-
Notifications
You must be signed in to change notification settings - Fork 3
/
ffmpeg4_modifications.diff
1362 lines (1350 loc) · 50.5 KB
/
ffmpeg4_modifications.diff
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
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
diff --git a/configure b/configure
index 06e3a7b..9f548b3 100755
--- a/configure
+++ b/configure
@@ -266,6 +266,7 @@ External library support:
--enable-libspeex enable Speex de/encoding via libspeex [no]
--enable-libsrt enable Haivision SRT protocol via libsrt [no]
--enable-libssh enable SFTP protocol via libssh [no]
+ --enable-libsvtav1 enable AV1 encoding via svt [no]
--enable-libtensorflow enable TensorFlow as a DNN module backend
for DNN based filters like sr [no]
--enable-libtesseract enable Tesseract, needed for ocr filter [no]
@@ -1799,6 +1800,7 @@ EXTERNAL_LIBRARY_LIST="
libspeex
libsrt
libssh
+ libsvtav1
libtensorflow
libtesseract
libtheora
@@ -3224,6 +3226,7 @@ libshine_encoder_select="audio_frame_queue"
libspeex_decoder_deps="libspeex"
libspeex_encoder_deps="libspeex"
libspeex_encoder_select="audio_frame_queue"
+libsvt_av1_encoder_deps="libsvtav1"
libtheora_encoder_deps="libtheora"
libtwolame_encoder_deps="libtwolame"
libvo_amrwbenc_encoder_deps="libvo_amrwbenc"
@@ -3530,6 +3533,8 @@ nnedi_filter_deps="gpl"
ocr_filter_deps="libtesseract"
ocv_filter_deps="libopencv"
openclsrc_filter_deps="opencl"
+phqm_filter_deps="libopencv"
+phqm_filter_extralibs="-lstdc++ -lopencv_img_hash"
overlay_opencl_filter_deps="opencl"
overlay_qsv_filter_deps="libmfx"
overlay_qsv_filter_select="qsvvpp"
@@ -6329,6 +6334,7 @@ enabled libsoxr && require libsoxr soxr.h soxr_create -lsoxr
enabled libssh && require_pkg_config libssh libssh libssh/sftp.h sftp_init
enabled libspeex && require_pkg_config libspeex speex speex/speex.h speex_decoder_init
enabled libsrt && require_pkg_config libsrt "srt >= 1.3.0" srt/srt.h srt_socket
+enabled libsvtav1 && require_pkg_config libsvtav1 SvtAv1Enc EbSvtAv1Enc.h svt_av1_enc_init_handle
enabled libtensorflow && require libtensorflow tensorflow/c/c_api.h TF_Version -ltensorflow
enabled libtesseract && require_pkg_config libtesseract tesseract tesseract/capi.h TessBaseAPICreate
enabled libtheora && require libtheora theora/theoraenc.h th_info_init -ltheoraenc -ltheoradec -logg
diff --git a/libavcodec/Makefile b/libavcodec/Makefile
index 0fd374f..d362a83 100644
--- a/libavcodec/Makefile
+++ b/libavcodec/Makefile
@@ -1005,6 +1005,7 @@ OBJS-$(CONFIG_LIBRAV1E_ENCODER) += librav1e.o
OBJS-$(CONFIG_LIBSHINE_ENCODER) += libshine.o
OBJS-$(CONFIG_LIBSPEEX_DECODER) += libspeexdec.o
OBJS-$(CONFIG_LIBSPEEX_ENCODER) += libspeexenc.o
+OBJS-$(CONFIG_LIBSVT_AV1_ENCODER) += libsvt_av1.o
OBJS-$(CONFIG_LIBTHEORA_ENCODER) += libtheoraenc.o
OBJS-$(CONFIG_LIBTWOLAME_ENCODER) += libtwolame.o
OBJS-$(CONFIG_LIBVO_AMRWBENC_ENCODER) += libvo-amrwbenc.o
diff --git a/libavcodec/allcodecs.c b/libavcodec/allcodecs.c
index f4cf180..5ff2e60 100644
--- a/libavcodec/allcodecs.c
+++ b/libavcodec/allcodecs.c
@@ -717,6 +717,7 @@ extern AVCodec ff_librsvg_decoder;
extern AVCodec ff_libshine_encoder;
extern AVCodec ff_libspeex_encoder;
extern AVCodec ff_libspeex_decoder;
+extern AVCodec ff_libsvt_av1_encoder;
extern AVCodec ff_libtheora_encoder;
extern AVCodec ff_libtwolame_encoder;
extern AVCodec ff_libvo_amrwbenc_encoder;
diff --git a/libavcodec/libsvt_av1.c b/libavcodec/libsvt_av1.c
new file mode 100644
index 0000000..5f1d8e0
--- /dev/null
+++ b/libavcodec/libsvt_av1.c
@@ -0,0 +1,548 @@
+/*
+* Scalable Video Technology for AV1 encoder library plugin
+*
+* Copyright (c) 2018 Intel Corporation
+*
+* 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 this program; if not, write to the Free Software
+* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
+*/
+
+#include <stdint.h>
+#include "EbSvtAv1ErrorCodes.h"
+#include "EbSvtAv1Enc.h"
+
+#include "libavutil/common.h"
+#include "libavutil/frame.h"
+#include "libavutil/opt.h"
+#include "libavutil/pixdesc.h"
+#include "libavutil/avassert.h"
+
+#include "internal.h"
+#include "avcodec.h"
+
+typedef enum eos_status {
+ EOS_NOT_REACHED = 0,
+ EOS_SENT,
+ EOS_RECEIVED
+}EOS_STATUS;
+
+typedef struct SvtContext {
+ AVClass *class;
+
+ EbSvtAv1EncConfiguration enc_params;
+ EbComponentType *svt_handle;
+
+ EbBufferHeaderType *in_buf;
+ int raw_size;
+ int max_tu_size;
+
+ AVBufferPool* pool;
+
+ EOS_STATUS eos_flag;
+
+ // User options.
+ int hierarchical_level;
+ int la_depth;
+ int enc_mode;
+ int rc_mode;
+ int scd;
+ int qp;
+
+ int forced_idr;
+
+ int tier;
+ int level;
+ int profile;
+
+
+} SvtContext;
+
+static const struct {
+ EbErrorType eb_err;
+ int av_err;
+ const char *desc;
+} svt_errors[] = {
+ { EB_ErrorNone, 0, "success" },
+ { EB_ErrorInsufficientResources, AVERROR(ENOMEM), "insufficient resources" },
+ { EB_ErrorUndefined, AVERROR(EINVAL), "undefined error" },
+ { EB_ErrorInvalidComponent, AVERROR(EINVAL), "invalid component" },
+ { EB_ErrorBadParameter, AVERROR(EINVAL), "bad parameter" },
+ { EB_ErrorDestroyThreadFailed, AVERROR_EXTERNAL, "failed to destory thread" },
+ { EB_ErrorSemaphoreUnresponsive, AVERROR_EXTERNAL, "semaphore unresponsive" },
+ { EB_ErrorDestroySemaphoreFailed, AVERROR_EXTERNAL, "semaphore unresponsive" },
+ { EB_ErrorCreateMutexFailed, AVERROR_EXTERNAL, "failed to creat mutex" },
+ { EB_ErrorMutexUnresponsive, AVERROR_EXTERNAL, "mutex unresponsive" },
+ { EB_ErrorDestroyMutexFailed, AVERROR_EXTERNAL, "failed to destory muxtex" },
+ { EB_NoErrorEmptyQueue, AVERROR(EAGAIN), "empty queue" },
+};
+
+static int svt_map_error(EbErrorType eb_err, const char **desc)
+{
+ int i;
+
+ av_assert0(desc);
+ for (i = 0; i < FF_ARRAY_ELEMS(svt_errors); i++) {
+ if (svt_errors[i].eb_err == eb_err) {
+ *desc = svt_errors[i].desc;
+ return svt_errors[i].av_err;
+ }
+ }
+ *desc = "unknown error";
+ return AVERROR_UNKNOWN;
+}
+
+static int svt_print_error(void *log_ctx, EbErrorType err,
+ const char *error_string)
+{
+ const char *desc;
+ int ret;
+ ret = svt_map_error(err, &desc);
+ av_log(log_ctx, AV_LOG_ERROR, "%s: %s (0x%x)\n", error_string, desc, err);
+ return ret;
+}
+
+static void free_buffer(SvtContext *svt_enc)
+{
+ if (svt_enc->in_buf) {
+ EbSvtIOFormat *in_data = (EbSvtIOFormat *)svt_enc->in_buf->p_buffer;
+ av_freep(&in_data);
+ av_freep(&svt_enc->in_buf);
+ }
+ av_buffer_pool_uninit(&svt_enc->pool);
+}
+
+static int alloc_buffer(EbSvtAv1EncConfiguration *config, SvtContext *svt_enc)
+{
+ const int pack_mode_10bit =
+ (config->encoder_bit_depth > 8) && (config->compressed_ten_bit_format == 0) ? 1 : 0;
+ const size_t luma_size_8bit =
+ config->source_width * config->source_height * (1 << pack_mode_10bit);
+ const size_t luma_size_10bit =
+ (config->encoder_bit_depth > 8 && pack_mode_10bit == 0) ? luma_size_8bit : 0;
+
+ EbSvtIOFormat *in_data;
+
+ svt_enc->raw_size = (luma_size_8bit + luma_size_10bit) * 3 / 2;
+
+ // allocate buffer for in and out
+ svt_enc->in_buf = av_mallocz(sizeof(*svt_enc->in_buf));
+ if (!svt_enc->in_buf)
+ return AVERROR(ENOMEM);
+
+ svt_enc->in_buf->p_buffer = (unsigned char *)av_mallocz(sizeof(*in_data));
+ if (!svt_enc->in_buf->p_buffer)
+ return AVERROR(ENOMEM);
+
+ svt_enc->in_buf->size = sizeof(*svt_enc->in_buf);
+ svt_enc->in_buf->p_app_private = NULL;
+
+ return 0;
+
+}
+
+static int config_enc_params(EbSvtAv1EncConfiguration *param,
+ AVCodecContext *avctx)
+{
+ SvtContext *svt_enc = avctx->priv_data;
+ const AVPixFmtDescriptor *desc;
+ int ret;
+
+ param->source_width = avctx->width;
+ param->source_height = avctx->height;
+
+ desc = av_pix_fmt_desc_get(avctx->pix_fmt);
+ param->encoder_bit_depth = desc->comp[0].depth;
+ av_log(avctx, AV_LOG_DEBUG , "Encoder %d bits depth input\n", param->encoder_bit_depth);
+
+ if (desc->log2_chroma_w == 1 && desc->log2_chroma_h == 1)
+ param->encoder_color_format = EB_YUV420;
+ else if (desc->log2_chroma_w == 1 && desc->log2_chroma_h == 0)
+ param->encoder_color_format = EB_YUV422;
+ else if (!desc->log2_chroma_w && !desc->log2_chroma_h)
+ param->encoder_color_format = EB_YUV444;
+ else {
+ av_log(avctx, AV_LOG_ERROR , "Unsupported pixel format\n");
+ return AVERROR(EINVAL);
+ }
+ av_log(avctx, AV_LOG_DEBUG , "Encoder color format is %d \n", param->encoder_color_format);
+
+ param->profile = svt_enc->profile;
+
+ if ((param->encoder_color_format == EB_YUV422 || param->encoder_bit_depth > 10)
+ && param->profile != PROFESSIONAL_PROFILE ) {
+ av_log(avctx, AV_LOG_WARNING, "Force to be professional profile \n");
+ param->profile = PROFESSIONAL_PROFILE;
+ } else if (param->encoder_color_format == EB_YUV444 && param->profile != HIGH_PROFILE) {
+ av_log(avctx, AV_LOG_WARNING, "Force to be high profile \n");
+ param->profile = HIGH_PROFILE;
+ }
+
+ // Update param from options
+ param->hierarchical_levels = svt_enc->hierarchical_level;
+ param->enc_mode = svt_enc->enc_mode;
+ param->tier = svt_enc->tier;
+ param->level = svt_enc->level;
+ param->rate_control_mode = svt_enc->rc_mode;
+ param->scene_change_detection = svt_enc->scd;
+
+ param->qp = svt_enc->qp;
+
+
+ param->target_bit_rate = avctx->bit_rate;
+ if (avctx->gop_size > 0)
+ param->intra_period_length = avctx->gop_size - 1;
+
+ if (avctx->framerate.num > 0 && avctx->framerate.den > 0) {
+ param->frame_rate_numerator = avctx->framerate.num;
+ param->frame_rate_denominator = avctx->framerate.den * avctx->ticks_per_frame;
+ } else {
+ param->frame_rate_numerator = avctx->time_base.den;
+ param->frame_rate_denominator = avctx->time_base.num * avctx->ticks_per_frame;
+ }
+
+ if (param->rate_control_mode) {
+ param->max_qp_allowed = avctx->qmax;
+ param->min_qp_allowed = avctx->qmin;
+ }
+
+ param->intra_refresh_type = svt_enc->forced_idr + 1;
+
+ if (svt_enc->la_depth != -1)
+ param->look_ahead_distance = svt_enc->la_depth;
+
+ return 0;
+}
+
+static void read_in_data(const AVFrame *frame,
+ EbBufferHeaderType *header_ptr)
+{
+ EbSvtIOFormat *in_data = (EbSvtIOFormat *)header_ptr->p_buffer;
+ const AVPixFmtDescriptor *desc;
+ int i, bytes_shift, plane_h;
+
+ desc = av_pix_fmt_desc_get(frame->format);
+ bytes_shift = desc->comp[0].depth > 8 ? 1 : 0;
+
+ in_data->luma = frame->data[0];
+ in_data->cb = frame->data[1];
+ in_data->cr = frame->data[2];
+
+ in_data->y_stride = AV_CEIL_RSHIFT(frame->linesize[0], bytes_shift);
+ in_data->cb_stride = AV_CEIL_RSHIFT(frame->linesize[1], bytes_shift);
+ in_data->cr_stride = AV_CEIL_RSHIFT(frame->linesize[2], bytes_shift);
+
+ for (i = 0; i < desc->nb_components; i++) {
+ plane_h = frame->height;
+ if (i > 0)
+ plane_h = AV_CEIL_RSHIFT(plane_h, desc->log2_chroma_h);
+ header_ptr->n_filled_len += frame->linesize[i] * plane_h;
+ }
+}
+
+static av_cold int eb_enc_init(AVCodecContext *avctx)
+{
+ SvtContext *svt_enc = avctx->priv_data;
+ EbErrorType svt_ret;
+ int ret;
+
+ svt_enc->eos_flag = EOS_NOT_REACHED;
+
+ svt_ret = svt_av1_enc_init_handle(&svt_enc->svt_handle, svt_enc, &svt_enc->enc_params);
+ if (svt_ret != EB_ErrorNone) {
+ return svt_print_error(avctx, svt_ret, "Error init encoder handle");
+ }
+
+ ret = config_enc_params(&svt_enc->enc_params, avctx);
+ if (ret < 0) {
+ av_log(avctx, AV_LOG_ERROR, "Error configure encoder parameters\n");
+ return ret;
+ }
+
+ svt_ret = svt_av1_enc_set_parameter(svt_enc->svt_handle, &svt_enc->enc_params);
+ if (svt_ret != EB_ErrorNone) {
+ return svt_print_error(avctx, svt_ret, "Error setting encoder parameters");
+ }
+
+ svt_ret = svt_av1_enc_init(svt_enc->svt_handle);
+ if (svt_ret != EB_ErrorNone) {
+ svt_av1_enc_deinit_handle(svt_enc->svt_handle);
+ svt_enc->svt_handle = NULL;
+ return svt_print_error(avctx, svt_ret, "Error init encoder");
+ }
+
+ if (avctx->flags & AV_CODEC_FLAG_GLOBAL_HEADER) {
+ EbBufferHeaderType *headerPtr = NULL;
+
+ svt_ret = svt_av1_enc_stream_header(svt_enc->svt_handle, &headerPtr);
+ if (svt_ret != EB_ErrorNone) {
+ return svt_print_error(avctx, svt_ret, "Error when build stream header");
+ }
+
+ avctx->extradata_size = headerPtr->n_filled_len;
+ avctx->extradata = av_mallocz(avctx->extradata_size + AV_INPUT_BUFFER_PADDING_SIZE);
+ if (!avctx->extradata) {
+ av_log(avctx, AV_LOG_ERROR,
+ "Cannot allocate AV1 header of size %d.\n", avctx->extradata_size);
+ return AVERROR(ENOMEM);
+ }
+
+ memcpy(avctx->extradata, headerPtr->p_buffer, avctx->extradata_size);
+
+ svt_ret = svt_av1_enc_stream_header_release(headerPtr);
+ if (svt_ret != EB_ErrorNone) {
+ return svt_print_error(avctx, svt_ret, "Error when destroy stream header");
+ }
+ }
+
+ ret = alloc_buffer(&svt_enc->enc_params, svt_enc);
+ if (ret < 0)
+ return ret;
+ return 0;
+
+}
+
+static int eb_send_frame(AVCodecContext *avctx, const AVFrame *frame)
+{
+ SvtContext *svt_enc = avctx->priv_data;
+ EbBufferHeaderType *headerPtr = svt_enc->in_buf;
+
+ if (!frame) {
+ EbBufferHeaderType headerPtrLast;
+ headerPtrLast.n_alloc_len = 0;
+ headerPtrLast.n_filled_len = 0;
+ headerPtrLast.n_tick_count = 0;
+ headerPtrLast.p_app_private = NULL;
+ headerPtrLast.p_buffer = NULL;
+ headerPtrLast.flags = EB_BUFFERFLAG_EOS;
+
+ svt_av1_enc_send_picture(svt_enc->svt_handle, &headerPtrLast);
+ svt_enc->eos_flag = EOS_SENT;
+ av_log(avctx, AV_LOG_DEBUG, "Finish sending frames!!!\n");
+ return 0;
+ }
+
+ read_in_data(frame, headerPtr);
+
+ headerPtr->flags = 0;
+ headerPtr->p_app_private = NULL;
+ headerPtr->pts = frame->pts;
+
+ svt_av1_enc_send_picture(svt_enc->svt_handle, headerPtr);
+
+ return 0;
+}
+
+static AVBufferRef* get_output_ref(AVCodecContext *avctx, SvtContext *svt_enc, int filled_len)
+{
+ if (filled_len > svt_enc->max_tu_size) {
+ const int MAX_FRAMES = 8;
+ int max_tu_size;
+ if (filled_len > svt_enc->raw_size * MAX_FRAMES) {
+ av_log(avctx, AV_LOG_ERROR, "something wrong in libsvtav1, the tu size > %d raw frame size.\n", MAX_FRAMES);
+ return NULL;
+ }
+ max_tu_size = 1 << av_ceil_log2(filled_len);
+ av_buffer_pool_uninit(&svt_enc->pool);
+ svt_enc->pool = av_buffer_pool_init(max_tu_size, NULL);
+ if (!svt_enc->pool) {
+ return NULL;
+ }
+ svt_enc->max_tu_size = max_tu_size;
+ }
+ if (!svt_enc->pool) {
+ av_log(avctx, AV_LOG_ERROR, "bug, no buffer pool.\n");
+ return NULL;
+ }
+ return av_buffer_pool_get(svt_enc->pool);
+}
+
+static int eb_receive_packet(AVCodecContext *avctx, AVPacket *pkt)
+{
+ SvtContext *svt_enc = avctx->priv_data;
+ EbBufferHeaderType *headerPtr;
+ EbErrorType svt_ret;
+ int ret = 0, pict_type;
+ AVBufferRef* ref;
+
+ if (svt_enc->eos_flag == EOS_RECEIVED)
+ return AVERROR_EOF;
+
+ svt_ret = svt_av1_enc_get_packet(svt_enc->svt_handle, &headerPtr, svt_enc->eos_flag);
+ if (svt_ret == EB_NoErrorEmptyQueue)
+ return AVERROR(EAGAIN);
+
+ ref = get_output_ref(avctx, svt_enc, headerPtr->n_filled_len);
+ if (!ref) {
+ av_log(avctx, AV_LOG_ERROR, "Failed to allocate output packet.\n");
+ svt_av1_enc_release_out_buffer(&headerPtr);
+ return AVERROR(ENOMEM);
+ }
+ pkt->buf = ref;
+ pkt->data = ref->data;
+
+ memcpy(pkt->data, headerPtr->p_buffer, headerPtr->n_filled_len);
+ pkt->size = headerPtr->n_filled_len;
+ pkt->pts = headerPtr->pts;
+ pkt->dts = headerPtr->dts;
+ if (headerPtr->pic_type == EB_AV1_KEY_PICTURE) {
+ pkt->flags |= AV_PKT_FLAG_KEY;
+ pict_type = AV_PICTURE_TYPE_I;
+ } else if (headerPtr->pic_type == EB_AV1_INTRA_ONLY_PICTURE) {
+ pict_type = AV_PICTURE_TYPE_I;
+ } else if (headerPtr->pic_type == EB_AV1_INVALID_PICTURE) {
+ pict_type = AV_PICTURE_TYPE_NONE;
+ } else
+ pict_type = AV_PICTURE_TYPE_P;
+
+ if (headerPtr->pic_type == EB_AV1_NON_REF_PICTURE)
+ pkt->flags |= AV_PKT_FLAG_DISPOSABLE;
+
+ if (headerPtr->flags & EB_BUFFERFLAG_EOS)
+ svt_enc->eos_flag = EOS_RECEIVED;
+
+ ff_side_data_set_encoder_stats(pkt, headerPtr->qp * FF_QP2LAMBDA, NULL, 0, pict_type);
+
+ svt_av1_enc_release_out_buffer(&headerPtr);
+
+ return ret;
+}
+
+static av_cold int eb_enc_close(AVCodecContext *avctx)
+{
+ SvtContext *svt_enc = avctx->priv_data;
+
+ if (svt_enc) {
+ if (svt_enc->svt_handle) {
+ svt_av1_enc_deinit(svt_enc->svt_handle);
+ svt_av1_enc_deinit_handle(svt_enc->svt_handle);
+ }
+
+ free_buffer(svt_enc);
+ svt_enc = NULL;
+ }
+ return 0;
+}
+
+#define OFFSET(x) offsetof(SvtContext, x)
+#define VE AV_OPT_FLAG_VIDEO_PARAM | AV_OPT_FLAG_ENCODING_PARAM
+static const AVOption options[] = {
+ { "hielevel", "Hierarchical prediction levels setting", OFFSET(hierarchical_level),
+ AV_OPT_TYPE_INT, { .i64 = 4 }, 3, 4, VE , "hielevel"},
+ { "3level", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = 3 }, INT_MIN, INT_MAX, VE, "hielevel" },
+ { "4level", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = 4 }, INT_MIN, INT_MAX, VE, "hielevel" },
+
+ { "la_depth", "Look ahead distance [0, 120]", OFFSET(la_depth),
+ AV_OPT_TYPE_INT, { .i64 = -1 }, -1, 120, VE },
+
+ { "preset", "Encoding preset [0, 8]",
+ OFFSET(enc_mode), AV_OPT_TYPE_INT, { .i64 = MAX_ENC_PRESET }, 0, MAX_ENC_PRESET, VE },
+
+ { "profile", "Set profile restrictions", OFFSET(profile), AV_OPT_TYPE_INT, { .i64 = MAIN_PROFILE}, MAIN_PROFILE, PROFESSIONAL_PROFILE, VE, "profile" },
+ { "main" , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = MAIN_PROFILE}, INT_MIN, INT_MAX, VE, "profile" },
+ { "high" , NULL, 0, AV_OPT_TYPE_CONST, { .i64 = HIGH_PROFILE}, INT_MIN, INT_MAX, VE, "profile" },
+ { "professional", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = PROFESSIONAL_PROFILE }, INT_MIN, INT_MAX, VE, "profile" },
+
+ { "tier", "Set tier (general_tier_flag)", OFFSET(tier),
+ AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 1, VE, "tier" },
+ { "main", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = 0 }, 0, 0, VE, "tier" },
+ { "high", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = 1 }, 0, 0, VE, "tier" },
+
+ { "level", "Set level (level_idc)", OFFSET(level),
+ AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 0x1f, VE, "level" },
+
+#define LEVEL(name, value) name, NULL, 0, AV_OPT_TYPE_CONST, \
+ { .i64 = value }, 0, 0, VE, "level"
+ { LEVEL("2.0", 20) },
+ { LEVEL("2.1", 21) },
+ { LEVEL("2.2", 22) },
+ { LEVEL("2.3", 23) },
+ { LEVEL("3.0", 30) },
+ { LEVEL("3.1", 31) },
+ { LEVEL("3.2", 32) },
+ { LEVEL("3.3", 33) },
+ { LEVEL("4.0", 40) },
+ { LEVEL("4.1", 41) },
+ { LEVEL("4.2", 42) },
+ { LEVEL("4.3", 43) },
+ { LEVEL("5.0", 50) },
+ { LEVEL("5.1", 51) },
+ { LEVEL("5.2", 52) },
+ { LEVEL("5.3", 53) },
+ { LEVEL("6.0", 60) },
+ { LEVEL("6.1", 61) },
+ { LEVEL("6.2", 62) },
+ { LEVEL("6.3", 63) },
+ { LEVEL("7.0", 70) },
+ { LEVEL("7.1", 71) },
+ { LEVEL("7.2", 72) },
+ { LEVEL("7.3", 73) },
+#undef LEVEL
+
+ { "rc", "Bit rate control mode", OFFSET(rc_mode),
+ AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 3, VE , "rc"},
+ { "cqp", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = 0 }, INT_MIN, INT_MAX, VE, "rc" },
+ { "vbr", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = 2 }, INT_MIN, INT_MAX, VE, "rc" },
+ { "cvbr", NULL, 0, AV_OPT_TYPE_CONST,{ .i64 = 3 }, INT_MIN, INT_MAX, VE, "rc" },
+
+ { "qp", "QP value for intra frames", OFFSET(qp),
+ AV_OPT_TYPE_INT, { .i64 = 50 }, 0, 63, VE },
+
+ { "sc_detection", "Scene change detection", OFFSET(scd),
+ AV_OPT_TYPE_BOOL, { .i64 = 0 }, 0, 1, VE },
+
+
+
+
+ { "forced-idr", "If forcing keyframes, force them as IDR frames.", OFFSET(forced_idr),
+ AV_OPT_TYPE_BOOL, { .i64 = 0 }, 0, 1, VE },
+
+ {NULL},
+};
+
+static const AVClass class = {
+ .class_name = "libsvt_av1",
+ .item_name = av_default_item_name,
+ .option = options,
+ .version = LIBAVUTIL_VERSION_INT,
+};
+
+static const AVCodecDefault eb_enc_defaults[] = {
+ { "b", "7M" },
+ { "g", "-2" },
+ { "qmin", "0" },
+ { "qmax", "63" },
+ { NULL },
+};
+
+AVCodec ff_libsvt_av1_encoder = {
+ .name = "libsvt_av1",
+ .long_name = NULL_IF_CONFIG_SMALL("SVT-AV1(Scalable Video Technology for AV1) encoder"),
+ .priv_data_size = sizeof(SvtContext),
+ .type = AVMEDIA_TYPE_VIDEO,
+ .id = AV_CODEC_ID_AV1,
+ .init = eb_enc_init,
+ .send_frame = eb_send_frame,
+ .receive_packet = eb_receive_packet,
+ .close = eb_enc_close,
+ .capabilities = AV_CODEC_CAP_DELAY | AV_CODEC_CAP_AUTO_THREADS,
+ .pix_fmts = (const enum AVPixelFormat[]){ AV_PIX_FMT_YUV420P,
+ AV_PIX_FMT_YUV420P10,
+ AV_PIX_FMT_NONE },
+ .priv_class = &class,
+ .defaults = eb_enc_defaults,
+ .caps_internal = FF_CODEC_CAP_INIT_CLEANUP,
+ .wrapper_name = "libsvt_av1",
+};
diff --git a/libavcodec/libx264.c b/libavcodec/libx264.c
index a08fe0c..fc9f94f 100644
--- a/libavcodec/libx264.c
+++ b/libavcodec/libx264.c
@@ -207,24 +207,76 @@ static void reconfig_encoder(AVCodecContext *ctx, const AVFrame *frame)
x264_encoder_reconfig(x4->enc, &x4->params);
}
- if (x4->params.rc.i_vbv_buffer_size != ctx->rc_buffer_size / 1000 ||
- x4->params.rc.i_vbv_max_bitrate != ctx->rc_max_rate / 1000) {
- x4->params.rc.i_vbv_buffer_size = ctx->rc_buffer_size / 1000;
- x4->params.rc.i_vbv_max_bitrate = ctx->rc_max_rate / 1000;
- x264_encoder_reconfig(x4->enc, &x4->params);
+ if (frame->perceptual_score == -1) {
+ if (x4->params.rc.i_vbv_buffer_size != ctx->rc_buffer_size / 1000 ||
+ x4->params.rc.i_vbv_max_bitrate != ctx->rc_max_rate / 1000) {
+ x4->params.rc.i_vbv_buffer_size = ctx->rc_buffer_size / 1000;
+ x4->params.rc.i_vbv_max_bitrate = ctx->rc_max_rate / 1000;
+ x264_encoder_reconfig(x4->enc, &x4->params);
+ }
}
if (x4->params.rc.i_rc_method == X264_RC_ABR &&
- x4->params.rc.i_bitrate != ctx->bit_rate / 1000) {
- x4->params.rc.i_bitrate = ctx->bit_rate / 1000;
- x264_encoder_reconfig(x4->enc, &x4->params);
+ (frame->perceptual_score > -1 ||
+ x4->params.rc.i_bitrate != ctx->bit_rate / 1000)) {
+ if (frame->perceptual_score > -1) {
+ int bitrate = 0;
+ /* set ABR bitrate value from perceptual score */
+ /* decrease compression by raising the avg bitrate up to N times */
+ bitrate = (ctx->bit_rate / 1000) + ((frame->perceptual_score * frame->perceptual_score_factor) * (ctx->bit_rate / 1000.0));
+ x4->params.rc.i_bitrate = bitrate;
+ x4->params.rc.i_vbv_max_bitrate = bitrate * 1.5;
+ x4->params.rc.i_vbv_buffer_size = bitrate * 1.5 * 1.5;
+ av_log(ctx, AV_LOG_DEBUG,
+ "Perceptual: [%0.2f] bitrate %d maxbitrate %d from %"PRIu64"\n",
+ frame->perceptual_score,
+ x4->params.rc.i_bitrate,
+ x4->params.rc.i_vbv_max_bitrate,
+ ctx->bit_rate / 1000);
+
+ /* tag this frame with this specific config */
+ x4->pic.param = &x4->params;
+ x264_encoder_reconfig(x4->enc, &x4->params);
+ } else {
+ x4->params.rc.i_bitrate = ctx->bit_rate / 1000;
+ x264_encoder_reconfig(x4->enc, &x4->params);
+ }
}
if (x4->crf >= 0 &&
x4->params.rc.i_rc_method == X264_RC_CRF &&
- x4->params.rc.f_rf_constant != x4->crf) {
- x4->params.rc.f_rf_constant = x4->crf;
- x264_encoder_reconfig(x4->enc, &x4->params);
+ (frame->perceptual_score > -1 ||
+ x4->params.rc.f_rf_constant != x4->crf)) {
+ if (frame->perceptual_score > -1) {
+ float crf_value = 0.0;
+
+ /* set crf value from perceptual score */
+ /* decrease compression by lowering the score by up to N CRF points */
+ crf_value = x4->crf - ((frame->perceptual_score * 100.0) / (frame->perceptual_score_factor * 2.0));
+ x4->params.rc.f_rf_constant = crf_value;
+
+ if (ctx->rc_max_rate) {
+ int bitrate = 0;
+ /* set ABR bitrate value from perceptual score */
+ /* decrease compression by raising the avg bitrate up to N times */
+ bitrate = (ctx->rc_max_rate / 1000) + ((frame->perceptual_score * frame->perceptual_score_factor) * (ctx->rc_max_rate / 1000.0));
+ x4->params.rc.i_vbv_max_bitrate = bitrate;
+ x4->params.rc.i_vbv_buffer_size = bitrate * 1.5 * 1.5;
+ }
+ av_log(ctx, AV_LOG_DEBUG,
+ "Perceptual: [%0.2f] crf: %0.2f bitrate %d maxbitrate %d from %"PRIu64"\n",
+ frame->perceptual_score,
+ x4->params.rc.f_rf_constant,
+ x4->params.rc.i_bitrate,
+ x4->params.rc.i_vbv_max_bitrate,
+ ctx->rc_max_rate / 1000);
+
+ /* tag this frame with this specific config */
+ x4->pic.param = &x4->params;
+ } else {
+ x4->params.rc.f_rf_constant = x4->crf;
+ x264_encoder_reconfig(x4->enc, &x4->params);
+ }
}
if (x4->params.rc.i_rc_method == X264_RC_CQP &&
diff --git a/libavfilter/Makefile b/libavfilter/Makefile
index e6cfcd9..f6fe208 100644
--- a/libavfilter/Makefile
+++ b/libavfilter/Makefile
@@ -337,6 +337,7 @@ OBJS-$(CONFIG_PERMS_FILTER) += f_perms.o
OBJS-$(CONFIG_PERSPECTIVE_FILTER) += vf_perspective.o
OBJS-$(CONFIG_PHASE_FILTER) += vf_phase.o
OBJS-$(CONFIG_PHOTOSENSITIVITY_FILTER) += vf_photosensitivity.o
+OBJS-$(CONFIG_PHQM_FILTER) += vf_phqm.o img_hash.o
OBJS-$(CONFIG_PIXDESCTEST_FILTER) += vf_pixdesctest.o
OBJS-$(CONFIG_PIXSCOPE_FILTER) += vf_datascope.o
OBJS-$(CONFIG_PP_FILTER) += vf_pp.o
@@ -513,6 +514,7 @@ SKIPHEADERS-$(CONFIG_LIBVIDSTAB) += vidstabutils.h
OBJS-$(CONFIG_SHARED) += log2_tab.o
+SKIPHEADERS-$(CONFIG_LIBOPENCV) += img_hash.h
SKIPHEADERS-$(CONFIG_QSVVPP) += qsvvpp.h
SKIPHEADERS-$(CONFIG_OPENCL) += opencl.h
SKIPHEADERS-$(CONFIG_VAAPI) += vaapi_vpp.h
diff --git a/libavfilter/allfilters.c b/libavfilter/allfilters.c
index 501e5d0..126ca66 100644
--- a/libavfilter/allfilters.c
+++ b/libavfilter/allfilters.c
@@ -325,6 +325,7 @@ extern AVFilter ff_vf_pixdesctest;
extern AVFilter ff_vf_pixscope;
extern AVFilter ff_vf_pp;
extern AVFilter ff_vf_pp7;
+extern AVFilter ff_vf_phqm;
extern AVFilter ff_vf_premultiply;
extern AVFilter ff_vf_prewitt;
extern AVFilter ff_vf_prewitt_opencl;
diff --git a/libavfilter/img_hash.cpp b/libavfilter/img_hash.cpp
new file mode 100644
index 0000000..4d5843d
--- /dev/null
+++ b/libavfilter/img_hash.cpp
@@ -0,0 +1,98 @@
+/*
+ * Copyright (c) 2019 Christopher Kennedy
+ *
+ * OpenCV img_hash
+ *
+ * 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
+ */
+
+#include <opencv2/core.hpp>
+#include <opencv2/core/ocl.hpp>
+#include <opencv2/highgui.hpp>
+#include <opencv2/img_hash.hpp>
+#include <opencv2/imgproc.hpp>
+
+#include <iostream>
+
+#include "img_hash.h"
+#include "libavutil/pixdesc.h"
+extern "C" {
+#include "avfilter.h"
+}
+
+// convert from avframe to iplimage format
+static int fill_iplimage_from_frame(IplImage *img, const AVFrame *frame, enum AVPixelFormat pixfmt)
+{
+ IplImage *tmpimg;
+ int depth = IPL_DEPTH_8U, channels_nb;
+
+ switch (pixfmt) {
+ case AV_PIX_FMT_GRAY8: channels_nb = 1; break;
+ case AV_PIX_FMT_BGRA: channels_nb = 4; break;
+ case AV_PIX_FMT_BGR24: channels_nb = 3; break;
+ default: return -1;
+ }
+
+ tmpimg = cvCreateImageHeader((CvSize){frame->width, frame->height}, depth, channels_nb);
+ *img = *tmpimg;
+ img->imageData = img->imageDataOrigin = (char *) frame->data[0];
+ img->dataOrder = IPL_DATA_ORDER_PIXEL;
+ img->origin = IPL_ORIGIN_TL;
+ img->widthStep = frame->linesize[0];
+
+ return 0;
+}
+
+// Get the score of two Video Frames by comparing the perceptual hashes and deriving a hamming distance
+// showing how similar they are or different. lower score is better for most algorithms
+extern "C" double getScore(const AVFrame *frame1, const AVFrame *frame2, enum AVPixelFormat pixfmt, int hash_type) {
+ cv::Ptr<cv::img_hash::ImgHashBase> algo;
+ IplImage ipl1, ipl2;
+ cv::Mat h1;
+ cv::Mat h2;
+ cv::Mat m1;
+ cv::Mat m2;
+
+ // Take FFmpeg video frame and convert into an IplImage for OpenCV
+ if (fill_iplimage_from_frame(&ipl1, frame1, pixfmt) != 0 ||
+ fill_iplimage_from_frame(&ipl2, frame2, pixfmt) != 0)
+ return DBL_MAX; // Return an invalid value if either fails
+
+ // Convert an IplImage to an Mat Image for OpenCV (newer format)
+ m1 = cv::cvarrToMat(&ipl1);
+ m2 = cv::cvarrToMat(&ipl2);
+
+ // substantiate the hash type algorithm
+ switch (hash_type) {
+ case PHASH: algo = cv::img_hash::PHash::create(); break;
+ case AVERAGE: algo = cv::img_hash::AverageHash::create(); break;
+ case MARRHILDRETH: algo = cv::img_hash::MarrHildrethHash::create(); break;
+ case RADIALVARIANCE: algo = cv::img_hash::RadialVarianceHash::create(); break;
+ // BlockMeanHash support mode 0 and mode 1, they associate to
+ // mode 1 and mode 2 of PHash library
+ case BLOCKMEAN1: algo = cv::img_hash::BlockMeanHash::create(0); break;
+ case BLOCKMEAN2: algo = cv::img_hash::BlockMeanHash::create(1); break;
+ case COLORMOMENT: algo = cv::img_hash::ColorMomentHash::create(); break;
+ }
+
+ // Compute the hash
+ algo->compute(m1, h1);
+ algo->compute(m2, h2);
+
+ // Compare the hashes and return the hamming distance
+ return algo->compare(h1, h2);
+}
diff --git a/libavfilter/img_hash.h b/libavfilter/img_hash.h
new file mode 100644
index 0000000..76f55c3
--- /dev/null
+++ b/libavfilter/img_hash.h
@@ -0,0 +1,46 @@
+/*
+ * Copyright (c) 2019 Christopher Kennedy
+ *
+ * PHQM Perceptual Hash Quality Metric
+ *
+ * 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
+ */
+
+#ifndef AVFILTER_IMG_HASH_H
+#define AVFILTER_IMG_HASH_H
+
+#include "avfilter.h"
+
+#if defined(__cplusplus)
+extern "C"
+{
+#endif
+
+#define AVERAGE 0
+#define BLOCKMEAN1 1
+#define BLOCKMEAN2 2
+#define COLORMOMENT 3
+#define MARRHILDRETH 4
+#define PHASH 5
+#define RADIALVARIANCE 6
+
+double getScore(const AVFrame *frame1, const AVFrame *frame2, enum AVPixelFormat pixfmt, int hash_type);
+#if defined(__cplusplus)
+}
+#endif
+
+#endif
diff --git a/libavfilter/vf_phqm.c b/libavfilter/vf_phqm.c
new file mode 100644
index 0000000..a7b73cf
--- /dev/null
+++ b/libavfilter/vf_phqm.c
@@ -0,0 +1,382 @@
+/*
+ * Copyright (c) 2019 Christopher Kennedy
+ *
+ * PHQM Perceptual Hash Quality Metric
+ *
+ * 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
+ * PHQM: Calculate the Image Hash Hamming Difference between two input videos.
+ */
+
+#include <float.h>
+#include "libavutil/avstring.h"
+#include "libavutil/opt.h"
+#include "libavutil/pixdesc.h"
+#include "avfilter.h"
+#include "drawutils.h"
+#include "formats.h"
+#include "framesync.h"
+#include "internal.h"
+#include "video.h"
+
+#include "img_hash.h"
+#include "scene_sad.h"
+
+typedef struct PHQMContext {
+ const AVClass *class;
+ FFFrameSync fs;
+ double shd, hd, min_hd, max_hd, smin_hd, smax_hd;
+ double hft, sft, phd, psad;
+ uint64_t nb_shd;
+ uint64_t nb_frames;
+ FILE *stats_file;
+ char *stats_file_str;
+ int hash_type;
+ ff_scene_sad_fn sad_ref; ///< Sum of the absolute difference function (scene detect only)
+ ff_scene_sad_fn sad_enc; ///< Sum of the absolute difference function (scene detect only)
+ double prev_mafd_ref; ///< previous MAFD (scene detect only)
+ double prev_mafd_enc; ///< previous MAFD (scene detect only)
+ AVFrame *prev_pic_ref; ///< ref previous frame (scene detect only)
+ AVFrame *prev_pic_enc; ///< enc previous frame (scene detect only)
+ double scd_thresh;
+ double scene_score_ref;
+ double scene_score_enc;
+ double prev_hamm_ref;
+ double prev_hamm_enc;
+} PHQMContext;
+
+#define OFFSET(x) offsetof(PHQMContext, x)
+#define FLAGS AV_OPT_FLAG_FILTERING_PARAM|AV_OPT_FLAG_VIDEO_PARAM
+
+static const AVOption phqm_options[] = {
+ { "stats_file", "Set file where to store per-frame difference information.", OFFSET(stats_file_str), AV_OPT_TYPE_STRING, {.str=NULL}, 0, 0, FLAGS },
+ { "f", "Set file where to store per-frame difference information.", OFFSET(stats_file_str), AV_OPT_TYPE_STRING, {.str=NULL}, 0, 0, FLAGS },
+ { "scd_thresh", "Scene Change Detection Threshold.", OFFSET(scd_thresh), AV_OPT_TYPE_DOUBLE, {.dbl=0.5}, 0, 1, FLAGS },
+ { "hash_type", "Type of Image Hash to use from OpenCV.", OFFSET(hash_type), AV_OPT_TYPE_INT, {.i64 = PHASH}, 0, 6, FLAGS, "hash_type" },
+ { "average", "Average Hash", 0, AV_OPT_TYPE_CONST, {.i64 = AVERAGE}, 0, 0, FLAGS, "hash_type" },
+ { "blockmean1", "Block Mean Hash 1", 0, AV_OPT_TYPE_CONST, {.i64 = BLOCKMEAN1}, 0, 0, FLAGS, "hash_type" },
+ { "blockmean2", "Block Mean Hash 2", 0, AV_OPT_TYPE_CONST, {.i64 = BLOCKMEAN2}, 0, 0, FLAGS, "hash_type" },
+ { "colormoment", "Color Moment Hash", 0, AV_OPT_TYPE_CONST, {.i64 = COLORMOMENT}, 0, 0, FLAGS, "hash_type" },
+ { "marrhildreth", "Marr Hildreth Hash", 0, AV_OPT_TYPE_CONST, {.i64 = MARRHILDRETH}, 0, 0, FLAGS, "hash_type" },
+ { "phash", "Perceptual Hash (PHash)", 0, AV_OPT_TYPE_CONST, {.i64 = PHASH}, 0, 0, FLAGS, "hash_type" },
+ { "radialvariance", "Radial Variance Hash", 0, AV_OPT_TYPE_CONST, {.i64 = RADIALVARIANCE}, 0, 0, FLAGS, "hash_type" },
+ { NULL }
+};
+
+FRAMESYNC_DEFINE_CLASS(phqm, PHQMContext, fs);
+
+static void set_meta(AVDictionary **metadata, const char *key, char comp, float d)
+{
+ char value[128];
+ snprintf(value, sizeof(value), "%0.2f", d);
+ if (comp) {
+ char key2[128];
+ snprintf(key2, sizeof(key2), "%s%c", key, comp);