* [FFmpeg-devel] [PATCH v2 1/5] ccfifo: Properly handle CEA-708 captions through framerate conversion
2023-04-07 21:58 [FFmpeg-devel] [PATCH v2 0/5] Add support for Closed Caption FIFO Devin Heitmueller
@ 2023-04-07 21:58 ` Devin Heitmueller
2023-04-14 13:22 ` James Almer
2023-04-07 21:58 ` [FFmpeg-devel] [PATCH v2 2/5] vf_fps: properly preserve CEA-708 captions Devin Heitmueller
` (4 subsequent siblings)
5 siblings, 1 reply; 13+ messages in thread
From: Devin Heitmueller @ 2023-04-07 21:58 UTC (permalink / raw)
To: ffmpeg-devel; +Cc: Devin Heitmueller
When transcoding video that contains 708 closed captions, the
caption data is tied to the frames as side data. Simply dropping
or adding frames to change the framerate will result in loss of
data, so the caption data needs to be preserved and reformatted.
For example, without this patch converting 720p59 to 1080i59
would result in loss of 50% of the caption bytes, resulting in
garbled 608 captions and 708 probably wouldn't render at all.
Further, the frames that are there will have an illegal
cc_count for the target framerate, so some decoders may ignore
the packets entirely.
Extract the 608 and 708 tuples and insert them onto queues. Then
after dropping/adding frames, re-write the tuples back into the
resulting frames at the appropriate rate given the target
framerate. This includes both having the correct cc_count as
well as clocking out the 608 pairs at the appropriate rate.
Signed-off-by: Devin Heitmueller <dheitmueller@ltnglobal.com>
---
libavfilter/Makefile | 1 +
libavfilter/ccfifo.c | 191 +++++++++++++++++++++++++++++++++++++++++++++++++++
libavfilter/ccfifo.h | 85 +++++++++++++++++++++++
3 files changed, 277 insertions(+)
create mode 100644 libavfilter/ccfifo.c
create mode 100644 libavfilter/ccfifo.h
diff --git a/libavfilter/Makefile b/libavfilter/Makefile
index 71e198b..628ade8 100644
--- a/libavfilter/Makefile
+++ b/libavfilter/Makefile
@@ -14,6 +14,7 @@ OBJS = allfilters.o \
buffersink.o \
buffersrc.o \
colorspace.o \
+ ccfifo.o \
drawutils.o \
fifo.o \
formats.o \
diff --git a/libavfilter/ccfifo.c b/libavfilter/ccfifo.c
new file mode 100644
index 0000000..5db4149
--- /dev/null
+++ b/libavfilter/ccfifo.c
@@ -0,0 +1,191 @@
+/*
+ * CEA-708 Closed Captioning FIFO
+ * Copyright (c) 2023 LTN Global Communications
+ *
+ * Author: Devin Heitmueller <dheitmueller@ltnglobal.com>
+ *
+ * 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 "ccfifo.h"
+
+struct AVCCFifo {
+ AVFifo *cc_608_fifo;
+ AVFifo *cc_708_fifo;
+ int expected_cc_count;
+ int expected_608;
+ int cc_detected;
+ void *log_ctx;
+};
+
+#define MAX_CC_ELEMENTS 128
+#define CC_BYTES_PER_ENTRY 3
+
+struct cc_lookup {
+ int num;
+ int den;
+ int cc_count;
+ int num_608;
+};
+
+const static struct cc_lookup cc_lookup_vals[] = {
+ { 15, 1, 40, 4 },
+ { 24, 1, 25, 3 },
+ { 24000, 1001, 25, 3 },
+ { 30, 1, 20, 2 },
+ { 30000, 1001, 20, 2},
+ { 60, 1, 10, 1 },
+ { 60000, 1001, 10, 1},
+};
+
+void av_ccfifo_freep(AVCCFifo **ccf)
+{
+ if (ccf && *ccf) {
+ AVCCFifo *tmp = *ccf;
+ if (tmp->cc_608_fifo)
+ av_fifo_freep2(&tmp->cc_608_fifo);
+ if (tmp->cc_708_fifo)
+ av_fifo_freep2(&tmp->cc_708_fifo);
+ av_freep(*ccf);
+ }
+}
+
+AVCCFifo *av_ccfifo_alloc(AVRational *framerate, void *log_ctx)
+{
+ AVCCFifo *ccf;
+ int i;
+
+ ccf = av_mallocz(sizeof(*ccf));
+ if (!ccf)
+ return NULL;
+
+ if (!(ccf->cc_708_fifo = av_fifo_alloc2(MAX_CC_ELEMENTS, CC_BYTES_PER_ENTRY, 0)))
+ goto error;
+
+ if (!(ccf->cc_608_fifo = av_fifo_alloc2(MAX_CC_ELEMENTS, CC_BYTES_PER_ENTRY, 0)))
+ goto error;
+
+ /* Based on the target FPS, figure out the expected cc_count and number of
+ 608 tuples per packet. See ANSI/CTA-708-E Sec 4.3.6.1. */
+ for (i = 0; i < (sizeof(cc_lookup_vals) / sizeof(struct cc_lookup)); i++) {
+ if (framerate->num == cc_lookup_vals[i].num &&
+ framerate->den == cc_lookup_vals[i].den) {
+ ccf->expected_cc_count = cc_lookup_vals[i].cc_count;
+ ccf->expected_608 = cc_lookup_vals[i].num_608;
+ break;
+ }
+ }
+
+ if (ccf->expected_608 == 0) {
+ av_log(ccf->log_ctx, AV_LOG_WARNING, "cc_fifo cannot transcode captions fps=%d/%d\n",
+ framerate->num, framerate->den);
+ return NULL;
+ }
+
+ return ccf;
+
+error:
+ av_ccfifo_freep(&ccf);
+ return NULL;
+}
+
+int av_ccfifo_inject(AVCCFifo *ccf, AVFrame *frame)
+{
+ AVFrameSideData *sd;
+ int cc_filled = 0;
+ int i;
+
+ if (!ccf)
+ return 0;
+
+ if (ccf->cc_detected == 0 || ccf->expected_cc_count == 0)
+ return 0;
+
+ sd = av_frame_new_side_data(frame, AV_FRAME_DATA_A53_CC,
+ ccf->expected_cc_count * CC_BYTES_PER_ENTRY);
+ if (!sd)
+ return 0;
+
+ for (i = 0; i < ccf->expected_608; i++) {
+ if (av_fifo_can_read(ccf->cc_608_fifo) >= CC_BYTES_PER_ENTRY) {
+ av_fifo_read(ccf->cc_608_fifo, &sd->data[cc_filled * CC_BYTES_PER_ENTRY],
+ CC_BYTES_PER_ENTRY);
+ cc_filled++;
+ } else {
+ break;
+ }
+ }
+
+ /* Insert any available data from the 708 FIFO */
+ while (cc_filled < ccf->expected_cc_count) {
+ if (av_fifo_can_read(ccf->cc_708_fifo) >= CC_BYTES_PER_ENTRY) {
+ av_fifo_read(ccf->cc_708_fifo, &sd->data[cc_filled * CC_BYTES_PER_ENTRY],
+ CC_BYTES_PER_ENTRY);
+ cc_filled++;
+ } else {
+ break;
+ }
+ }
+
+ /* Insert 708 padding into any remaining fields */
+ while (cc_filled < ccf->expected_cc_count) {
+ sd->data[cc_filled * CC_BYTES_PER_ENTRY] = 0xfa;
+ sd->data[cc_filled * CC_BYTES_PER_ENTRY + 1] = 0x00;
+ sd->data[cc_filled * CC_BYTES_PER_ENTRY + 2] = 0x00;
+ cc_filled++;
+ }
+
+ return 0;
+}
+
+int av_ccfifo_extract(AVCCFifo *ccf, AVFrame *frame)
+{
+ int i;
+
+ if (!ccf)
+ return 0;
+
+ /* Read the A53 side data, discard padding, and put 608/708 into
+ queues so we can ensure they get into the output frames at
+ the correct rate... */
+ if (ccf->expected_cc_count > 0) {
+ AVFrameSideData *side_data = av_frame_get_side_data(frame, AV_FRAME_DATA_A53_CC);
+ if (side_data) {
+ uint8_t *cc_bytes = side_data->data;
+ int cc_count = side_data->size / CC_BYTES_PER_ENTRY;
+ ccf->cc_detected = 1;
+
+ for (i = 0; i < cc_count; i++) {
+ /* See ANSI/CTA-708-E Sec 4.3, Table 3 */
+ uint8_t cc_valid = (cc_bytes[CC_BYTES_PER_ENTRY*i] & 0x04) >> 2;
+ uint8_t cc_type = cc_bytes[CC_BYTES_PER_ENTRY*i] & 0x03;
+ if (cc_type == 0x00 || cc_type == 0x01) {
+ av_fifo_write(ccf->cc_608_fifo, &cc_bytes[CC_BYTES_PER_ENTRY*i],
+ CC_BYTES_PER_ENTRY);
+ } else if (cc_valid && (cc_type == 0x02 || cc_type == 0x03)) {
+ av_fifo_write(ccf->cc_708_fifo, &cc_bytes[CC_BYTES_PER_ENTRY*i],
+ CC_BYTES_PER_ENTRY);
+ }
+ }
+
+ /* Remove the side data, as we will re-create it on the
+ output as needed */
+ av_frame_remove_side_data(frame, AV_FRAME_DATA_A53_CC);
+ }
+ }
+ return 0;
+}
diff --git a/libavfilter/ccfifo.h b/libavfilter/ccfifo.h
new file mode 100644
index 0000000..66403be
--- /dev/null
+++ b/libavfilter/ccfifo.h
@@ -0,0 +1,85 @@
+/*
+ * CEA-708 Closed Captioning FIFO
+ * Copyright (c) 2023 LTN Global Communications
+ *
+ * Author: Devin Heitmueller <dheitmueller@ltnglobal.com>
+ *
+ * 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
+ * CC FIFO Buffer
+ */
+
+#ifndef AVUTIL_CCFIFO_H
+#define AVUTIL_CCFIFO_H
+
+#include "libavutil/avutil.h"
+#include "libavutil/frame.h"
+#include "libavutil/fifo.h"
+
+typedef struct AVCCFifo AVCCFifo;
+
+/**
+ * Allocate an AVCCFifo.
+ *
+ * @param sample_fmt sample format
+ * @param channels number of channels
+ * @param nb_samples initial allocation size, in samples
+ * @return newly allocated AVCCFifo, or NULL on error
+ */
+AVCCFifo *av_ccfifo_alloc(AVRational *framerate, void *log_ctx);
+
+/**
+ * Free an AVCCFifo
+ *
+ * @param ccf Pointer to the pointer to the AVCCFifo which should be freed
+ * @note `*ptr = NULL` is safe and leads to no action.
+ */
+void av_ccfifo_freep(AVCCFifo **ccf);
+
+
+/**
+ * Read a frame into a CC Fifo
+ *
+ * Extract CC bytes from the AVFrame, insert them into our queue, and
+ * remove the side data from the AVFrame. The side data is removed
+ * as it will be re-inserted at the appropriate rate later in the
+ * filter.
+ *
+ * @param af AVCCFifo to write to
+ * @param frame AVFrame with the video frame to operate on
+ * @return Zero on success, or negative AVERROR
+ * code on failure.
+ */
+int av_ccfifo_extract(AVCCFifo *af, AVFrame *frame);
+
+/**
+ * Insert CC data from the FIFO into an AVFrame (as side data)
+ *
+ * Dequeue the appropriate number of CC tuples based on the
+ * frame rate, and insert them into the AVFrame
+ *
+ * @param af AVCCFifo to read from
+ * @param frame AVFrame with the video frame to operate on
+ * @return Zero on success, or negative AVERROR
+ * code on failure.
+ */
+int av_ccfifo_inject(AVCCFifo *af, AVFrame *frame);
+
+#endif /* AVUTIL_CCFIFO_H */
--
1.8.3.1
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
^ permalink raw reply [flat|nested] 13+ messages in thread* Re: [FFmpeg-devel] [PATCH v2 1/5] ccfifo: Properly handle CEA-708 captions through framerate conversion
2023-04-07 21:58 ` [FFmpeg-devel] [PATCH v2 1/5] ccfifo: Properly handle CEA-708 captions through framerate conversion Devin Heitmueller
@ 2023-04-14 13:22 ` James Almer
2023-04-14 15:19 ` Devin Heitmueller
0 siblings, 1 reply; 13+ messages in thread
From: James Almer @ 2023-04-14 13:22 UTC (permalink / raw)
To: ffmpeg-devel
On 4/7/2023 6:58 PM, Devin Heitmueller wrote:
> When transcoding video that contains 708 closed captions, the
> caption data is tied to the frames as side data. Simply dropping
> or adding frames to change the framerate will result in loss of
> data, so the caption data needs to be preserved and reformatted.
>
> For example, without this patch converting 720p59 to 1080i59
> would result in loss of 50% of the caption bytes, resulting in
> garbled 608 captions and 708 probably wouldn't render at all.
> Further, the frames that are there will have an illegal
> cc_count for the target framerate, so some decoders may ignore
> the packets entirely.
>
> Extract the 608 and 708 tuples and insert them onto queues. Then
> after dropping/adding frames, re-write the tuples back into the
> resulting frames at the appropriate rate given the target
> framerate. This includes both having the correct cc_count as
> well as clocking out the 608 pairs at the appropriate rate.
>
> Signed-off-by: Devin Heitmueller <dheitmueller@ltnglobal.com>
> ---
> libavfilter/Makefile | 1 +
> libavfilter/ccfifo.c | 191 +++++++++++++++++++++++++++++++++++++++++++++++++++
> libavfilter/ccfifo.h | 85 +++++++++++++++++++++++
> 3 files changed, 277 insertions(+)
> create mode 100644 libavfilter/ccfifo.c
> create mode 100644 libavfilter/ccfifo.h
>
> diff --git a/libavfilter/Makefile b/libavfilter/Makefile
> index 71e198b..628ade8 100644
> --- a/libavfilter/Makefile
> +++ b/libavfilter/Makefile
> @@ -14,6 +14,7 @@ OBJS = allfilters.o \
> buffersink.o \
> buffersrc.o \
> colorspace.o \
> + ccfifo.o \
> drawutils.o \
> fifo.o \
> formats.o \
> diff --git a/libavfilter/ccfifo.c b/libavfilter/ccfifo.c
> new file mode 100644
> index 0000000..5db4149
> --- /dev/null
> +++ b/libavfilter/ccfifo.c
> @@ -0,0 +1,191 @@
> +/*
> + * CEA-708 Closed Captioning FIFO
> + * Copyright (c) 2023 LTN Global Communications
> + *
> + * Author: Devin Heitmueller <dheitmueller@ltnglobal.com>
> + *
> + * 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 "ccfifo.h"
> +
> +struct AVCCFifo {
> + AVFifo *cc_608_fifo;
> + AVFifo *cc_708_fifo;
> + int expected_cc_count;
> + int expected_608;
> + int cc_detected;
> + void *log_ctx;
> +};
> +
> +#define MAX_CC_ELEMENTS 128
> +#define CC_BYTES_PER_ENTRY 3
> +
> +struct cc_lookup {
> + int num;
> + int den;
> + int cc_count;
> + int num_608;
> +};
> +
> +const static struct cc_lookup cc_lookup_vals[] = {
> + { 15, 1, 40, 4 },
> + { 24, 1, 25, 3 },
> + { 24000, 1001, 25, 3 },
> + { 30, 1, 20, 2 },
> + { 30000, 1001, 20, 2},
> + { 60, 1, 10, 1 },
> + { 60000, 1001, 10, 1},
> +};
> +
> +void av_ccfifo_freep(AVCCFifo **ccf)
> +{
> + if (ccf && *ccf) {
> + AVCCFifo *tmp = *ccf;
> + if (tmp->cc_608_fifo)
> + av_fifo_freep2(&tmp->cc_608_fifo);
> + if (tmp->cc_708_fifo)
> + av_fifo_freep2(&tmp->cc_708_fifo);
> + av_freep(*ccf);
> + }
> +}
> +
> +AVCCFifo *av_ccfifo_alloc(AVRational *framerate, void *log_ctx)
None of these functions should use the av_* prefix if they are meant to
be internal, otherwise the symbols will be exported. Use ff_* instead.
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
^ permalink raw reply [flat|nested] 13+ messages in thread* Re: [FFmpeg-devel] [PATCH v2 1/5] ccfifo: Properly handle CEA-708 captions through framerate conversion
2023-04-14 13:22 ` James Almer
@ 2023-04-14 15:19 ` Devin Heitmueller
2023-04-16 9:37 ` Thilo Borgmann
0 siblings, 1 reply; 13+ messages in thread
From: Devin Heitmueller @ 2023-04-14 15:19 UTC (permalink / raw)
To: FFmpeg development discussions and patches
On Fri, Apr 14, 2023 at 9:22 AM James Almer <jamrial@gmail.com> wrote:
> None of these functions should use the av_* prefix if they are meant to
> be internal, otherwise the symbols will be exported. Use ff_* instead.
Thanks for taking the time to review. I will rename the functions in
the next series I post to the ML.
Regards,
Devin
--
Devin Heitmueller, Senior Software Engineer
LTN Global Communications
o: +1 (301) 363-1001
w: https://ltnglobal.com e: devin.heitmueller@ltnglobal.com
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [FFmpeg-devel] [PATCH v2 1/5] ccfifo: Properly handle CEA-708 captions through framerate conversion
2023-04-14 15:19 ` Devin Heitmueller
@ 2023-04-16 9:37 ` Thilo Borgmann
2023-04-17 0:29 ` Devin Heitmueller
0 siblings, 1 reply; 13+ messages in thread
From: Thilo Borgmann @ 2023-04-16 9:37 UTC (permalink / raw)
To: FFmpeg development discussions and patches
Hi,
> Am 14.04.2023 um 08:20 schrieb Devin Heitmueller <devin.heitmueller@ltnglobal.com>:
>
> On Fri, Apr 14, 2023 at 9:22 AM James Almer <jamrial@gmail.com> wrote:
>> None of these functions should use the av_* prefix if they are meant to
>> be internal, otherwise the symbols will be exported. Use ff_* instead.
>
> Thanks for taking the time to review. I will rename the functions in
> the next series I post to the ML.
Has that „trivial memory leak“ been addressed? 0_o
-Thilo
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
^ permalink raw reply [flat|nested] 13+ messages in thread
* Re: [FFmpeg-devel] [PATCH v2 1/5] ccfifo: Properly handle CEA-708 captions through framerate conversion
2023-04-16 9:37 ` Thilo Borgmann
@ 2023-04-17 0:29 ` Devin Heitmueller
0 siblings, 0 replies; 13+ messages in thread
From: Devin Heitmueller @ 2023-04-17 0:29 UTC (permalink / raw)
To: FFmpeg development discussions and patches
On Sun, Apr 16, 2023 at 5:38 AM Thilo Borgmann <thilo.borgmann@mail.de> wrote:
> Has that „trivial memory leak“ been addressed? 0_o
>
> -Thilo
I was waiting to see if there was any other feedback. I didn't think
it was worthwhile to submit an updated patch series with a one-line
change.
That said, if there are no further comments then I will generate a new
series with the leak fix and functions renamed to ff_.
Devin
--
Devin Heitmueller, Senior Software Engineer
LTN Global Communications
o: +1 (301) 363-1001
w: https://ltnglobal.com e: devin.heitmueller@ltnglobal.com
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
^ permalink raw reply [flat|nested] 13+ messages in thread
* [FFmpeg-devel] [PATCH v2 2/5] vf_fps: properly preserve CEA-708 captions
2023-04-07 21:58 [FFmpeg-devel] [PATCH v2 0/5] Add support for Closed Caption FIFO Devin Heitmueller
2023-04-07 21:58 ` [FFmpeg-devel] [PATCH v2 1/5] ccfifo: Properly handle CEA-708 captions through framerate conversion Devin Heitmueller
@ 2023-04-07 21:58 ` Devin Heitmueller
2023-04-07 21:58 ` [FFmpeg-devel] [PATCH v2 3/5] yadif: Properly preserve CEA-708 closed captions Devin Heitmueller
` (3 subsequent siblings)
5 siblings, 0 replies; 13+ messages in thread
From: Devin Heitmueller @ 2023-04-07 21:58 UTC (permalink / raw)
To: ffmpeg-devel; +Cc: Devin Heitmueller
The existing implementation made an attempt to remove duplicate
captions if increasing the framerate, but made no attempt to
handle reducing the framerate, nor did it rewrite the caption
payloads to have the appropriate cc_count (e.g. the cc_count needs
to change from 20 to 10 when going from 1080i59 to 720p59 and
vice-versa).
Make use of the new ccfifo mechanism to ensure that caption data
is properly preserved.
Signed-off-by: Devin Heitmueller <dheitmueller@ltnglobal.com>
---
libavfilter/vf_fps.c | 9 ++++++++-
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/libavfilter/vf_fps.c b/libavfilter/vf_fps.c
index 051d278..3f18bfc 100644
--- a/libavfilter/vf_fps.c
+++ b/libavfilter/vf_fps.c
@@ -34,6 +34,7 @@
#include "libavutil/mathematics.h"
#include "libavutil/opt.h"
#include "avfilter.h"
+#include "ccfifo.h"
#include "filters.h"
#include "internal.h"
@@ -85,6 +86,7 @@ typedef struct FPSContext {
AVFrame *frames[2]; ///< buffered frames
int frames_count; ///< number of buffered frames
+ AVCCFifo *cc_fifo; ///< closed captions
int64_t next_pts; ///< pts of the next frame to output
@@ -165,6 +167,7 @@ static av_cold void uninit(AVFilterContext *ctx)
frame = shift_frame(ctx, s);
av_frame_free(&frame);
}
+ av_ccfifo_freep(&s->cc_fifo);
av_log(ctx, AV_LOG_VERBOSE, "%d frames in, %d frames out; %d frames dropped, "
"%d frames duplicated.\n", s->frames_in, s->frames_out, s->drop, s->dup);
@@ -210,6 +213,9 @@ static int config_props(AVFilterLink* outlink)
s->in_pts_off, s->out_pts_off, s->start_time);
}
+ if (!(s->cc_fifo = av_ccfifo_alloc(&outlink->frame_rate, ctx)))
+ av_log(ctx, AV_LOG_VERBOSE, "Failure to setup CC FIFO queue. Captions will be passed through\n");
+
av_log(ctx, AV_LOG_VERBOSE, "fps=%d/%d\n", outlink->frame_rate.num, outlink->frame_rate.den);
return 0;
@@ -242,6 +248,7 @@ static int read_frame(AVFilterContext *ctx, FPSContext *s, AVFilterLink *inlink,
av_log(ctx, AV_LOG_DEBUG, "Read frame with in pts %"PRId64", out pts %"PRId64"\n",
in_pts, frame->pts);
+ av_ccfifo_extract(s->cc_fifo, frame);
s->frames[s->frames_count++] = frame;
s->frames_in++;
@@ -289,7 +296,7 @@ static int write_frame(AVFilterContext *ctx, FPSContext *s, AVFilterLink *outlin
if (!frame)
return AVERROR(ENOMEM);
// Make sure Closed Captions will not be duplicated
- av_frame_remove_side_data(s->frames[0], AV_FRAME_DATA_A53_CC);
+ av_ccfifo_inject(s->cc_fifo, frame);
frame->pts = s->next_pts++;
frame->duration = 1;
--
1.8.3.1
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
^ permalink raw reply [flat|nested] 13+ messages in thread* [FFmpeg-devel] [PATCH v2 3/5] yadif: Properly preserve CEA-708 closed captions
2023-04-07 21:58 [FFmpeg-devel] [PATCH v2 0/5] Add support for Closed Caption FIFO Devin Heitmueller
2023-04-07 21:58 ` [FFmpeg-devel] [PATCH v2 1/5] ccfifo: Properly handle CEA-708 captions through framerate conversion Devin Heitmueller
2023-04-07 21:58 ` [FFmpeg-devel] [PATCH v2 2/5] vf_fps: properly preserve CEA-708 captions Devin Heitmueller
@ 2023-04-07 21:58 ` Devin Heitmueller
2023-04-07 21:58 ` [FFmpeg-devel] [PATCH v2 4/5] tinterlace: " Devin Heitmueller
` (2 subsequent siblings)
5 siblings, 0 replies; 13+ messages in thread
From: Devin Heitmueller @ 2023-04-07 21:58 UTC (permalink / raw)
To: ffmpeg-devel; +Cc: Devin Heitmueller
Various deinterlacing modes have the effect of doubling the
framerate, and we need to ensure that the caption data isn't
duplicated (or else you get double captions on-screen).
Use the new ccfifo mechanism for yadif (and yadif_cuda and bwdif
since they use the same yadif core) so that CEA-708 data is
properly preserved through this filter.
Signed-off-by: Devin Heitmueller <dheitmueller@ltnglobal.com>
---
libavfilter/vf_bwdif.c | 7 +++++++
libavfilter/vf_yadif.c | 6 ++++++
libavfilter/vf_yadif_cuda.c | 8 ++++++++
libavfilter/yadif.h | 2 ++
libavfilter/yadif_common.c | 5 +++++
5 files changed, 28 insertions(+)
diff --git a/libavfilter/vf_bwdif.c b/libavfilter/vf_bwdif.c
index 34e8c5e..c625b68 100644
--- a/libavfilter/vf_bwdif.c
+++ b/libavfilter/vf_bwdif.c
@@ -297,6 +297,7 @@ static av_cold void uninit(AVFilterContext *ctx)
av_frame_free(&yadif->prev);
av_frame_free(&yadif->cur );
av_frame_free(&yadif->next);
+ av_ccfifo_freep(&yadif->cc_fifo);
}
static const enum AVPixelFormat pix_fmts[] = {
@@ -332,6 +333,12 @@ static int config_props(AVFilterLink *link)
if(yadif->mode&1)
link->frame_rate = av_mul_q(link->src->inputs[0]->frame_rate, (AVRational){2,1});
+ else
+ link->frame_rate = ctx->inputs[0]->frame_rate;
+
+ if (!(yadif->cc_fifo = av_ccfifo_alloc(&link->frame_rate, ctx)))
+ av_log(ctx, AV_LOG_VERBOSE, "Failure to setup CC FIFO queue. Captions will be passed through\n");
+
if (link->w < 3 || link->h < 4) {
av_log(ctx, AV_LOG_ERROR, "Video of less than 3 columns or 4 lines is not supported\n");
diff --git a/libavfilter/vf_yadif.c b/libavfilter/vf_yadif.c
index 1be02de..b51d21f 100644
--- a/libavfilter/vf_yadif.c
+++ b/libavfilter/vf_yadif.c
@@ -261,6 +261,7 @@ static av_cold void uninit(AVFilterContext *ctx)
av_frame_free(&yadif->prev);
av_frame_free(&yadif->cur );
av_frame_free(&yadif->next);
+ av_ccfifo_freep(&yadif->cc_fifo);
}
static const enum AVPixelFormat pix_fmts[] = {
@@ -293,6 +294,11 @@ static int config_output(AVFilterLink *outlink)
if(s->mode & 1)
outlink->frame_rate = av_mul_q(ctx->inputs[0]->frame_rate,
(AVRational){2, 1});
+ else
+ outlink->frame_rate = ctx->inputs[0]->frame_rate;
+
+ if (!(s->cc_fifo = av_ccfifo_alloc(&outlink->frame_rate, ctx)))
+ av_log(ctx, AV_LOG_VERBOSE, "Failure to setup CC FIFO queue. Captions will be passed through\n");
if (outlink->w < 3 || outlink->h < 3) {
av_log(ctx, AV_LOG_ERROR, "Video of less than 3 columns or lines is not supported\n");
diff --git a/libavfilter/vf_yadif_cuda.c b/libavfilter/vf_yadif_cuda.c
index 685b8a2..d96ec39 100644
--- a/libavfilter/vf_yadif_cuda.c
+++ b/libavfilter/vf_yadif_cuda.c
@@ -206,6 +206,9 @@ static av_cold void deint_cuda_uninit(AVFilterContext *ctx)
av_frame_free(&y->cur);
av_frame_free(&y->next);
+ if (yadif->cc_fifo)
+ av_cc_fifo_free(yadif->cc_fifo);
+
av_buffer_unref(&s->device_ref);
s->hwctx = NULL;
av_buffer_unref(&s->input_frames_ref);
@@ -291,6 +294,11 @@ static int config_output(AVFilterLink *link)
if(y->mode & 1)
link->frame_rate = av_mul_q(ctx->inputs[0]->frame_rate,
(AVRational){2, 1});
+ else
+ outlink->frame_rate = ctx->inputs[0]->frame_rate;
+
+ if (!(s->cc_fifo = av_cc_fifo_alloc(&outlink->frame_rate, ctx)))
+ av_log(ctx, AV_LOG_VERBOSE, "Failure to setup CC FIFO queue. Captions will be passed through\n");
if (link->w < 3 || link->h < 3) {
av_log(ctx, AV_LOG_ERROR, "Video of less than 3 columns or lines is not supported\n");
diff --git a/libavfilter/yadif.h b/libavfilter/yadif.h
index c928911..1077576 100644
--- a/libavfilter/yadif.h
+++ b/libavfilter/yadif.h
@@ -22,6 +22,7 @@
#include "libavutil/opt.h"
#include "libavutil/pixdesc.h"
#include "avfilter.h"
+#include "ccfifo.h"
enum YADIFMode {
YADIF_MODE_SEND_FRAME = 0, ///< send 1 frame for each frame
@@ -76,6 +77,7 @@ typedef struct YADIFContext {
int eof;
uint8_t *temp_line;
int temp_line_size;
+ AVCCFifo *cc_fifo;
/*
* An algorithm that treats first and/or last fields in a sequence
diff --git a/libavfilter/yadif_common.c b/libavfilter/yadif_common.c
index a10cf7a..24cc72a 100644
--- a/libavfilter/yadif_common.c
+++ b/libavfilter/yadif_common.c
@@ -60,6 +60,8 @@ static int return_frame(AVFilterContext *ctx, int is_second)
yadif->out->pts = AV_NOPTS_VALUE;
}
}
+
+ av_ccfifo_inject(yadif->cc_fifo, yadif->out);
ret = ff_filter_frame(ctx->outputs[0], yadif->out);
yadif->frame_pending = (yadif->mode&1) && !is_second;
@@ -96,6 +98,8 @@ int ff_yadif_filter_frame(AVFilterLink *link, AVFrame *frame)
av_assert0(frame);
+ av_ccfifo_extract(yadif->cc_fifo, frame);
+
if (yadif->frame_pending)
return_frame(ctx, 1);
@@ -137,6 +141,7 @@ int ff_yadif_filter_frame(AVFilterLink *link, AVFrame *frame)
if (!yadif->out)
return AVERROR(ENOMEM);
+ av_ccfifo_inject(yadif->cc_fifo, yadif->out);
av_frame_free(&yadif->prev);
if (yadif->out->pts != AV_NOPTS_VALUE)
yadif->out->pts *= 2;
--
1.8.3.1
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
^ permalink raw reply [flat|nested] 13+ messages in thread* [FFmpeg-devel] [PATCH v2 4/5] tinterlace: Properly preserve CEA-708 closed captions
2023-04-07 21:58 [FFmpeg-devel] [PATCH v2 0/5] Add support for Closed Caption FIFO Devin Heitmueller
` (2 preceding siblings ...)
2023-04-07 21:58 ` [FFmpeg-devel] [PATCH v2 3/5] yadif: Properly preserve CEA-708 closed captions Devin Heitmueller
@ 2023-04-07 21:58 ` Devin Heitmueller
2023-04-10 1:28 ` Lance Wang
2023-04-07 21:58 ` [FFmpeg-devel] [PATCH v2 5/5] vf_ccrepack: Add new filter to repack CEA-708 side data Devin Heitmueller
2023-04-14 13:19 ` [FFmpeg-devel] [PATCH v2 0/5] Add support for Closed Caption FIFO Devin Heitmueller
5 siblings, 1 reply; 13+ messages in thread
From: Devin Heitmueller @ 2023-04-07 21:58 UTC (permalink / raw)
To: ffmpeg-devel; +Cc: Devin Heitmueller
Because the interlacing filter halves the effective framerate, we
need to ensure that no CEA-708 data is lost as frames are merged.
Make use of the new ccfifo mechanism to ensure that caption data
is properly preserved as frames pass through the filter.
Thanks to Thomas Mundt for review and noticing a couple of
missed codepaths for injection on output.
Signed-off-by: Devin Heitmueller <dheitmueller@ltnglobal.com>
---
libavfilter/tinterlace.h | 2 ++
libavfilter/vf_tinterlace.c | 8 ++++++++
2 files changed, 10 insertions(+)
diff --git a/libavfilter/tinterlace.h b/libavfilter/tinterlace.h
index 37b6c10..9f5ce7e 100644
--- a/libavfilter/tinterlace.h
+++ b/libavfilter/tinterlace.h
@@ -32,6 +32,7 @@
#include "libavutil/pixdesc.h"
#include "drawutils.h"
#include "avfilter.h"
+#include "ccfifo.h"
#define TINTERLACE_FLAG_VLPF 01
#define TINTERLACE_FLAG_CVLPF 2
@@ -77,6 +78,7 @@ typedef struct TInterlaceContext {
const AVPixFmtDescriptor *csp;
void (*lowpass_line)(uint8_t *dstp, ptrdiff_t width, const uint8_t *srcp,
ptrdiff_t mref, ptrdiff_t pref, int clip_max);
+ AVCCFifo *cc_fifo;
} TInterlaceContext;
void ff_tinterlace_init_x86(TInterlaceContext *interlace);
diff --git a/libavfilter/vf_tinterlace.c b/libavfilter/vf_tinterlace.c
index 0326292..226983c 100644
--- a/libavfilter/vf_tinterlace.c
+++ b/libavfilter/vf_tinterlace.c
@@ -291,6 +291,9 @@ static int config_out_props(AVFilterLink *outlink)
#endif
}
+ if (!(tinterlace->cc_fifo = av_ccfifo_alloc(&outlink->frame_rate, ctx)))
+ av_log(ctx, AV_LOG_VERBOSE, "Failure to setup CC FIFO queue. Captions will be passed through\n");
+
av_log(ctx, AV_LOG_VERBOSE, "mode:%d filter:%s h:%d -> h:%d\n", tinterlace->mode,
(tinterlace->flags & TINTERLACE_FLAG_CVLPF) ? "complex" :
(tinterlace->flags & TINTERLACE_FLAG_VLPF) ? "linear" : "off",
@@ -375,6 +378,8 @@ static int filter_frame(AVFilterLink *inlink, AVFrame *picref)
tinterlace->cur = tinterlace->next;
tinterlace->next = picref;
+ av_ccfifo_extract(tinterlace->cc_fifo, picref);
+
cur = tinterlace->cur;
next = tinterlace->next;
/* we need at least two frames */
@@ -451,6 +456,7 @@ static int filter_frame(AVFilterLink *inlink, AVFrame *picref)
if (!out)
return AVERROR(ENOMEM);
out->pts /= 2; // adjust pts to new framerate
+ av_ccfifo_inject(tinterlace->cc_fifo, out);
ret = ff_filter_frame(outlink, out);
return ret;
}
@@ -486,6 +492,7 @@ static int filter_frame(AVFilterLink *inlink, AVFrame *picref)
out->pts = cur->pts*2;
out->pts = av_rescale_q(out->pts, tinterlace->preout_time_base, outlink->time_base);
+ av_ccfifo_inject(tinterlace->cc_fifo, out);
if ((ret = ff_filter_frame(outlink, out)) < 0)
return ret;
@@ -521,6 +528,7 @@ static int filter_frame(AVFilterLink *inlink, AVFrame *picref)
out->pts = av_rescale_q(out->pts, tinterlace->preout_time_base, outlink->time_base);
out->duration = av_rescale_q(1, av_inv_q(outlink->frame_rate), outlink->time_base);
+ av_ccfifo_inject(tinterlace->cc_fifo, out);
ret = ff_filter_frame(outlink, out);
return ret;
--
1.8.3.1
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
^ permalink raw reply [flat|nested] 13+ messages in thread* Re: [FFmpeg-devel] [PATCH v2 4/5] tinterlace: Properly preserve CEA-708 closed captions
2023-04-07 21:58 ` [FFmpeg-devel] [PATCH v2 4/5] tinterlace: " Devin Heitmueller
@ 2023-04-10 1:28 ` Lance Wang
2023-04-10 12:45 ` Devin Heitmueller
0 siblings, 1 reply; 13+ messages in thread
From: Lance Wang @ 2023-04-10 1:28 UTC (permalink / raw)
To: FFmpeg development discussions and patches
On Sat, Apr 8, 2023 at 5:04 AM Devin Heitmueller <
devin.heitmueller@ltnglobal.com> wrote:
> Because the interlacing filter halves the effective framerate, we
> need to ensure that no CEA-708 data is lost as frames are merged.
>
> Make use of the new ccfifo mechanism to ensure that caption data
> is properly preserved as frames pass through the filter.
>
> Thanks to Thomas Mundt for review and noticing a couple of
> missed codepaths for injection on output.
>
> Signed-off-by: Devin Heitmueller <dheitmueller@ltnglobal.com>
> ---
> libavfilter/tinterlace.h | 2 ++
> libavfilter/vf_tinterlace.c | 8 ++++++++
> 2 files changed, 10 insertions(+)
>
> diff --git a/libavfilter/tinterlace.h b/libavfilter/tinterlace.h
> index 37b6c10..9f5ce7e 100644
> --- a/libavfilter/tinterlace.h
> +++ b/libavfilter/tinterlace.h
> @@ -32,6 +32,7 @@
> #include "libavutil/pixdesc.h"
> #include "drawutils.h"
> #include "avfilter.h"
> +#include "ccfifo.h"
>
> #define TINTERLACE_FLAG_VLPF 01
> #define TINTERLACE_FLAG_CVLPF 2
> @@ -77,6 +78,7 @@ typedef struct TInterlaceContext {
> const AVPixFmtDescriptor *csp;
> void (*lowpass_line)(uint8_t *dstp, ptrdiff_t width, const uint8_t
> *srcp,
> ptrdiff_t mref, ptrdiff_t pref, int clip_max);
> + AVCCFifo *cc_fifo;
> } TInterlaceContext;
>
> void ff_tinterlace_init_x86(TInterlaceContext *interlace);
> diff --git a/libavfilter/vf_tinterlace.c b/libavfilter/vf_tinterlace.c
> index 0326292..226983c 100644
> --- a/libavfilter/vf_tinterlace.c
> +++ b/libavfilter/vf_tinterlace.c
> @@ -291,6 +291,9 @@ static int config_out_props(AVFilterLink *outlink)
> #endif
> }
>
> + if (!(tinterlace->cc_fifo = av_ccfifo_alloc(&outlink->frame_rate,
> ctx)))
> + av_log(ctx, AV_LOG_VERBOSE, "Failure to setup CC FIFO queue.
> Captions will be passed through\n");
> +
>
free cc_fifo?
> av_log(ctx, AV_LOG_VERBOSE, "mode:%d filter:%s h:%d -> h:%d\n",
> tinterlace->mode,
> (tinterlace->flags & TINTERLACE_FLAG_CVLPF) ? "complex" :
> (tinterlace->flags & TINTERLACE_FLAG_VLPF) ? "linear" : "off",
> @@ -375,6 +378,8 @@ static int filter_frame(AVFilterLink *inlink, AVFrame
> *picref)
> tinterlace->cur = tinterlace->next;
> tinterlace->next = picref;
>
> + av_ccfifo_extract(tinterlace->cc_fifo, picref);
> +
> cur = tinterlace->cur;
> next = tinterlace->next;
> /* we need at least two frames */
> @@ -451,6 +456,7 @@ static int filter_frame(AVFilterLink *inlink, AVFrame
> *picref)
> if (!out)
> return AVERROR(ENOMEM);
> out->pts /= 2; // adjust pts to new framerate
> + av_ccfifo_inject(tinterlace->cc_fifo, out);
> ret = ff_filter_frame(outlink, out);
> return ret;
> }
> @@ -486,6 +492,7 @@ static int filter_frame(AVFilterLink *inlink, AVFrame
> *picref)
> out->pts = cur->pts*2;
>
> out->pts = av_rescale_q(out->pts, tinterlace->preout_time_base,
> outlink->time_base);
> + av_ccfifo_inject(tinterlace->cc_fifo, out);
> if ((ret = ff_filter_frame(outlink, out)) < 0)
> return ret;
>
> @@ -521,6 +528,7 @@ static int filter_frame(AVFilterLink *inlink, AVFrame
> *picref)
>
> out->pts = av_rescale_q(out->pts, tinterlace->preout_time_base,
> outlink->time_base);
> out->duration = av_rescale_q(1, av_inv_q(outlink->frame_rate),
> outlink->time_base);
> + av_ccfifo_inject(tinterlace->cc_fifo, out);
> ret = ff_filter_frame(outlink, out);
>
> return ret;
> --
> 1.8.3.1
>
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel@ffmpeg.org
> https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>
> To unsubscribe, visit link above, or email
> ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
>
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
^ permalink raw reply [flat|nested] 13+ messages in thread* Re: [FFmpeg-devel] [PATCH v2 4/5] tinterlace: Properly preserve CEA-708 closed captions
2023-04-10 1:28 ` Lance Wang
@ 2023-04-10 12:45 ` Devin Heitmueller
0 siblings, 0 replies; 13+ messages in thread
From: Devin Heitmueller @ 2023-04-10 12:45 UTC (permalink / raw)
To: FFmpeg development discussions and patches
Hi Lance,
Thanks for taking the time to review.
On Sun, Apr 9, 2023 at 9:28 PM Lance Wang <lance.lmwang@gmail.com> wrote:
>
> free cc_fifo?
You're right, I lost the cc_fifo() free call during some refactoring
before I submitted the patch series. I will include it in the next
version of the series.
Devin
--
Devin Heitmueller, Senior Software Engineer
LTN Global Communications
o: +1 (301) 363-1001
w: https://ltnglobal.com e: devin.heitmueller@ltnglobal.com
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
^ permalink raw reply [flat|nested] 13+ messages in thread
* [FFmpeg-devel] [PATCH v2 5/5] vf_ccrepack: Add new filter to repack CEA-708 side data
2023-04-07 21:58 [FFmpeg-devel] [PATCH v2 0/5] Add support for Closed Caption FIFO Devin Heitmueller
` (3 preceding siblings ...)
2023-04-07 21:58 ` [FFmpeg-devel] [PATCH v2 4/5] tinterlace: " Devin Heitmueller
@ 2023-04-07 21:58 ` Devin Heitmueller
2023-04-14 13:19 ` [FFmpeg-devel] [PATCH v2 0/5] Add support for Closed Caption FIFO Devin Heitmueller
5 siblings, 0 replies; 13+ messages in thread
From: Devin Heitmueller @ 2023-04-07 21:58 UTC (permalink / raw)
To: ffmpeg-devel; +Cc: Devin Heitmueller
THis filter can correct certain issues seen from upstream sources
where the cc_count is not properly set or the CEA-608 tuples are
not at the start of the payload as expected.
Make use of the ccfifo to extract and immediately repack the CEA-708
side data, thereby removing any extra padding and ensuring the 608
tuples are at the front of the payload.
Signed-off-by: Devin Heitmueller <dheitmueller@ltnglobal.com>
---
doc/filters.texi | 10 +++++
libavfilter/Makefile | 1 +
libavfilter/allfilters.c | 1 +
libavfilter/vf_ccrepack.c | 100 ++++++++++++++++++++++++++++++++++++++++++++++
4 files changed, 112 insertions(+)
create mode 100644 libavfilter/vf_ccrepack.c
diff --git a/doc/filters.texi b/doc/filters.texi
index 120fe66..e72e73d 100644
--- a/doc/filters.texi
+++ b/doc/filters.texi
@@ -8930,6 +8930,16 @@ Only deinterlace frames marked as interlaced.
The default value is @code{all}.
@end table
+@section ccrepack
+
+Repack CEA-708 closed captioning side data
+
+This filter fixes various issues seen with commerical encoders
+related to upstream malformed CEA-708 payloads, specifically
+incorrect number of tuples (wrong cc_count for the target FPS),
+and incorrect ordering of tuples (i.e. the CEA-608 tuples are not at
+the first entries in the payload).
+
@section cas
Apply Contrast Adaptive Sharpen filter to video stream.
diff --git a/libavfilter/Makefile b/libavfilter/Makefile
index 628ade8..3d0b3ad 100644
--- a/libavfilter/Makefile
+++ b/libavfilter/Makefile
@@ -212,6 +212,7 @@ OBJS-$(CONFIG_BOXBLUR_OPENCL_FILTER) += vf_avgblur_opencl.o opencl.o \
opencl/avgblur.o boxblur.o
OBJS-$(CONFIG_BWDIF_FILTER) += vf_bwdif.o yadif_common.o
OBJS-$(CONFIG_CAS_FILTER) += vf_cas.o
+OBJS-$(CONFIG_CCREPACK_FILTER) += vf_ccrepack.o
OBJS-$(CONFIG_CHROMABER_VULKAN_FILTER) += vf_chromaber_vulkan.o vulkan.o vulkan_filter.o
OBJS-$(CONFIG_CHROMAHOLD_FILTER) += vf_chromakey.o
OBJS-$(CONFIG_CHROMAKEY_FILTER) += vf_chromakey.o
diff --git a/libavfilter/allfilters.c b/libavfilter/allfilters.c
index d7db46c..b38550b 100644
--- a/libavfilter/allfilters.c
+++ b/libavfilter/allfilters.c
@@ -196,6 +196,7 @@ extern const AVFilter ff_vf_boxblur;
extern const AVFilter ff_vf_boxblur_opencl;
extern const AVFilter ff_vf_bwdif;
extern const AVFilter ff_vf_cas;
+extern const AVFilter ff_vf_ccrepack;
extern const AVFilter ff_vf_chromaber_vulkan;
extern const AVFilter ff_vf_chromahold;
extern const AVFilter ff_vf_chromakey;
diff --git a/libavfilter/vf_ccrepack.c b/libavfilter/vf_ccrepack.c
new file mode 100644
index 0000000..063c360
--- /dev/null
+++ b/libavfilter/vf_ccrepack.c
@@ -0,0 +1,100 @@
+/*
+ * CEA-708 Closed Caption Repacker
+ * Copyright (c) 2023 LTN Global Communications
+ *
+ * Author: Devin Heitmueller <dheitmueller@ltnglobal.com>
+ *
+ * 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
+ */
+
+/*
+ * Repackage CEA-708 arrays, which deals with incorrect cc_count for a given
+ * output framerate, and incorrect 708 padding.
+ *
+ * See CEA CEA-10-A "EIA-708-B Implementation Guidance", Section 26.5
+ * "Grouping DTVCC Data Within user_data() Structure"
+ */
+
+#include "avfilter.h"
+#include "internal.h"
+#include "ccfifo.h"
+#include "libavutil/opt.h"
+
+typedef struct CCRepackContext
+{
+ const AVClass *class;
+ AVCCFifo *cc_fifo;
+} CCRepackContext;
+
+static const AVOption ccrepack_options[] = {
+ { NULL }
+};
+
+AVFILTER_DEFINE_CLASS(ccrepack);
+
+static int config_input(AVFilterLink *link)
+{
+ CCRepackContext *ctx = link->dst->priv;
+
+ if (!(ctx->cc_fifo = av_ccfifo_alloc(&link->frame_rate, ctx)))
+ av_log(ctx, AV_LOG_VERBOSE, "Failure to setup CC FIFO queue. Captions will be passed through\n");
+
+ return 0;
+}
+
+static int filter_frame(AVFilterLink *inlink, AVFrame *frame)
+{
+ CCRepackContext *ctx = inlink->dst->priv;
+ AVFilterLink *outlink = inlink->dst->outputs[0];
+
+ av_ccfifo_extract(ctx->cc_fifo, frame);
+ av_ccfifo_inject(ctx->cc_fifo, frame);
+
+ return ff_filter_frame(outlink, frame);
+}
+
+static av_cold void uninit(AVFilterContext *ctx)
+{
+ CCRepackContext *s = ctx->priv;
+ av_ccfifo_freep(&s->cc_fifo);
+}
+
+static const AVFilterPad avfilter_vf_ccrepack_inputs[] = {
+ {
+ .name = "default",
+ .type = AVMEDIA_TYPE_VIDEO,
+ .filter_frame = filter_frame,
+ .config_props = config_input,
+ },
+};
+
+static const AVFilterPad avfilter_vf_ccrepack_outputs[] = {
+ {
+ .name = "default",
+ .type = AVMEDIA_TYPE_VIDEO,
+ },
+};
+
+AVFilter ff_vf_ccrepack = {
+ .name = "ccrepack",
+ .description = NULL_IF_CONFIG_SMALL("Repack CEA-708 closed caption metadata"),
+ .uninit = uninit,
+ .priv_size = sizeof(CCRepackContext),
+ .priv_class = &ccrepack_class,
+ FILTER_INPUTS(avfilter_vf_ccrepack_inputs),
+ FILTER_OUTPUTS(avfilter_vf_ccrepack_outputs),
+};
--
1.8.3.1
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
^ permalink raw reply [flat|nested] 13+ messages in thread* Re: [FFmpeg-devel] [PATCH v2 0/5] Add support for Closed Caption FIFO
2023-04-07 21:58 [FFmpeg-devel] [PATCH v2 0/5] Add support for Closed Caption FIFO Devin Heitmueller
` (4 preceding siblings ...)
2023-04-07 21:58 ` [FFmpeg-devel] [PATCH v2 5/5] vf_ccrepack: Add new filter to repack CEA-708 side data Devin Heitmueller
@ 2023-04-14 13:19 ` Devin Heitmueller
5 siblings, 0 replies; 13+ messages in thread
From: Devin Heitmueller @ 2023-04-14 13:19 UTC (permalink / raw)
To: ffmpeg-devel; +Cc: Devin Heitmueller
On Fri, Apr 7, 2023 at 5:03 PM Devin Heitmueller
<devin.heitmueller@ltnglobal.com> wrote:
>
> This patch series address all known issues reported to date, and in
> particular moves the API into libavfilter where it can remain
> private (at least for now).
>
> Comments and/or further review is welcomed.
>
> Devin Heitmueller (5):
> ccfifo: Properly handle CEA-708 captions through framerate conversion
> vf_fps: properly preserve CEA-708 captions
> yadif: Properly preserve CEA-708 closed captions
> tinterlace: Properly preserve CEA-708 closed captions
> vf_ccrepack: Add new filter to repack CEA-708 side data
>
> doc/filters.texi | 10 +++
> libavfilter/Makefile | 2 +
> libavfilter/allfilters.c | 1 +
> libavfilter/ccfifo.c | 191 ++++++++++++++++++++++++++++++++++++++++++++
> libavfilter/ccfifo.h | 85 ++++++++++++++++++++
> libavfilter/tinterlace.h | 2 +
> libavfilter/vf_bwdif.c | 7 ++
> libavfilter/vf_ccrepack.c | 100 +++++++++++++++++++++++
> libavfilter/vf_fps.c | 9 ++-
> libavfilter/vf_tinterlace.c | 8 ++
> libavfilter/vf_yadif.c | 6 ++
> libavfilter/vf_yadif_cuda.c | 8 ++
> libavfilter/yadif.h | 2 +
> libavfilter/yadif_common.c | 5 ++
> 14 files changed, 435 insertions(+), 1 deletion(-)
> create mode 100644 libavfilter/ccfifo.c
> create mode 100644 libavfilter/ccfifo.h
> create mode 100644 libavfilter/vf_ccrepack.c
>
> --
> 1.8.3.1
>
Does anyone have any further comments on this patch series? I moved
the API to be internal to libavfilter as requested, and there's one
trivial memory leak, but I don't think it's gotten any substantive
review.
I would like to see it merged, so I welcome further comments/feedback.
Thanks,
Devin
--
Devin Heitmueller, Senior Software Engineer
LTN Global Communications
o: +1 (301) 363-1001
w: https://ltnglobal.com e: devin.heitmueller@ltnglobal.com
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-request@ffmpeg.org with subject "unsubscribe".
^ permalink raw reply [flat|nested] 13+ messages in thread