ff_calculate_bounding_box(FFBoundingBox *bbox, const uint8_t *data, int linesize, int w, int h, int min_val, int depth)
Calculate the smallest rectangle that will encompass the region with values > min_val.
Definition: bbox.c:81
AV_PIX_FMT_YUVA420P16
#define AV_PIX_FMT_YUVA420P16
Definition: pixfmt.h:525
w
uint8_t w
Definition: llviddspenc.c:38
AV_PIX_FMT_YUVA420P10
#define AV_PIX_FMT_YUVA420P10
Definition: pixfmt.h:520
AVOption
AVOption.
Definition: opt.h:294
AV_PIX_FMT_YUV420P10
#define AV_PIX_FMT_YUV420P10
Definition: pixfmt.h:482
AV_PIX_FMT_YUV440P
@ AV_PIX_FMT_YUV440P
planar YUV 4:4:0 (1 Cr & Cb sample per 1x2 Y samples)
Definition: pixfmt.h:106
BBoxContext
Definition: vf_bbox.c:33
AVDictionary
Definition: dict.c:34
AVFilter::name
const char * name
Filter name.
Definition: avfilter.h:170
AV_PIX_FMT_YUVA422P10
#define AV_PIX_FMT_YUVA422P10
Definition: pixfmt.h:521
AV_PIX_FMT_GRAY9
#define AV_PIX_FMT_GRAY9
Definition: pixfmt.h:462
AVFrame::data
uint8_t * data[AV_NUM_DATA_POINTERS]
pointer to the picture/channel planes.
Definition: frame.h:361
AV_PIX_FMT_YUVA420P9
#define AV_PIX_FMT_YUVA420P9
Definition: pixfmt.h:517
FLAGS
#define FLAGS
Definition: vf_bbox.c:40
AV_PIX_FMT_YUVA444P16
#define AV_PIX_FMT_YUVA444P16
Definition: pixfmt.h:527
AV_PIX_FMT_YUV422P9
#define AV_PIX_FMT_YUV422P9
Definition: pixfmt.h:480
FFBoundingBox::x2
int x2
Definition: bbox.h:27
SET_META
#define SET_META(key, value)
Definition: vf_bbox.c:73
AV_PIX_FMT_GRAY16
#define AV_PIX_FMT_GRAY16
Definition: pixfmt.h:466
config_output
static int config_output(AVFilterLink *outlink)
Definition: vf_bbox.c:116
AVFilterPad
A filter pad used for either input or output.
Definition: internal.h:33
AV_PIX_FMT_YUV444P10
#define AV_PIX_FMT_YUV444P10
Definition: pixfmt.h:485
AV_PIX_FMT_YUVJ411P
@ AV_PIX_FMT_YUVJ411P
planar YUV 4:1:1, 12bpp, (1 Cr & Cb sample per 4x1 Y samples) full scale (JPEG), deprecated in favor ...
Definition: pixfmt.h:283
AV_PIX_FMT_YUV422P16
#define AV_PIX_FMT_YUV422P16
Definition: pixfmt.h:494
AV_PIX_FMT_YUVJ422P
@ AV_PIX_FMT_YUVJ422P
planar YUV 4:2:2, 16bpp, full scale (JPEG), deprecated in favor of AV_PIX_FMT_YUV422P and setting col...
Definition: pixfmt.h:86
s
#define s(width, name)
Definition: cbs_vp9.c:198
AV_PIX_FMT_YUVA420P
@ AV_PIX_FMT_YUVA420P
planar YUV 4:2:0, 20bpp, (1 Cr & Cb sample per 2x2 Y & A samples)
Definition: pixfmt.h:108
AV_PIX_FMT_YUV444P16
#define AV_PIX_FMT_YUV444P16
Definition: pixfmt.h:495
AV_PIX_FMT_YUVA444P12
#define AV_PIX_FMT_YUVA444P12
Definition: pixfmt.h:524
AV_PIX_FMT_YUV420P9
#define AV_PIX_FMT_YUV420P9
Definition: pixfmt.h:479
AV_PIX_FMT_YUV420P16
#define AV_PIX_FMT_YUV420P16
Definition: pixfmt.h:493
ctx
AVFormatContext * ctx
Definition: movenc.c:48
AV_PIX_FMT_GRAY14
#define AV_PIX_FMT_GRAY14
Definition: pixfmt.h:465
AV_PIX_FMT_YUV420P
@ AV_PIX_FMT_YUV420P
planar YUV 4:2:0, 12bpp, (1 Cr & Cb sample per 2x2 Y samples)
Definition: pixfmt.h:73
FILTER_INPUTS
#define FILTER_INPUTS(array)
Definition: internal.h:182
AV_PIX_FMT_YUVJ444P
@ AV_PIX_FMT_YUVJ444P
planar YUV 4:4:4, 24bpp, full scale (JPEG), deprecated in favor of AV_PIX_FMT_YUV444P and setting col...
Definition: pixfmt.h:87
frame
static AVFrame * frame
Definition: demux_decode.c:54
AV_PIX_FMT_GRAY10
#define AV_PIX_FMT_GRAY10
Definition: pixfmt.h:463
OFFSET
#define OFFSET(x)
Definition: vf_bbox.c:39
AVClass
Describe the class of an AVClass context structure.
Definition: log.h:66
NULL
#define NULL
Definition: coverity.c:32
AV_PIX_FMT_YUVJ420P
@ AV_PIX_FMT_YUVJ420P
planar YUV 4:2:0, 12bpp, full scale (JPEG), deprecated in favor of AV_PIX_FMT_YUV420P and setting col...
Definition: pixfmt.h:85
AV_PIX_FMT_YUV440P10
#define AV_PIX_FMT_YUV440P10
Definition: pixfmt.h:484
pix_fmts
static enum AVPixelFormat pix_fmts[]
Definition: vf_bbox.c:49
AV_PIX_FMT_YUV422P10
#define AV_PIX_FMT_YUV422P10
Definition: pixfmt.h:483
AV_PIX_FMT_GRAY8
@ AV_PIX_FMT_GRAY8
Y , 8bpp.
Definition: pixfmt.h:81
BBoxContext::depth
int depth
Definition: vf_bbox.c:36
FFBoundingBox::x1
int x1
Definition: bbox.h:27
av_ts2timestr
#define av_ts2timestr(ts, tb)
Convenience macro, the return value should be used only directly in function arguments but never stan...
Definition: timestamp.h:77
NULL_IF_CONFIG_SMALL
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification.
Definition: internal.h:106
AV_PIX_FMT_YUV422P12
#define AV_PIX_FMT_YUV422P12
Definition: pixfmt.h:487
AV_PIX_FMT_YUV444P12
#define AV_PIX_FMT_YUV444P12
Definition: pixfmt.h:489
ff_filter_process_command
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:879
AV_PIX_FMT_YUVA444P
@ AV_PIX_FMT_YUVA444P
planar YUV 4:4:4 32bpp, (1 Cr & Cb sample per 1x1 Y & A samples)
Definition: pixfmt.h:174
AV_PIX_FMT_YUVA444P10
#define AV_PIX_FMT_YUVA444P10
Definition: pixfmt.h:522
AVFILTER_DEFINE_CLASS
AVFILTER_DEFINE_CLASS(bbox)
AV_LOG_INFO
#define AV_LOG_INFO
Standard information.
Definition: log.h:191
internal.h
AVFILTER_FLAG_SUPPORT_TIMELINE_GENERIC
#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:147
FFBoundingBox::y1
int y1
Definition: bbox.h:27
AV_PIX_FMT_YUVJ440P
@ AV_PIX_FMT_YUVJ440P
planar YUV 4:4:0 full scale (JPEG), deprecated in favor of AV_PIX_FMT_YUV440P and setting color_range
Definition: pixfmt.h:107
bbox_options
static const AVOption bbox_options[]
Definition: vf_bbox.c:42
BBoxContext::min_val
int min_val
Definition: vf_bbox.c:35
bbox_outputs
static const AVFilterPad bbox_outputs[]
Definition: vf_bbox.c:138
AVFilterPad::name
const char * name
Pad name.
Definition: internal.h:39
AV_PIX_FMT_YUV444P9
#define AV_PIX_FMT_YUV444P9
Definition: pixfmt.h:481
AVFilter
Filter definition.
Definition: avfilter.h:166
AV_PIX_FMT_YUVA444P9
#define AV_PIX_FMT_YUVA444P9
Definition: pixfmt.h:519
AV_PIX_FMT_YUV420P12
#define AV_PIX_FMT_YUV420P12
Definition: pixfmt.h:486
AV_PIX_FMT_YUV422P14
#define AV_PIX_FMT_YUV422P14
Definition: pixfmt.h:491
AV_PIX_FMT_NONE
@ AV_PIX_FMT_NONE
Definition: pixfmt.h:72
AV_PIX_FMT_YUVA422P12
#define AV_PIX_FMT_YUVA422P12
Definition: pixfmt.h:523
AV_OPT_TYPE_INT
@ AV_OPT_TYPE_INT
Definition: opt.h:225
avfilter.h
AVFrame::metadata
AVDictionary * metadata
metadata.
Definition: frame.h:708
AVFILTER_FLAG_METADATA_ONLY
#define AVFILTER_FLAG_METADATA_ONLY
The filter is a "metadata" filter - it does not modify the frame data in any way.
Definition: avfilter.h:133
ff_vf_bbox
const AVFilter ff_vf_bbox
Definition: vf_bbox.c:146
AV_PIX_FMT_YUV444P
@ AV_PIX_FMT_YUV444P
planar YUV 4:4:4, 24bpp, (1 Cr & Cb sample per 1x1 Y samples)
Definition: pixfmt.h:78
AVFilterContext
An instance of a filter.
Definition: avfilter.h:409
desc
const char * desc
Definition: libsvtav1.c:83
AVMEDIA_TYPE_VIDEO
@ AVMEDIA_TYPE_VIDEO
Definition: avutil.h:201
AV_PIX_FMT_YUV422P
@ AV_PIX_FMT_YUV422P
planar YUV 4:2:2, 16bpp, (1 Cr & Cb sample per 2x1 Y samples)
Definition: pixfmt.h:77
FFBoundingBox
Definition: bbox.h:26
AVPixFmtDescriptor
Descriptor that unambiguously describes how the bits of a pixel are stored in the up to 4 data planes...
Definition: pixdesc.h:69
FILTER_OUTPUTS
#define FILTER_OUTPUTS(array)
Definition: internal.h:183
AV_PIX_FMT_YUV411P
@ AV_PIX_FMT_YUV411P
planar YUV 4:1:1, 12bpp, (1 Cr & Cb sample per 4x1 Y samples)
Definition: pixfmt.h:80
timestamp.h
AVERROR_BUG
#define AVERROR_BUG
Internal bug, also see AVERROR_BUG2.
Definition: error.h:52
AVFrame::linesize
int linesize[AV_NUM_DATA_POINTERS]
For video, a positive or negative value, which is typically indicating the size in bytes of each pict...
Definition: frame.h:385
AV_PIX_FMT_YUV410P
@ AV_PIX_FMT_YUV410P
planar YUV 4:1:0, 9bpp, (1 Cr & Cb sample per 4x4 Y samples)
Definition: pixfmt.h:79
av_log
#define av_log(a,...)
Definition: tableprint_vlc.h:27
av_ts2str
#define av_ts2str(ts)
Convenience macro, the return value should be used only directly in function arguments but never stan...
Definition: timestamp.h:54
AV_PIX_FMT_YUV440P12
#define AV_PIX_FMT_YUV440P12
Definition: pixfmt.h:488
h
h
Definition: vp9dsp_template.c:2038
AV_PIX_FMT_YUV444P14
#define AV_PIX_FMT_YUV444P14
Definition: pixfmt.h:492
AV_PIX_FMT_GRAY12
#define AV_PIX_FMT_GRAY12
Definition: pixfmt.h:464
bbox_inputs
static const AVFilterPad bbox_inputs[]
Definition: vf_bbox.c:130
AV_PIX_FMT_YUVA422P
@ AV_PIX_FMT_YUVA422P
planar YUV 4:2:2 24bpp, (1 Cr & Cb sample per 2x1 Y & A samples)
Definition: pixfmt.h:173
FFBoundingBox::y2
int y2
Definition: bbox.h:27
AV_PIX_FMT_YUV420P14
#define AV_PIX_FMT_YUV420P14
Definition: pixfmt.h:490