aboutsummaryrefslogtreecommitdiffstats
path: root/libavfilter/vf_dnn_detect.c
blob: b5dae42c6506a1443c4711a4abf763d303777fa9 (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
/*
 * This file is part of FFmpeg.
 *
 * FFmpeg is free software; you can redistribute it and/or
 * modify it under the terms of the GNU Lesser General Public
 * License as published by the Free Software Foundation; either
 * version 2.1 of the License, or (at your option) any later version.
 *
 * FFmpeg is distributed in the hope that it will be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
 * Lesser General Public License for more details.
 *
 * You should have received a copy of the GNU Lesser General Public
 * License along with FFmpeg; if not, write to the Free Software
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

/**
 * @file
 * implementing an object detecting filter using deep learning networks.
 */

#include "libavutil/file_open.h"
#include "libavutil/opt.h"
#include "filters.h"
#include "dnn_filter_common.h"
#include "internal.h"
#include "video.h"
#include "libavutil/time.h"
#include "libavutil/avstring.h"
#include "libavutil/detection_bbox.h"

typedef struct DnnDetectContext {
    const AVClass *class;
    DnnContext dnnctx;
    float confidence;
    char *labels_filename;
    char **labels;
    int label_count;
} DnnDetectContext;

#define OFFSET(x) offsetof(DnnDetectContext, dnnctx.x)
#define OFFSET2(x) offsetof(DnnDetectContext, x)
#define FLAGS AV_OPT_FLAG_FILTERING_PARAM | AV_OPT_FLAG_VIDEO_PARAM
static const AVOption dnn_detect_options[] = {
    { "dnn_backend", "DNN backend",                OFFSET(backend_type),     AV_OPT_TYPE_INT,       { .i64 = DNN_OV },    INT_MIN, INT_MAX, FLAGS, "backend" },
#if (CONFIG_LIBTENSORFLOW == 1)
    { "tensorflow",  "tensorflow backend flag",    0,                        AV_OPT_TYPE_CONST,     { .i64 = DNN_TF },    0, 0, FLAGS, "backend" },
#endif
#if (CONFIG_LIBOPENVINO == 1)
    { "openvino",    "openvino backend flag",      0,                        AV_OPT_TYPE_CONST,     { .i64 = DNN_OV },    0, 0, FLAGS, "backend" },
#endif
    DNN_COMMON_OPTIONS
    { "confidence",  "threshold of confidence",    OFFSET2(confidence),      AV_OPT_TYPE_FLOAT,     { .dbl = 0.5 },  0, 1, FLAGS},
    { "labels",      "path to labels file",        OFFSET2(labels_filename), AV_OPT_TYPE_STRING,    { .str = NULL }, 0, 0, FLAGS },
    { NULL }
};

AVFILTER_DEFINE_CLASS(dnn_detect);

static int dnn_detect_post_proc_ov(AVFrame *frame, DNNData *output, AVFilterContext *filter_ctx)
{
    DnnDetectContext *ctx = filter_ctx->priv;
    float conf_threshold = ctx->confidence;
    int proposal_count = output->height;
    int detect_size = output->width;
    float *detections = output->data;
    int nb_bboxes = 0;
    AVFrameSideData *sd;
    AVDetectionBBox *bbox;
    AVDetectionBBoxHeader *header;

    sd = av_frame_get_side_data(frame, AV_FRAME_DATA_DETECTION_BBOXES);
    if (sd) {
        av_log(filter_ctx, AV_LOG_ERROR, "already have bounding boxes in side data.\n");
        return -1;
    }

    for (int i = 0; i < proposal_count; ++i) {
        float conf = detections[i * detect_size + 2];
        if (conf < conf_threshold) {
            continue;
        }
        nb_bboxes++;
    }

    if (nb_bboxes == 0) {
        av_log(filter_ctx, AV_LOG_VERBOSE, "nothing detected in this frame.\n");
        return 0;
    }

    header = av_detection_bbox_create_side_data(frame, nb_bboxes);
    if (!header) {
        av_log(filter_ctx, AV_LOG_ERROR, "failed to create side data with %d bounding boxes\n", nb_bboxes);
        return -1;
    }

    av_strlcpy(header->source, ctx->dnnctx.model_filename, sizeof(header->source));

    for (int i = 0; i < proposal_count; ++i) {
        int av_unused image_id = (int)detections[i * detect_size + 0];
        int label_id = (int)detections[i * detect_size + 1];
        float conf   =      detections[i * detect_size + 2];
        float x0     =      detections[i * detect_size + 3];
        float y0     =      detections[i * detect_size + 4];
        float x1     =      detections[i * detect_size + 5];
        float y1     =      detections[i * detect_size + 6];

        if (conf < conf_threshold) {
            continue;
        }

        bbox = av_get_detection_bbox(header, header->nb_bboxes - nb_bboxes);
        bbox->x = (int)(x0 * frame->width);
        bbox->w = (int)(x1 * frame->width) - bbox->x;
        bbox->y = (int)(y0 * frame->height);
        bbox->h = (int)(y1 * frame->height) - bbox->y;

        bbox->detect_confidence = av_make_q((int)(conf * 10000), 10000);
        bbox->classify_count = 0;

        if (ctx->labels && label_id < ctx->label_count) {
            av_strlcpy(bbox->detect_label, ctx->labels[label_id], sizeof(bbox->detect_label));
        } else {
            snprintf(bbox->detect_label, sizeof(bbox->detect_label), "%d", label_id);
        }

        nb_bboxes--;
        if (nb_bboxes == 0) {
            break;
        }
    }

    return 0;
}

static int dnn_detect_post_proc_tf(AVFrame *frame, DNNData *output, AVFilterContext *filter_ctx)
{
    DnnDetectContext *ctx = filter_ctx->priv;
    int proposal_count;
    float conf_threshold = ctx->confidence;
    float *conf, *position, *label_id, x0, y0, x1, y1;
    int nb_bboxes = 0;
    AVFrameSideData *sd;
    AVDetectionBBox *bbox;
    AVDetectionBBoxHeader *header;

    proposal_count = *(float *)(output[0].data);
    conf           = output[1].data;
    position       = output[3].data;
    label_id       = output[2].data;

    sd = av_frame_get_side_data(frame, AV_FRAME_DATA_DETECTION_BBOXES);
    if (sd) {
        av_log(filter_ctx, AV_LOG_ERROR, "already have dnn bounding boxes in side data.\n");
        return -1;
    }

    for (int i = 0; i < proposal_count; ++i) {
        if (conf[i] < conf_threshold)
            continue;
        nb_bboxes++;
    }

    if (nb_bboxes == 0) {
        av_log(filter_ctx, AV_LOG_VERBOSE, "nothing detected in this frame.\n");
        return 0;
    }

    header = av_detection_bbox_create_side_data(frame, nb_bboxes);
    if (!header) {
        av_log(filter_ctx, AV_LOG_ERROR, "failed to create side data with %d bounding boxes\n", nb_bboxes);
        return -1;
    }

    av_strlcpy(header->source, ctx->dnnctx.model_filename, sizeof(header->source));

    for (int i = 0; i < proposal_count; ++i) {
        y0 = position[i * 4];
        x0 = position[i * 4 + 1];
        y1 = position[i * 4 + 2];
        x1 = position[i * 4 + 3];

        bbox = av_get_detection_bbox(header, i);

        if (conf[i] < conf_threshold) {
            continue;
        }

        bbox->x = (int)(x0 * frame->width);
        bbox->w = (int)(x1 * frame->width) - bbox->x;
        bbox->y = (int)(y0 * frame->height);
        bbox->h = (int)(y1 * frame->height) - bbox->y;

        bbox->detect_confidence = av_make_q((int)(conf[i] * 10000), 10000);
        bbox->classify_count = 0;

        if (ctx->labels && label_id[i] < ctx->label_count) {
            av_strlcpy(bbox->detect_label, ctx->labels[(int)label_id[i]], sizeof(bbox->detect_label));
        } else {
            snprintf(bbox->detect_label, sizeof(bbox->detect_label), "%d", (int)label_id[i]);
        }

        nb_bboxes--;
        if (nb_bboxes == 0) {
            break;
        }
    }
    return 0;
}

static int dnn_detect_post_proc(AVFrame *frame, DNNData *output, uint32_t nb, AVFilterContext *filter_ctx)
{
    DnnDetectContext *ctx = filter_ctx->priv;
    DnnContext *dnn_ctx = &ctx->dnnctx;
    switch (dnn_ctx->backend_type) {
    case DNN_OV:
        return dnn_detect_post_proc_ov(frame, output, filter_ctx);
    case DNN_TF:
        return dnn_detect_post_proc_tf(frame, output, filter_ctx);
    default:
        avpriv_report_missing_feature(filter_ctx, "Current dnn backend does not support detect filter\n");
        return AVERROR(EINVAL);
    }
}

static void free_detect_labels(DnnDetectContext *ctx)
{
    for (int i = 0; i < ctx->label_count; i++) {
        av_freep(&ctx->labels[i]);
    }
    ctx->label_count = 0;
    av_freep(&ctx->labels);
}

static int read_detect_label_file(AVFilterContext *context)
{
    int line_len;
    FILE *file;
    DnnDetectContext *ctx = context->priv;

    file = avpriv_fopen_utf8(ctx->labels_filename, "r");
    if (!file){
        av_log(context, AV_LOG_ERROR, "failed to open file %s\n", ctx->labels_filename);
        return AVERROR(EINVAL);
    }

    while (!feof(file)) {
        char *label;
        char buf[256];
        if (!fgets(buf, 256, file)) {
            break;
        }

        line_len = strlen(buf);
        while (line_len) {
            int i = line_len - 1;
            if (buf[i] == '\n' || buf[i] == '\r' || buf[i] == ' ') {
                buf[i] = '\0';
                line_len--;
            } else {
                break;
            }
        }

        if (line_len == 0)  // empty line
            continue;

        if (line_len >= AV_DETECTION_BBOX_LABEL_NAME_MAX_SIZE) {
            av_log(context, AV_LOG_ERROR, "label %s too long\n", buf);
            fclose(file);
            return AVERROR(EINVAL);
        }

        label = av_strdup(buf);
        if (!label) {
            av_log(context, AV_LOG_ERROR, "failed to allocate memory for label %s\n", buf);
            fclose(file);
            return AVERROR(ENOMEM);
        }

        if (av_dynarray_add_nofree(&ctx->labels, &ctx->label_count, label) < 0) {
            av_log(context, AV_LOG_ERROR, "failed to do av_dynarray_add\n");
            fclose(file);
            av_freep(&label);
            return AVERROR(ENOMEM);
        }
    }

    fclose(file);
    return 0;
}

static int check_output_nb(DnnDetectContext *ctx, DNNBackendType backend_type, int output_nb)
{
    switch(backend_type) {
    case DNN_TF:
        if (output_nb != 4) {
            av_log(ctx, AV_LOG_ERROR, "Only support tensorflow detect model with 4 outputs, \
                                       but get %d instead\n", output_nb);
            return AVERROR(EINVAL);
        }
        return 0;
    case DNN_OV:
        if (output_nb != 1) {
            av_log(ctx, AV_LOG_ERROR, "Dnn detect filter with openvino backend needs 1 output only, \
                                       but get %d instead\n", output_nb);
            return AVERROR(EINVAL);
        }
        return 0;
    default:
        avpriv_report_missing_feature(ctx, "Dnn detect filter does not support current backend\n");
        return AVERROR(EINVAL);
    }
    return 0;
}

static av_cold int dnn_detect_init(AVFilterContext *context)
{
    DnnDetectContext *ctx = context->priv;
    DnnContext *dnn_ctx = &ctx->dnnctx;
    int ret;

    ret = ff_dnn_init(&ctx->dnnctx, DFT_ANALYTICS_DETECT, context);
    if (ret < 0)
        return ret;
    ret = check_output_nb(ctx, dnn_ctx->backend_type, dnn_ctx->nb_outputs);
    if (ret < 0)
        return ret;
    ff_dnn_set_detect_post_proc(&ctx->dnnctx, dnn_detect_post_proc);

    if (ctx->labels_filename) {
        return read_detect_label_file(context);
    }
    return 0;
}

static const enum AVPixelFormat pix_fmts[] = {
    AV_PIX_FMT_RGB24, AV_PIX_FMT_BGR24,
    AV_PIX_FMT_GRAY8, AV_PIX_FMT_GRAYF32,
    AV_PIX_FMT_YUV420P, AV_PIX_FMT_YUV422P,
    AV_PIX_FMT_YUV444P, AV_PIX_FMT_YUV410P, AV_PIX_FMT_YUV411P,
    AV_PIX_FMT_NV12,
    AV_PIX_FMT_NONE
};

static int dnn_detect_flush_frame(AVFilterLink *outlink, int64_t pts, int64_t *out_pts)
{
    DnnDetectContext *ctx = outlink->src->priv;
    int ret;
    DNNAsyncStatusType async_state;

    ret = ff_dnn_flush(&ctx->dnnctx);
    if (ret != 0) {
        return -1;
    }

    do {
        AVFrame *in_frame = NULL;
        AVFrame *out_frame = NULL;
        async_state = ff_dnn_get_result(&ctx->dnnctx, &in_frame, &out_frame);
        if (async_state == DAST_SUCCESS) {
            ret = ff_filter_frame(outlink, in_frame);
            if (ret < 0)
                return ret;
            if (out_pts)
                *out_pts = in_frame->pts + pts;
        }
        av_usleep(5000);
    } while (async_state >= DAST_NOT_READY);

    return 0;
}

static int dnn_detect_activate(AVFilterContext *filter_ctx)
{
    AVFilterLink *inlink = filter_ctx->inputs[0];
    AVFilterLink *outlink = filter_ctx->outputs[0];
    DnnDetectContext *ctx = filter_ctx->priv;
    AVFrame *in = NULL;
    int64_t pts;
    int ret, status;
    int got_frame = 0;
    int async_state;

    FF_FILTER_FORWARD_STATUS_BACK(outlink, inlink);

    do {
        // drain all input frames
        ret = ff_inlink_consume_frame(inlink, &in);
        if (ret < 0)
            return ret;
        if (ret > 0) {
            if (ff_dnn_execute_model(&ctx->dnnctx, in, NULL) != 0) {
                return AVERROR(EIO);
            }
        }
    } while (ret > 0);

    // drain all processed frames
    do {
        AVFrame *in_frame = NULL;
        AVFrame *out_frame = NULL;
        async_state = ff_dnn_get_result(&ctx->dnnctx, &in_frame, &out_frame);
        if (async_state == DAST_SUCCESS) {
            ret = ff_filter_frame(outlink, in_frame);
            if (ret < 0)
                return ret;
            got_frame = 1;
        }
    } while (async_state == DAST_SUCCESS);

    // if frame got, schedule to next filter
    if (got_frame)
        return 0;

    if (ff_inlink_acknowledge_status(inlink, &status, &pts)) {
        if (status == AVERROR_EOF) {
            int64_t out_pts = pts;
            ret = dnn_detect_flush_frame(outlink, pts, &out_pts);
            ff_outlink_set_status(outlink, status, out_pts);
            return ret;
        }
    }

    FF_FILTER_FORWARD_WANTED(outlink, inlink);

    return 0;
}

static av_cold void dnn_detect_uninit(AVFilterContext *context)
{
    DnnDetectContext *ctx = context->priv;
    ff_dnn_uninit(&ctx->dnnctx);
    free_detect_labels(ctx);
}

const AVFilter ff_vf_dnn_detect = {
    .name          = "dnn_detect",
    .description   = NULL_IF_CONFIG_SMALL("Apply DNN detect filter to the input."),
    .priv_size     = sizeof(DnnDetectContext),
    .init          = dnn_detect_init,
    .uninit        = dnn_detect_uninit,
    FILTER_INPUTS(ff_video_default_filterpad),
    FILTER_OUTPUTS(ff_video_default_filterpad),
    FILTER_PIXFMTS_ARRAY(pix_fmts),
    .priv_class    = &dnn_detect_class,
    .activate      = dnn_detect_activate,
};