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 33typedef struct { 34 const AVClass *class; 35 int min_val; 36} BBoxContext; 37 38#define OFFSET(x) offsetof(BBoxContext, x) 39#define FLAGS AV_OPT_FLAG_VIDEO_PARAM|AV_OPT_FLAG_FILTERING_PARAM 40 41static const AVOption bbox_options[] = { 42 { "min_val", "set minimum luminance value for bounding box", OFFSET(min_val), AV_OPT_TYPE_INT, { .i64 = 16 }, 0, 254, FLAGS }, 43 { NULL } 44}; 45 46AVFILTER_DEFINE_CLASS(bbox); 47 48static int query_formats(AVFilterContext *ctx) 49{ 50 static const enum AVPixelFormat pix_fmts[] = { 51 AV_PIX_FMT_YUV420P, 52 AV_PIX_FMT_YUV444P, 53 AV_PIX_FMT_YUV440P, 54 AV_PIX_FMT_YUV422P, 55 AV_PIX_FMT_YUV411P, 56 AV_PIX_FMT_NONE, 57 }; 58 59 ff_set_common_formats(ctx, ff_make_format_list(pix_fmts)); 60 return 0; 61} 62 63#define SET_META(key, value) \ 64 snprintf(buf, sizeof(buf), "%d", value); \ 65 av_dict_set(metadata, key, buf, 0); 66 67static int filter_frame(AVFilterLink *inlink, AVFrame *frame) 68{ 69 AVFilterContext *ctx = inlink->dst; 70 BBoxContext *bbox = ctx->priv; 71 FFBoundingBox box; 72 int has_bbox, w, h; 73 char buf[32]; 74 75 has_bbox = 76 ff_calculate_bounding_box(&box, 77 frame->data[0], frame->linesize[0], 78 inlink->w, inlink->h, bbox->min_val); 79 w = box.x2 - box.x1 + 1; 80 h = box.y2 - box.y1 + 1; 81 82 av_log(ctx, AV_LOG_INFO, 83 "n:%"PRId64" pts:%s pts_time:%s", inlink->frame_count, 84 av_ts2str(frame->pts), av_ts2timestr(frame->pts, &inlink->time_base)); 85 86 if (has_bbox) { 87 AVDictionary **metadata = avpriv_frame_get_metadatap(frame); 88 89 SET_META("lavfi.bbox.x1", box.x1) 90 SET_META("lavfi.bbox.x2", box.x2) 91 SET_META("lavfi.bbox.y1", box.y1) 92 SET_META("lavfi.bbox.y2", box.y2) 93 SET_META("lavfi.bbox.w", w) 94 SET_META("lavfi.bbox.h", h) 95 96 av_log(ctx, AV_LOG_INFO, 97 " x1:%d x2:%d y1:%d y2:%d w:%d h:%d" 98 " crop=%d:%d:%d:%d drawbox=%d:%d:%d:%d", 99 box.x1, box.x2, box.y1, box.y2, w, h, 100 w, h, box.x1, box.y1, /* crop params */ 101 box.x1, box.y1, w, h); /* drawbox params */ 102 } 103 av_log(ctx, AV_LOG_INFO, "\n"); 104 105 return ff_filter_frame(inlink->dst->outputs[0], frame); 106} 107 108static const AVFilterPad bbox_inputs[] = { 109 { 110 .name = "default", 111 .type = AVMEDIA_TYPE_VIDEO, 112 .filter_frame = filter_frame, 113 }, 114 { NULL } 115}; 116 117static const AVFilterPad bbox_outputs[] = { 118 { 119 .name = "default", 120 .type = AVMEDIA_TYPE_VIDEO, 121 }, 122 { NULL } 123}; 124 125AVFilter ff_vf_bbox = { 126 .name = "bbox", 127 .description = NULL_IF_CONFIG_SMALL("Compute bounding box for each frame."), 128 .priv_size = sizeof(BBoxContext), 129 .priv_class = &bbox_class, 130 .query_formats = query_formats, 131 .inputs = bbox_inputs, 132 .outputs = bbox_outputs, 133 .flags = AVFILTER_FLAG_SUPPORT_TIMELINE_GENERIC, 134}; 135