FFMPEG C++ Non monotonically increasing dts to muxer - c++

I have two inputs:
Stream with audio and video
Stream with only video
I want to generate an ouput stream mixing Audio from the first stream and Video from the second stream.
But when i generate the output it says i am non monotonically increasing dts to muxer. I think im messing up with the dts.
Please help!!
Here is my code:
#include <iostream>
//Linux...
#ifdef __cplusplus
extern "C"
{
#endif
#include <libavformat/avformat.h>
#include <libavutil/mathematics.h>
#include <libavutil/time.h>
#ifdef __cplusplus
}
#endif
using namespace std;
int main(int argc, char* argv[])
{
AVOutputFormat* ofmt = NULL;
//Input AVFormatContext and Output AVFormatContext
AVFormatContext *audio_video_ifmt_ctx = NULL,
* only_video_ifmt_ctx = NULL,
* ofmt_ctx = NULL;
AVPacket audio_video_pkt, only_video_pkt;
const char *audio_video_in_filename, * only_video_in_filename, * out_filename;
int ret, i;
int audio_video_audioindex = -1;
int audio_video_videoindex = -1;
int only_video_videoindex = -1;
int audio_video_dataindex = -1;
int only_video_dataindex = -1;
int audio_video_frame_index = 0, only_video_frame_index = 0;
audio_video_in_filename = "rtmp://localhost/live/STREAM_NAME.flv";
only_video_in_filename = "rtmp://localhost/live/STREAM_NAME2.flv";
out_filename = "rtmp://localhost/live/STREAM_NAME3.flv";
//Network
avformat_network_init();
//Input ONLY VIDEO
if ((ret = avformat_open_input(&only_video_ifmt_ctx, only_video_in_filename, 0, 0)) < 0) {
printf("Could not open audio and video input file.");
return 0;
}
if ((ret = avformat_find_stream_info(only_video_ifmt_ctx, 0)) < 0) {
printf("Failed to retrieve audio and video input stream information");
return 0;
}
//Input AUDIO AND VIDEO
if ((ret = avformat_open_input(&audio_video_ifmt_ctx, audio_video_in_filename, 0, 0)) < 0) {
printf("Could not open only video input file.");
return 0;
}
if ((ret = avformat_find_stream_info(audio_video_ifmt_ctx, 0)) < 0) {
printf("Failed to retrieve only video input stream information");
return 0;
}
//Search audio_video stream index
for (i = 0; i < audio_video_ifmt_ctx->nb_streams; i++) {
if (audio_video_ifmt_ctx->streams[i]->codecpar->codec_type == AVMEDIA_TYPE_VIDEO) {
audio_video_videoindex = i;
}
if (audio_video_ifmt_ctx->streams[i]->codecpar->codec_type == AVMEDIA_TYPE_AUDIO) {
audio_video_audioindex = i;
}
if (audio_video_ifmt_ctx->streams[i]->codecpar->codec_type == AVMEDIA_TYPE_DATA) {
audio_video_dataindex = i;
}
}
//Search only_video stream index
for (i = 0; i < only_video_ifmt_ctx->nb_streams; i++) {
if (only_video_ifmt_ctx->streams[i]->codecpar->codec_type == AVMEDIA_TYPE_VIDEO) {
only_video_videoindex = i;
}
else if(audio_video_ifmt_ctx->streams[i]->codecpar->codec_type == AVMEDIA_TYPE_DATA) {
only_video_dataindex = i;
}
}
av_dump_format(audio_video_ifmt_ctx, 0, audio_video_in_filename, 0);
av_dump_format(only_video_ifmt_ctx, 0, only_video_in_filename, 0);
//Allocate output stream context
avformat_alloc_output_context2(&ofmt_ctx, NULL, NULL, out_filename); //RTMP
if (!ofmt_ctx) {
printf("Could not create output context\n");
ret = AVERROR_UNKNOWN;
return 0;
}
ofmt = ofmt_ctx->oformat;
//Create Output Data and Audio Stream
for (i = 0; i < audio_video_ifmt_ctx->nb_streams; i++) {
AVStream* video_in_stream;
//Create output AVStream according to input AVStream
if (i == audio_video_dataindex) {
video_in_stream = audio_video_ifmt_ctx->streams[audio_video_dataindex];
}
else if (i == audio_video_audioindex) {
video_in_stream = audio_video_ifmt_ctx->streams[audio_video_audioindex];
}
else {
video_in_stream = NULL;
continue;
}
//AVStream *out_stream = avformat_new_stream(ofmt_ctx, in_stream->codec->codec);
AVCodec* codec = avcodec_find_decoder(video_in_stream->codecpar->codec_id);
AVStream* out_stream = avformat_new_stream(ofmt_ctx, codec);
if (!out_stream) {
printf("Failed allocating output stream\n");
ret = AVERROR_UNKNOWN;
return 0;
}
AVCodecContext* p_codec_ctx = avcodec_alloc_context3(codec);
ret = avcodec_parameters_to_context(p_codec_ctx, video_in_stream->codecpar);
//Copy the settings of AVCodecContext
if (ret < 0) {
printf("Failed to copy context from input to output stream codec context\n");
return 0;
}
p_codec_ctx->codec_tag = 0;
if (ofmt_ctx->oformat->flags & AVFMT_GLOBALHEADER)
p_codec_ctx->flags |= AV_CODEC_FLAG_GLOBAL_HEADER;
ret = avcodec_parameters_from_context(out_stream->codecpar, p_codec_ctx);
if (ret < 0) {
av_log(NULL, AV_LOG_ERROR, "eno:[%d] error to paramters codec paramter \n", ret);
}
}
//Create Output Video Stream
for (i = 0; i < only_video_ifmt_ctx->nb_streams; i++) {
AVStream* video_in_stream;
//Create output AVStream according to input AVStream
if (i == only_video_dataindex) {
continue;
video_in_stream = only_video_ifmt_ctx->streams[only_video_dataindex];
}
else if (i == only_video_videoindex) {
video_in_stream = only_video_ifmt_ctx->streams[only_video_videoindex];
}
else {
video_in_stream = NULL;
continue;
}
//AVStream *out_stream = avformat_new_stream(ofmt_ctx, in_stream->codec->codec);
AVCodec* codec = avcodec_find_decoder(video_in_stream->codecpar->codec_id);
AVStream* out_stream = avformat_new_stream(ofmt_ctx, codec);
if (!out_stream) {
printf("Failed allocating output stream\n");
ret = AVERROR_UNKNOWN;
return 0;
}
AVCodecContext* p_codec_ctx = avcodec_alloc_context3(codec);
ret = avcodec_parameters_to_context(p_codec_ctx, video_in_stream->codecpar);
//Copy the settings of AVCodecContext
if (ret < 0) {
printf("Failed to copy context from input to output stream codec context\n");
return 0;
}
p_codec_ctx->codec_tag = 0;
if (ofmt_ctx->oformat->flags & AVFMT_GLOBALHEADER)
p_codec_ctx->flags |= AV_CODEC_FLAG_GLOBAL_HEADER;
ret = avcodec_parameters_from_context(out_stream->codecpar, p_codec_ctx);
if (ret < 0) {
av_log(NULL, AV_LOG_ERROR, "eno:[%d] error to paramters codec paramter \n", ret);
}
}
//Dump Format------------------
av_dump_format(ofmt_ctx, 0, out_filename, 1);
//Open output URL
if (!(ofmt->flags & AVFMT_NOFILE)) {
ret = avio_open(&ofmt_ctx->pb, out_filename, AVIO_FLAG_WRITE);
if (ret < 0) {
printf("Could not open output URL '%s'", out_filename);
goto end;
}
}
//Write output file header
ret = avformat_write_header(ofmt_ctx, NULL);
if (ret < 0) {
printf("Error occurred when opening output URL\n");
goto end;
}
//Pull stream
while (1)
{
AVStream *audio_video_in_stream, * only_video_in_stream, * out_stream;
ret = av_read_frame(audio_video_ifmt_ctx, &audio_video_pkt);
ret = av_read_frame(only_video_ifmt_ctx, &only_video_pkt);
if (ret < 0)
break;
audio_video_in_stream = audio_video_ifmt_ctx->streams[audio_video_pkt.stream_index];
only_video_in_stream = only_video_ifmt_ctx->streams[only_video_pkt.stream_index];
out_stream = NULL;
///// TREAT AUDIO PACKETS /////
if (audio_video_pkt.stream_index == audio_video_dataindex) {
out_stream = ofmt_ctx->streams[0];
}
if (audio_video_pkt.stream_index == audio_video_audioindex) {
out_stream = ofmt_ctx->streams[1];
}
//Convert PTS/DTS
if (out_stream != NULL) {
audio_video_pkt.pts = av_rescale_q_rnd(audio_video_pkt.pts, audio_video_in_stream->time_base, out_stream->time_base, (AVRounding)(AV_ROUND_NEAR_INF | AV_ROUND_PASS_MINMAX));
audio_video_pkt.dts = av_rescale_q_rnd(audio_video_pkt.dts, audio_video_in_stream->time_base, out_stream->time_base, (AVRounding)(AV_ROUND_NEAR_INF | AV_ROUND_PASS_MINMAX));
audio_video_pkt.duration = av_rescale_q(audio_video_pkt.duration, audio_video_in_stream->time_base, out_stream->time_base);
audio_video_pkt.pos = -1;
}
//Writing audio stream
if (out_stream != NULL) {
ret = av_interleaved_write_frame(ofmt_ctx, &audio_video_pkt);
}
///// TREAT VIDEO PACKETS /////
if (only_video_pkt.stream_index == only_video_videoindex) {
out_stream = ofmt_ctx->streams[2];
}
//Convert PTS/DTS
if (out_stream != NULL) {
only_video_pkt.pts = av_rescale_q_rnd(only_video_pkt.pts, only_video_in_stream->time_base, out_stream->time_base, (AVRounding)(AV_ROUND_NEAR_INF | AV_ROUND_PASS_MINMAX));
only_video_pkt.dts = av_rescale_q_rnd(only_video_pkt.dts, only_video_in_stream->time_base, out_stream->time_base, (AVRounding)(AV_ROUND_NEAR_INF | AV_ROUND_PASS_MINMAX));
only_video_pkt.duration = av_rescale_q(only_video_pkt.duration, only_video_in_stream->time_base, out_stream->time_base);
only_video_pkt.pos = -1;
}
//Writing video
if (out_stream != NULL) {
ret = av_interleaved_write_frame(ofmt_ctx, &only_video_pkt);
}
//Desreferenciamos el paquete
av_packet_unref(&audio_video_pkt);
av_packet_unref(&only_video_pkt);
if (ret < 0) {
printf("Error muxing packet\n");
break;
}
}
//Write file trailer
av_write_trailer(ofmt_ctx);
end:
avformat_close_input(&audio_video_ifmt_ctx);
avformat_close_input(&only_video_ifmt_ctx);
/* close output */
if (ofmt_ctx && !(ofmt->flags & AVFMT_NOFILE))
avio_close(ofmt_ctx->pb);
avformat_free_context(ofmt_ctx);
if (ret < 0 && ret != AVERROR_EOF) {
printf("Error occurred.\n");
return -1;
}
return 0;
}
Here is the output:
Input #0, flv, from 'rtmp://localhost/live/STREAM_NAME.flv':
Metadata:
fileSize : 0
audiochannels : 2
2.1 : false
3.1 : false
4.0 : false
4.1 : false
5.1 : false
7.1 : false
encoder : obs-output module (libobs version 25.0.4)
Duration: 00:00:00.00, start: 2910.332000, bitrate: N/A
Stream #0:0: Data: none
Stream #0:1: Audio: aac (LC), 44100 Hz, stereo, fltp, 163 kb/s
Stream #0:2: Video: h264 (High), yuv420p(progressive), 1920x1080, 2560 kb/s, 23.98 fps, 23.98 tbr, 1k tbn, 47.95 tbc
Input #0, flv, from 'rtmp://localhost/live/STREAM_NAME2.flv':
Metadata:
encoder : Lavf58.65.101
Duration: 00:00:00.00, start: 2726.365000, bitrate: N/A
Stream #0:0: Data: none
Stream #0:1: Video: h264 (High), yuv420p(progressive), 1920x1080, 52 fps, 52 tbr, 1k tbn, 104 tbc
Output #0, flv, to 'rtmp://localhost/live/STREAM_NAME3.flv':
Stream #0:0: Data: none
Stream #0:1: Audio: aac (LC), 44100 Hz, stereo, fltp, 163 kb/s
Stream #0:2: Video: h264 (High), yuv420p(progressive), 1920x1080, q=2-31
[flv # 0000014F2CDB9500] Application provided invalid, non monotonically increasing dts to muxer in stream 1: 2910332 >= 2726346
Error muxing packet
[flv # 0000014F2CDB9500] Failed to update header with correct duration.
[flv # 0000014F2CDB9500] Failed to update header with correct filesize.
Error occurred.

Related

Adding unregistered SEI data to every frame (ffmpeg / C++ / Windows)

I am working with FFMPEG 5.2 using it with C++ in Visual Studio. What I require to do is to add a SEI Unregistered message (5) to every frame of a stream, for that, I am demuxing a MP4 container, then taking the video stream, decoding every packet to get a frame, then add SEI message to every frame, encoding and remuxing a new video stream (video only) and saving the new stream to a separate container.
To add the SEI data I use this specific code:
const char* sideDataMsg = "139FB1A9446A4DEC8CBF65B1E12D2CFDHola";;
size_t sideDataSize = sizeof(sideDataMsg);
AVBufferRef* sideDataBuffer = av_buffer_alloc(20);
sideDataBuffer->data = (uint8_t*)sideDataMsg;
AVFrameSideData* sideData = av_frame_new_side_data_from_buf(frame, AV_FRAME_DATA_SEI_UNREGISTERED, sideDataBuffer);
regarding the format of the sideDataMsg I have tried several apporaches including setting it like: "139FB1A9-446A-4DEC-8CBF65B1E12D2CFD+Hola!" which is indicated to be the required format in H.264 specs, however, even when in memory I see the SEI data is added to every frame as we observe as follows:
the resulting stream/container does not shows the expected data, below my entire code, this is mostly code taken/adapted from doc/examples folder of FFMPEG library.
BTW: I also tried setting AVCodecContext->export_side_data to different bit values (0 to FF) understanding that this can indicate the encoder to export the SEI data in every frame to be encoded but no luck.
I appreciate in advance any help from you!
// FfmpegTests.cpp : This file contains the 'main' function. Program execution begins and ends there.
//
#pragma warning(disable : 4996)
extern "C"
{
#include "libavformat/avformat.h"
#include "libavcodec/avcodec.h"
#include "libavfilter/avfilter.h"
#include "libavutil/opt.h"
#include "libavutil/avutil.h"
#include "libavutil/error.h"
#include "libavfilter/buffersrc.h"
#include "libavfilter/buffersink.h"
#include "libswscale/swscale.h"
}
#pragma comment(lib, "avcodec.lib")
#pragma comment(lib, "avformat.lib")
#pragma comment(lib, "avfilter.lib")
#pragma comment(lib, "avutil.lib")
#pragma comment(lib, "swscale.lib")
#include <cstdio>
#include <iostream>
#include <chrono>
#include <thread>
static AVFormatContext* fmt_ctx;
static AVCodecContext* dec_ctx;
AVFilterGraph* filter_graph;
AVFilterContext* buffersrc_ctx;
AVFilterContext* buffersink_ctx;
static int video_stream_index = -1;
const char* filter_descr = "scale=78:24,transpose=cclock";
static int64_t last_pts = AV_NOPTS_VALUE;
// FOR SEI NAL INSERTION
const AVOutputFormat* ofmt = NULL;
AVFormatContext* ofmt_ctx = NULL;
int stream_index = 0;
int* stream_mapping = NULL;
int stream_mapping_size = 0;
int FRAMES_COUNT = 0;
const AVCodec* codec_enc;
AVCodecContext* c = NULL;
static int open_input_file(const char* filename)
{
const AVCodec* dec;
int ret;
if ((ret = avformat_open_input(&fmt_ctx, filename, NULL, NULL)) < 0) {
av_log(NULL, AV_LOG_ERROR, "Cannot open input file\n");
return ret;
}
if ((ret = avformat_find_stream_info(fmt_ctx, NULL)) < 0) {
av_log(NULL, AV_LOG_ERROR, "Cannot find stream information\n");
return ret;
}
/* select the video stream */
ret = av_find_best_stream(fmt_ctx, AVMEDIA_TYPE_VIDEO, -1, -1, &dec, 0);
if (ret < 0) {
av_log(NULL, AV_LOG_ERROR, "Cannot find a video stream in the input file\n");
return ret;
}
video_stream_index = ret;
/* create decoding context */
dec_ctx = avcodec_alloc_context3(dec);
if (!dec_ctx)
return AVERROR(ENOMEM);
avcodec_parameters_to_context(dec_ctx, fmt_ctx->streams[video_stream_index]->codecpar);
FRAMES_COUNT = fmt_ctx->streams[video_stream_index]->nb_frames;
/* init the video decoder */
if ((ret = avcodec_open2(dec_ctx, dec, NULL)) < 0) {
av_log(NULL, AV_LOG_ERROR, "Cannot open video decoder\n");
return ret;
}
return 0;
}
static int init_filters(const char* filters_descr)
{
char args[512];
int ret = 0;
const AVFilter* buffersrc = avfilter_get_by_name("buffer");
const AVFilter* buffersink = avfilter_get_by_name("buffersink");
AVFilterInOut* outputs = avfilter_inout_alloc();
AVFilterInOut* inputs = avfilter_inout_alloc();
AVRational time_base = fmt_ctx->streams[video_stream_index]->time_base;
enum AVPixelFormat pix_fmts[] = { AV_PIX_FMT_GRAY8, AV_PIX_FMT_NONE };
filter_graph = avfilter_graph_alloc();
if (!outputs || !inputs || !filter_graph) {
ret = AVERROR(ENOMEM);
goto end;
}
/* buffer video source: the decoded frames from the decoder will be inserted here. */
snprintf(args, sizeof(args),
"video_size=%dx%d:pix_fmt=%d:time_base=%d/%d:pixel_aspect=%d/%d",
dec_ctx->width, dec_ctx->height, dec_ctx->pix_fmt,
time_base.num, time_base.den,
dec_ctx->sample_aspect_ratio.num, dec_ctx->sample_aspect_ratio.den);
ret = avfilter_graph_create_filter(&buffersrc_ctx, buffersrc, "in",
args, NULL, filter_graph);
if (ret < 0) {
av_log(NULL, AV_LOG_ERROR, "Cannot create buffer source\n");
goto end;
}
/* buffer video sink: to terminate the filter chain. */
ret = avfilter_graph_create_filter(&buffersink_ctx, buffersink, "out",
NULL, NULL, filter_graph);
if (ret < 0) {
av_log(NULL, AV_LOG_ERROR, "Cannot create buffer sink\n");
goto end;
}
ret = av_opt_set_int_list(buffersink_ctx, "pix_fmts", pix_fmts, AV_PIX_FMT_NONE, AV_OPT_SEARCH_CHILDREN);
if (ret < 0) {
av_log(NULL, AV_LOG_ERROR, "Cannot set output pixel format\n");
goto end;
}
outputs->name = av_strdup("in");
outputs->filter_ctx = buffersrc_ctx;
outputs->pad_idx = 0;
outputs->next = NULL;
inputs->name = av_strdup("out");
inputs->filter_ctx = buffersink_ctx;
inputs->pad_idx = 0;
inputs->next = NULL;
if ((ret = avfilter_graph_parse_ptr(filter_graph, filters_descr,
&inputs, &outputs, NULL)) < 0)
goto end;
if ((ret = avfilter_graph_config(filter_graph, NULL)) < 0)
goto end;
end:
avfilter_inout_free(&inputs);
avfilter_inout_free(&outputs);
return ret;
}
static void display_frame(const AVFrame* frame, AVRational time_base)
{
int x, y;
uint8_t* p0, * p;
int64_t delay;
if (frame->pts != AV_NOPTS_VALUE) {
if (last_pts != AV_NOPTS_VALUE) {
/* sleep roughly the right amount of time;
* usleep is in microseconds, just like AV_TIME_BASE. */
AVRational timeBaseQ;
timeBaseQ.num = 1;
timeBaseQ.den = AV_TIME_BASE;
delay = av_rescale_q(frame->pts - last_pts, time_base, timeBaseQ);
if (delay > 0 && delay < 1000000)
std::this_thread::sleep_for(std::chrono::microseconds(delay));
}
last_pts = frame->pts;
}
/* Trivial ASCII grayscale display. */
p0 = frame->data[0];
puts("\033c");
for (y = 0; y < frame->height; y++) {
p = p0;
for (x = 0; x < frame->width; x++)
putchar(" .-+#"[*(p++) / 52]);
putchar('\n');
p0 += frame->linesize[0];
}
fflush(stdout);
}
int save_frame_as_jpeg(AVCodecContext* pCodecCtx, AVFrame* pFrame, int FrameNo) {
int ret = 0;
const AVCodec* jpegCodec = avcodec_find_encoder(AV_CODEC_ID_MJPEG);
if (!jpegCodec) {
return -1;
}
AVCodecContext* jpegContext = avcodec_alloc_context3(jpegCodec);
if (!jpegContext) {
return -1;
}
jpegContext->pix_fmt = pCodecCtx->pix_fmt;
jpegContext->height = pFrame->height;
jpegContext->width = pFrame->width;
jpegContext->time_base = AVRational{ 1,10 };
jpegContext->strict_std_compliance = FF_COMPLIANCE_UNOFFICIAL;
ret = avcodec_open2(jpegContext, jpegCodec, NULL);
if (ret < 0) {
return ret;
}
FILE* JPEGFile;
char JPEGFName[256];
AVPacket packet;
packet.data = NULL;
packet.size = 0;
av_init_packet(&packet);
int gotFrame;
ret = avcodec_send_frame(jpegContext, pFrame);
if (ret < 0) {
return ret;
}
ret = avcodec_receive_packet(jpegContext, &packet);
if (ret < 0) {
return ret;
}
sprintf(JPEGFName, "c:\\folder\\dvr-%06d.jpg", FrameNo);
JPEGFile = fopen(JPEGFName, "wb");
fwrite(packet.data, 1, packet.size, JPEGFile);
fclose(JPEGFile);
av_packet_unref(&packet);
avcodec_close(jpegContext);
return 0;
}
int initialize_output_stream(AVFormatContext* input_fctx, const char* out_filename) {
int ret = 0;
avformat_alloc_output_context2(&ofmt_ctx, NULL, NULL, out_filename);
if (!ofmt_ctx) {
fprintf(stderr, "Could not create output context\n");
return -1;
}
stream_mapping_size = input_fctx->nb_streams;
stream_mapping = (int*)av_calloc(stream_mapping_size, sizeof(*stream_mapping));
if (!stream_mapping) {
ret = AVERROR(ENOMEM);
return -1;
}
for (int i = 0; i < input_fctx->nb_streams; i++) {
AVStream* out_stream;
AVStream* in_stream = input_fctx->streams[i];
AVCodecParameters* in_codecpar = in_stream->codecpar;
if (in_codecpar->codec_type != AVMEDIA_TYPE_AUDIO &&
in_codecpar->codec_type != AVMEDIA_TYPE_VIDEO &&
in_codecpar->codec_type != AVMEDIA_TYPE_SUBTITLE) {
stream_mapping[i] = -1;
continue;
}
stream_mapping[i] = stream_index++;
out_stream = avformat_new_stream(ofmt_ctx, NULL);
if (!out_stream) {
fprintf(stderr, "Failed allocating output stream\n");
ret = AVERROR_UNKNOWN;
return ret;
}
ret = avcodec_parameters_copy(out_stream->codecpar, in_codecpar);
if (ret < 0) {
fprintf(stderr, "Failed to copy codec parameters\n");
return -1;
}
out_stream->codecpar->codec_tag = 0;
}
ret = avio_open(&ofmt_ctx->pb, out_filename, AVIO_FLAG_WRITE);
if (ret < 0) {
fprintf(stderr, "Could not open output file '%s'", out_filename);
return -1;
}
ret = avformat_write_header(ofmt_ctx, NULL);
if (ret < 0) {
fprintf(stderr, "Error occurred when opening output file\n");
return -1;
}
// ENCODER
codec_enc = avcodec_find_encoder_by_name("libx264");
if (!codec_enc) {
fprintf(stderr, "Codec '%s' not found\n", "libx264");
return -1;
}
c = avcodec_alloc_context3(codec_enc);
if (!c) {
fprintf(stderr, "Could not allocate video codec context\n");
exit(1);
}
c->bit_rate = dec_ctx->bit_rate;
c->width = dec_ctx->width;
c->height = dec_ctx->height;
c->time_base = dec_ctx->time_base;
c->framerate = dec_ctx->framerate;
c->gop_size = dec_ctx->gop_size;
c->max_b_frames = dec_ctx->max_b_frames;
c->pix_fmt = dec_ctx->pix_fmt;
c->time_base = AVRational{ 1,1 };
c->export_side_data = 255;
if (codec_enc->id == AV_CODEC_ID_H264)
av_opt_set(c->priv_data, "preset", "slow", 0);
ret = avcodec_open2(c, codec_enc, NULL);
if (ret < 0) {
fprintf(stderr, "Could not open codec\n");
return ret;
}
}
int add_frame_output_stream(AVFrame* frame) {
int ret;
AVPacket* pkt;
pkt = av_packet_alloc();
ret = avcodec_send_frame(c, frame);
if (ret < 0) {
fprintf(stderr, "Error sending a frame for decoding\n");
return ret;
}
while (ret >= 0) {
ret = avcodec_receive_packet(c, pkt);
if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF)
return 0;
else if (ret < 0) {
fprintf(stderr, "Error during decoding\n");
return -1;
}
pkt->stream_index = stream_mapping[pkt->stream_index];
ret = av_interleaved_write_frame(ofmt_ctx, pkt);
av_packet_unref(pkt);
}
return 0;
}
int main(int argc, char** argv)
{
AVFrame* frame;
AVFrame* filt_frame;
AVPacket* packet;
int ret, count = 0;
// FOR SEI NAL INSERTION
const char* out_filename;
if (argc < 2) {
fprintf(stderr, "Usage: %s file\n", argv[0]);
exit(1);
}
frame = av_frame_alloc();
filt_frame = av_frame_alloc();
packet = av_packet_alloc();
if (!frame || !filt_frame || !packet) {
fprintf(stderr, "Could not allocate frame or packet\n");
exit(1);
}
if ((ret = open_input_file(argv[1])) < 0)
goto end;
if ((ret = init_filters(filter_descr)) < 0)
goto end;
out_filename = argv[2];
initialize_output_stream(fmt_ctx, out_filename);
while (count < FRAMES_COUNT)
{
if ((ret = av_read_frame(fmt_ctx, packet)) < 0)
break;
if (packet->stream_index == video_stream_index) {
ret = avcodec_send_packet(dec_ctx, packet);
if (ret < 0) {
av_log(NULL, AV_LOG_ERROR, "Error while sending a packet to the decoder\n");
break;
}
while (ret >= 0)
{
ret = avcodec_receive_frame(dec_ctx, frame);
if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF) {
break;
}
else if (ret < 0) {
av_log(NULL, AV_LOG_ERROR, "Error while receiving a frame from the decoder\n");
goto end;
}
frame->pts = frame->best_effort_timestamp;
/* push the decoded frame into the filtergraph */
if (av_buffersrc_add_frame_flags(buffersrc_ctx, frame, AV_BUFFERSRC_FLAG_KEEP_REF) < 0) {
av_log(NULL, AV_LOG_ERROR, "Error while feeding the filtergraph\n");
break;
}
/* pull filtered frames from the filtergraph */
while (1) {
ret = av_buffersink_get_frame(buffersink_ctx, filt_frame);
if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF)
break;
if (ret < 0)
goto end;
// display_frame(filt_frame, buffersink_ctx->inputs[0]->time_base);
av_frame_unref(filt_frame);
/* ret = save_frame_as_jpeg(dec_ctx, frame, dec_ctx->frame_number);
if (ret < 0)
goto end; */
//2. Add metadata to frames SEI
ret = av_frame_make_writable(frame);
if (ret < 0)
exit(1);
char sideDataSei[43] = "139FB1A9446A4DEC8CBF65B1E12D2CFDHola";
const char* sideDataMsg = "139FB1A9446A4DEC8CBF65B1E12D2CFDHola";
size_t sideDataSize = sizeof(sideDataMsg);
AVBufferRef* sideDataBuffer = av_buffer_alloc(20);
sideDataBuffer->data = (uint8_t*)sideDataMsg;
AVFrameSideData* sideData = av_frame_new_side_data_from_buf(frame, AV_FRAME_DATA_SEI_UNREGISTERED, sideDataBuffer);
ret = add_frame_output_stream(frame);
if (ret < 0)
goto end;
}
av_frame_unref(frame);
count++;
}
}
av_packet_unref(packet);
}
av_write_trailer(ofmt_ctx);
end:
avfilter_graph_free(&filter_graph);
avcodec_free_context(&dec_ctx);
avformat_close_input(&fmt_ctx);
av_frame_free(&frame);
av_frame_free(&filt_frame);
av_packet_free(&packet);
if (ret < 0 && ret != AVERROR_EOF) {
char errBuf[AV_ERROR_MAX_STRING_SIZE]{ 0 };
int res = av_strerror(ret, errBuf, AV_ERROR_MAX_STRING_SIZE);
fprintf(stderr, "Error: %s\n", errBuf);
exit(1);
}
exit(0);
}
Well, I came up with this solution from a friend, just had to add:
*av_opt_set_int(c->priv_data, "udu_sei", 1, 0);*
In the function initialize_output_stream after all parameters are set for AVCodecContext (c) that is being used for the output stream encoding.
Hope this helps someone!

H265 codec changes from hvc1 to hev1

I try to copy mp4 file with hevc video.
Source file have hvc1 codec.
Stream #0:0(und): Video: hevc (Main) (hvc1 / 0x31637668)
After copying the codec gets the value of hev1
Stream #0:0(und): Video: hevc (Main) (hev1 / 0x31766568)
How can i get hvc1?
I found that ffmpeg can do it with "-tag:v hvc1"
(https://stackoverflow.com/questions/32152090/encode-h265-to-hvc1-codec)
And how to set tag in c++ code (FFMpeg copy streams without transcode)
But it does not effect
My code:
AVOutputFormat *ofmt = NULL;
AVFormatContext *ifmt_ctx = NULL, *ofmt_ctx = NULL;
AVPacket pkt;
const char *in_filename, *out_filename;
int ret, i;
int stream_index = 0;
int *stream_mapping = NULL;
int stream_mapping_size = 0;
if (argc < 3) {
printf("usage: %s input output\n"
"API example program to remux a media file with libavformat and libavcodec.\n"
"The output format is guessed according to the file extension.\n"
"\n", argv[0]);
return 1;
}
in_filename = argv[1];
out_filename = argv[2];
av_register_all();
if ((ret = avformat_open_input(&ifmt_ctx, in_filename, 0, 0)) < 0) {
fprintf(stderr, "Could not open input file '%s'", in_filename);
goto end;
}
if ((ret = avformat_find_stream_info(ifmt_ctx, 0)) < 0) {
fprintf(stderr, "Failed to retrieve input stream information");
goto end;
}
av_dump_format(ifmt_ctx, 0, in_filename, 0);
avformat_alloc_output_context2(&ofmt_ctx, NULL, NULL, out_filename);
if (!ofmt_ctx) {
fprintf(stderr, "Could not create output context\n");
ret = AVERROR_UNKNOWN;
goto end;
}
stream_mapping_size = ifmt_ctx->nb_streams;
stream_mapping = (int*)av_mallocz_array(stream_mapping_size, sizeof(*stream_mapping));
if (!stream_mapping) {
ret = AVERROR(ENOMEM);
goto end;
}
ofmt = ofmt_ctx->oformat;
for (i = 0; i < (int)ifmt_ctx->nb_streams; i++) {
AVStream *out_stream;
AVStream *in_stream = ifmt_ctx->streams[i];
AVCodecParameters *in_codecpar = in_stream->codecpar;
AVCodec *decoder = avcodec_find_decoder(in_stream->codecpar->codec_id);
AVCodec *encoder = avcodec_find_encoder(in_stream->codecpar->codec_id);
if (in_codecpar->codec_type != AVMEDIA_TYPE_AUDIO &&
in_codecpar->codec_type != AVMEDIA_TYPE_VIDEO &&
in_codecpar->codec_type != AVMEDIA_TYPE_SUBTITLE) {
stream_mapping[i] = -1;
continue;
}
stream_mapping[i] = stream_index++;
out_stream = avformat_new_stream(ofmt_ctx, NULL);
if (!out_stream) {
fprintf(stderr, "Failed allocating output stream\n");
ret = AVERROR_UNKNOWN;
goto end;
}
ret = avcodec_parameters_copy(out_stream->codecpar, in_codecpar);
if (ret < 0) {
fprintf(stderr, "Failed to copy codec parameters\n");
goto end;
}
unsigned int tag = 0;
// for ffmpeg new api 3.x
AVCodecParameters *parameters = out_stream->codecpar;
if (av_codec_get_tag2(ofmt_ctx->oformat->codec_tag, encoder->id, &tag) == 0) {
av_log(NULL, AV_LOG_ERROR, "could not find codec tag for codec id %d, default to 0.\n", encoder->id);
}
parameters->codec_tag = tag;
out_stream->codec = avcodec_alloc_context3(encoder);
// more setting for stream->codec
avcodec_parameters_to_context(out_stream->codec, parameters);
}
av_dump_format(ofmt_ctx, 0, out_filename, 1);
if (!(ofmt->flags & AVFMT_NOFILE)) {
ret = avio_open(&ofmt_ctx->pb, out_filename, AVIO_FLAG_WRITE);
if (ret < 0) {
fprintf(stderr, "Could not open output file '%s'", out_filename);
goto end;
}
}
ret = avformat_write_header(ofmt_ctx, NULL);
if (ret < 0) {
fprintf(stderr, "Error occurred when opening output file\n");
goto end;
}
while (1) {
AVStream *in_stream, *out_stream;
ret = av_read_frame(ifmt_ctx, &pkt);
if (ret < 0)
break;
in_stream = ifmt_ctx->streams[pkt.stream_index];
if (pkt.stream_index >= stream_mapping_size ||
stream_mapping[pkt.stream_index] < 0) {
av_packet_unref(&pkt);
continue;
}
pkt.stream_index = stream_mapping[pkt.stream_index];
out_stream = ofmt_ctx->streams[pkt.stream_index];
log_packet(ifmt_ctx, &pkt, "in");
/* copy packet */
pkt.pts = av_rescale_q_rnd(pkt.pts, in_stream->time_base, out_stream->time_base, (AVRounding)(AV_ROUND_NEAR_INF | AV_ROUND_PASS_MINMAX));
pkt.dts = av_rescale_q_rnd(pkt.dts, in_stream->time_base, out_stream->time_base, (AVRounding)(AV_ROUND_NEAR_INF | AV_ROUND_PASS_MINMAX));
pkt.duration = av_rescale_q(pkt.duration, in_stream->time_base, out_stream->time_base);
pkt.pos = -1;
log_packet(ofmt_ctx, &pkt, "out");
ret = av_interleaved_write_frame(ofmt_ctx, &pkt);
if (ret < 0) {
fprintf(stderr, "Error muxing packet\n");
break;
}
av_packet_unref(&pkt);
}
av_write_trailer(ofmt_ctx);
end:
avformat_close_input(&ifmt_ctx);
/* close output */
if (ofmt_ctx && !(ofmt->flags & AVFMT_NOFILE))
avio_closep(&ofmt_ctx->pb);
avformat_free_context(ofmt_ctx);
av_freep(&stream_mapping);
if (ret < 0 && ret != AVERROR_EOF) {
fprintf(stderr, "Error occurred: %s\n", av_errTOstr(ret).c_str());
return 1;
}
After changes:
out_stream = avformat_new_stream(ofmt_ctx, NULL);
ret = avcodec_parameters_copy(out_stream->codecpar, in_codecpar);
AVCodecParameters *parameters = out_stream->codecpar;
unsigned int tag = 0;
av_codec_get_tag2(ofmt_ctx->oformat->codec_tag, encoder->id, &tag) == 0);
parameters->codec_tag = tag;
out_stream->codec = avcodec_alloc_context3(encoder);
avcodec_parameters_to_context(out_stream->codec, parameters);
if (in_codecpar->codec_type == AVMEDIA_TYPE_VIDEO) {
parameters->codec_tag = MKTAG('h', 'v', 'c', '1');
}
}
av_dump_format(ofmt_ctx, 0, out_filename, 1);
ret = avio_open(&ofmt_ctx->pb, out_filename, AVIO_FLAG_WRITE);
ret = avformat_write_header(ofmt_ctx, NULL);
if (ret < 0) {
fprintf(stderr, "Error write header: %s\n", av_errTOstr(ret). c_str());
goto end;
}

How to set start time of video saved from RTSP stream with FFMPEG

I use FFMPEG to record video from a RTSP stream. What my code does is get current day time, create a folder with this format year/month/day/hour/minute and save the video to that folder.
When a new minute arrive, I create the new folder base on the new minute and run the record again to the new folder.
Basically It works, but the next video start time is continue the end of previous video. For example:
video1: 00:00 -> 00:55
video2: 00:56 -> ...
I hope I can set for all videos start from 00:00. Can I do that?
Here my code
ffmpeg.h
class CtFfmpeg {
public:
CtFfmpeg();
~CtFfmpeg();
void init();
int getInput();
int getOutputName(const char *filename);
int release();
int ret;
AVFormatContext *ifmt_ctx, *ofmt_ctx;
AVStream *in_stream, *out_stream;
AVPacket pkt;
const char *in_filename;
char *out_filename;
private:
int setOutput(const char *outfilename);
AVOutputFormat *ofmt;
};
ffmpeg.cpp
#include "ctffmpeg.h"
CtFfmpeg::CtFfmpeg() {
in_filename = new char [1024];
out_filename = new char [1024];
}
CtFfmpeg::~CtFfmpeg() {
delete [] in_filename;
delete [] out_filename;
}
void CtFfmpeg::init() {
avcodec_register_all();
av_register_all();
avformat_network_init();
pkt = { 0 };
av_init_packet(&pkt);
ofmt = NULL;
ifmt_ctx = NULL;
ofmt_ctx = NULL;
return;
}
int CtFfmpeg::release() {
av_write_trailer(ofmt_ctx);
avcodec_close(out_stream->codec);
// avcodec_close(in_stream->codec);
// avformat_close_input(&ifmt_ctx);
/* close output */
if (!(ofmt->flags & AVFMT_NOFILE))
avio_close(ofmt_ctx->pb);
avformat_free_context(ofmt_ctx);
av_free_packet(&pkt);
if (ret < 0 && ret != AVERROR_EOF) {
fprintf(stderr, "Error occurred\n");
return 1;
}
}
int CtFfmpeg::getInput() {
if ((ret = avformat_open_input(&ifmt_ctx, in_filename, 0, 0)) < 0) {
fprintf(stderr, "Could not open input file '%s'", in_filename);
release();
}
if ((ret = avformat_find_stream_info(ifmt_ctx, 0)) < 0) {
fprintf(stderr, "Failed to retrieve input stream information");
release();
}
av_dump_format(ifmt_ctx, 0, in_filename, 0);
}
int CtFfmpeg::setOutput(const char *outfilename) {
avformat_alloc_output_context2(&ofmt_ctx, NULL, NULL, outfilename);
if (!ofmt_ctx) {
fprintf(stderr, "Could not create output context\n");
ret = AVERROR_UNKNOWN;
release();
}
ofmt = ofmt_ctx->oformat;
for (int i = 0; i < ifmt_ctx->nb_streams; i++) {
in_stream = ifmt_ctx->streams[i];
out_stream = avformat_new_stream(ofmt_ctx, in_stream->codec->codec);
if (!out_stream) {
fprintf(stderr, "Failed allocating output stream\n");
ret = AVERROR_UNKNOWN;
release();
}
ret = avcodec_copy_context(out_stream->codec, in_stream->codec);
if (ret < 0) {
fprintf(stderr, "Failed to copy context from input to output stream codec context\n");
release();
}
out_stream->codec->codec_tag = 0;
if (ofmt_ctx->oformat->flags & AVFMT_GLOBALHEADER)
out_stream->codec->flags |= CODEC_FLAG_GLOBAL_HEADER;
} // for
av_dump_format(ofmt_ctx, 0, outfilename, 1);
if (!(ofmt->flags & AVFMT_NOFILE)) {
ret = avio_open(&ofmt_ctx->pb, outfilename, AVIO_FLAG_WRITE);
if (ret < 0) {
fprintf(stderr, "Could not open output file '%s'", outfilename);
release();
}
}
ret = avformat_write_header(ofmt_ctx, NULL);
if (ret < 0) {
fprintf(stderr, "Error occurred when opening output file\n");
release();
}
}
int CtFfmpeg::getOutputName(const char *filename){
sprintf(out_filename,filename);
setOutput(out_filename);
}
main.cpp
#include "ctfolder.h"
#include "ctffmpeg.h"
CtFfmpeg * ff;
int main(int argc, char** argv) {
if (argc < 2) {
printf("usage: %s <RTSP link> \n", argv[0]);
return 1;
}
ff = new CtFfmpeg();
ff->in_filename = argv[1]; //RTSP input link
ff->init();
ff->getInput();
string filename;
videoRecorder obj;
int start, now;
start = obj.get_current_min();
if(obj.create_folder(0755))
cout << "Cannot create folder, maybe it already exists" << endl;
else
cout << "Create folder succesfully" << endl;
int skip = 0;
while(1){
filename = obj.update_filename();
ff->getOutputName(filename.c_str());
while((now = obj.get_current_min()) == start) {
ff->ret = av_read_frame(ff->ifmt_ctx, &(ff->pkt));
skip++;
if(skip==1)
continue;
if(skip>2)
skip=2;
if (ff->ret < 0)
continue;
ff->pkt.pts = av_rescale_q_rnd(ff->pkt.pts, ff->in_stream->time_base, ff->out_stream->time_base, (AVRounding)(AV_ROUND_NEAR_INF|AV_ROUND_PASS_MINMAX));
ff->pkt.dts = av_rescale_q_rnd(ff->pkt.dts, ff->in_stream->time_base, ff->out_stream->time_base, (AVRounding)(AV_ROUND_NEAR_INF|AV_ROUND_PASS_MINMAX));
ff->pkt.duration = av_rescale_q(ff->pkt.duration, ff->in_stream->time_base, ff->out_stream->time_base);
ff->pkt.pos = -1;
ff->ret = av_interleaved_write_frame(ff->ofmt_ctx, &(ff->pkt));
if (ff->ret < 0) {
fprintf(stderr, "Error muxing packet\n");
continue;
}
av_free_packet(&(ff->pkt));
}
ff->release();
cout << "New minute!" << endl;
if(obj.create_folder(0755))
cout << "Cannot create folder, something's wrong" << endl;
else
cout << "Create folder succesfully" << endl;
start = now;
}
return 0;
}
You need to shift your recording packet's pts to 0.
while(<some condition>)
{
//...
int64_t pts_offset = AV_NOPTS_VALUE ;
while((now = obj.get_current_min()) == start)
{
//...
ff.pkt.pts = ...
//...
if( pts_offset == AV_NOPTS_VALUE )
{
pts_offset = ff.pkt.pts ;
}
ff.pkt.pts -= pts_offset ;
// ...
}
}
I tried to build your code and add Alexander Chernin suggestion to it but I face to muxer error!
When you decrease recording packet's pts, it's value go lower than recording packet's dts. In avcodec.h, above declaration of pts I found this comment:
pts MUST be larger or equal to dts as presentation cannot happen before decompression.
I solved this error by decreasing recording packet's dts.
ff->pkt.pts = av_rescale_q_rnd(ff->pkt.pts, ff->in_stream->ff->out_stream->(AVRounding) (AV_ROUND_NEAR_INF | AV_ROUND_PASS_MINMAX));
if (pts_offset == AV_NOPTS_VALUE) {
pts_offset = ff->pkt.pts;
}
ff->pkt.pts -= pts_offset;
ff->pkt.dts -= pts_offset;
ff->pkt.dts = av_rescale_q_rnd(ff->pkt.dts, ff->in_stream->time_base,ff->out_stream->time_base,(AVRounding) (AV_ROUND_NEAR_INF | AV_ROUND_PASS_MINMAX));
ff->pkt.duration = av_rescale_q(ff->pkt.duration,ff->in_stream->time_base,ff->out_stream->time_base);
ff->pkt.pos = -1;

FFMPEG. Read frame, process it, put it to output video. Copy sound stream unchanged

I want to apply processing to a video clip with sound track, extract and process frame by frame and write result to output file. Number of frames, size of frame and speed remains unchanged in output clip. Also I want to keep the same audio track as I have in source.
I can read clip, decode frames and process then using opencv. Audio packets are also writes fine. I'm stuck on forming output video stream.
The minimal runnable code I have for now (sorry it not so short, but cant do it shorter):
extern "C" {
#include <libavutil/timestamp.h>
#include <libavformat/avformat.h>
#include "libavcodec/avcodec.h"
#include <libavutil/opt.h>
#include <libavdevice/avdevice.h>
#include <libswscale/swscale.h>
}
#include "opencv2/opencv.hpp"
#if LIBAVCODEC_VERSION_INT < AV_VERSION_INT(55,28,1)
#define av_frame_alloc avcodec_alloc_frame
#endif
using namespace std;
using namespace cv;
static void log_packet(const AVFormatContext *fmt_ctx, const AVPacket *pkt, const char *tag)
{
AVRational *time_base = &fmt_ctx->streams[pkt->stream_index]->time_base;
char buf1[AV_TS_MAX_STRING_SIZE] = { 0 };
av_ts_make_string(buf1, pkt->pts);
char buf2[AV_TS_MAX_STRING_SIZE] = { 0 };
av_ts_make_string(buf1, pkt->dts);
char buf3[AV_TS_MAX_STRING_SIZE] = { 0 };
av_ts_make_string(buf1, pkt->duration);
char buf4[AV_TS_MAX_STRING_SIZE] = { 0 };
av_ts_make_time_string(buf1, pkt->pts, time_base);
char buf5[AV_TS_MAX_STRING_SIZE] = { 0 };
av_ts_make_time_string(buf1, pkt->dts, time_base);
char buf6[AV_TS_MAX_STRING_SIZE] = { 0 };
av_ts_make_time_string(buf1, pkt->duration, time_base);
printf("pts:%s pts_time:%s dts:%s dts_time:%s duration:%s duration_time:%s stream_index:%d\n",
buf1, buf4,
buf2, buf5,
buf3, buf6,
pkt->stream_index);
}
int main(int argc, char **argv)
{
AVOutputFormat *ofmt = NULL;
AVFormatContext *ifmt_ctx = NULL, *ofmt_ctx = NULL;
AVPacket pkt;
AVFrame *pFrame = NULL;
AVFrame *pFrameRGB = NULL;
int frameFinished = 0;
pFrame = av_frame_alloc();
pFrameRGB = av_frame_alloc();
const char *in_filename, *out_filename;
int ret, i;
in_filename = "../../TestClips/Audio Video Sync Test.mp4";
out_filename = "out.mp4";
// Initialize FFMPEG
av_register_all();
// Get input file format context
if ((ret = avformat_open_input(&ifmt_ctx, in_filename, 0, 0)) < 0)
{
fprintf(stderr, "Could not open input file '%s'", in_filename);
goto end;
}
// Extract streams description
if ((ret = avformat_find_stream_info(ifmt_ctx, 0)) < 0)
{
fprintf(stderr, "Failed to retrieve input stream information");
goto end;
}
// Print detailed information about the input or output format,
// such as duration, bitrate, streams, container, programs, metadata, side data, codec and time base.
av_dump_format(ifmt_ctx, 0, in_filename, 0);
// Allocate an AVFormatContext for an output format.
avformat_alloc_output_context2(&ofmt_ctx, NULL, NULL, out_filename);
if (!ofmt_ctx)
{
fprintf(stderr, "Could not create output context\n");
ret = AVERROR_UNKNOWN;
goto end;
}
// The output container format.
ofmt = ofmt_ctx->oformat;
// Allocating output streams
for (i = 0; i < ifmt_ctx->nb_streams; i++)
{
AVStream *in_stream = ifmt_ctx->streams[i];
AVStream *out_stream = avformat_new_stream(ofmt_ctx, in_stream->codec->codec);
if (!out_stream)
{
fprintf(stderr, "Failed allocating output stream\n");
ret = AVERROR_UNKNOWN;
goto end;
}
ret = avcodec_copy_context(out_stream->codec, in_stream->codec);
if (ret < 0)
{
fprintf(stderr, "Failed to copy context from input to output stream codec context\n");
goto end;
}
out_stream->codec->codec_tag = 0;
if (ofmt_ctx->oformat->flags & AVFMT_GLOBALHEADER)
{
out_stream->codec->flags |= AV_CODEC_FLAG_GLOBAL_HEADER;
}
}
// Show output format info
av_dump_format(ofmt_ctx, 0, out_filename, 1);
// Open output file
if (!(ofmt->flags & AVFMT_NOFILE))
{
ret = avio_open(&ofmt_ctx->pb, out_filename, AVIO_FLAG_WRITE);
if (ret < 0)
{
fprintf(stderr, "Could not open output file '%s'", out_filename);
goto end;
}
}
// Write output file header
ret = avformat_write_header(ofmt_ctx, NULL);
if (ret < 0)
{
fprintf(stderr, "Error occurred when opening output file\n");
goto end;
}
// Search for input video codec info
AVCodec *in_codec = nullptr;
AVCodecContext* avctx = nullptr;
int video_stream_index = -1;
for (int i = 0; i < ifmt_ctx->nb_streams; i++)
{
if (ifmt_ctx->streams[i]->codec->coder_type == AVMEDIA_TYPE_VIDEO)
{
video_stream_index = i;
avctx = ifmt_ctx->streams[i]->codec;
in_codec = avcodec_find_decoder(avctx->codec_id);
if (!in_codec)
{
fprintf(stderr, "in codec not found\n");
exit(1);
}
break;
}
}
// Search for output video codec info
AVCodec *out_codec = nullptr;
AVCodecContext* o_avctx = nullptr;
int o_video_stream_index = -1;
for (int i = 0; i < ofmt_ctx->nb_streams; i++)
{
if (ofmt_ctx->streams[i]->codec->coder_type == AVMEDIA_TYPE_VIDEO)
{
o_video_stream_index = i;
o_avctx = ofmt_ctx->streams[i]->codec;
out_codec = avcodec_find_encoder(o_avctx->codec_id);
if (!out_codec)
{
fprintf(stderr, "out codec not found\n");
exit(1);
}
break;
}
}
// openCV pixel format
AVPixelFormat pFormat = AV_PIX_FMT_RGB24;
// Data size
int numBytes = avpicture_get_size(pFormat, avctx->width, avctx->height);
// allocate buffer
uint8_t *buffer = (uint8_t *)av_malloc(numBytes * sizeof(uint8_t));
// fill frame structure
avpicture_fill((AVPicture *)pFrameRGB, buffer, pFormat, avctx->width, avctx->height);
// frame area
int y_size = avctx->width * avctx->height;
// Open input codec
avcodec_open2(avctx, in_codec, NULL);
// Main loop
while (1)
{
AVStream *in_stream, *out_stream;
ret = av_read_frame(ifmt_ctx, &pkt);
if (ret < 0)
{
break;
}
in_stream = ifmt_ctx->streams[pkt.stream_index];
out_stream = ofmt_ctx->streams[pkt.stream_index];
log_packet(ifmt_ctx, &pkt, "in");
// copy packet
pkt.pts = av_rescale_q_rnd(pkt.pts, in_stream->time_base, out_stream->time_base, AVRounding(AV_ROUND_NEAR_INF | AV_ROUND_PASS_MINMAX));
pkt.dts = av_rescale_q_rnd(pkt.dts, in_stream->time_base, out_stream->time_base, AVRounding(AV_ROUND_NEAR_INF | AV_ROUND_PASS_MINMAX));
pkt.duration = av_rescale_q(pkt.duration, in_stream->time_base, out_stream->time_base);
pkt.pos = -1;
log_packet(ofmt_ctx, &pkt, "out");
if (pkt.stream_index == video_stream_index)
{
avcodec_decode_video2(avctx, pFrame, &frameFinished, &pkt);
if (frameFinished)
{
struct SwsContext *img_convert_ctx;
img_convert_ctx = sws_getCachedContext(NULL,
avctx->width,
avctx->height,
avctx->pix_fmt,
avctx->width,
avctx->height,
AV_PIX_FMT_BGR24,
SWS_BICUBIC,
NULL,
NULL,
NULL);
sws_scale(img_convert_ctx,
((AVPicture*)pFrame)->data,
((AVPicture*)pFrame)->linesize,
0,
avctx->height,
((AVPicture *)pFrameRGB)->data,
((AVPicture *)pFrameRGB)->linesize);
sws_freeContext(img_convert_ctx);
// Do some image processing
cv::Mat img(pFrame->height, pFrame->width, CV_8UC3, pFrameRGB->data[0],false);
cv::GaussianBlur(img,img,Size(5,5),3);
cv::imshow("Display", img);
cv::waitKey(5);
// --------------------------------
// Transform back to initial format
// --------------------------------
img_convert_ctx = sws_getCachedContext(NULL,
avctx->width,
avctx->height,
AV_PIX_FMT_BGR24,
avctx->width,
avctx->height,
avctx->pix_fmt,
SWS_BICUBIC,
NULL,
NULL,
NULL);
sws_scale(img_convert_ctx,
((AVPicture*)pFrameRGB)->data,
((AVPicture*)pFrameRGB)->linesize,
0,
avctx->height,
((AVPicture *)pFrame)->data,
((AVPicture *)pFrame)->linesize);
// --------------------------------------------
// Something must be here
// --------------------------------------------
//
// Write fideo frame (How to write frame to output stream ?)
//
// --------------------------------------------
sws_freeContext(img_convert_ctx);
}
}
else // write sound frame
{
ret = av_interleaved_write_frame(ofmt_ctx, &pkt);
}
if (ret < 0)
{
fprintf(stderr, "Error muxing packet\n");
break;
}
// Decrease packet ref counter
av_packet_unref(&pkt);
}
av_write_trailer(ofmt_ctx);
end:
avformat_close_input(&ifmt_ctx);
// close output
if (ofmt_ctx && !(ofmt->flags & AVFMT_NOFILE))
{
avio_closep(&ofmt_ctx->pb);
}
avformat_free_context(ofmt_ctx);
if (ret < 0 && ret != AVERROR_EOF)
{
char buf_err[AV_ERROR_MAX_STRING_SIZE] = { 0 };
av_make_error_string(buf_err, AV_ERROR_MAX_STRING_SIZE, ret);
fprintf(stderr, "Error occurred: %s\n", buf_err);
return 1;
}
avcodec_close(avctx);
av_free(pFrame);
av_free(pFrameRGB);
return 0;
}
Your original code segfaults in my case. Initializing the output codec context seems to fix it. The code below works for me but I didn't test the OpenCV stuff as I don't have the lib installed.
Get the codec context:
// Search for output video codec info
AVCodec *out_codec = NULL;
AVCodecContext* o_avctx = NULL;
int o_video_stream_index = -1;
for (int i = 0; i < ofmt_ctx->nb_streams; i++)
{
if (ofmt_ctx->streams[i]->codec->coder_type == AVMEDIA_TYPE_VIDEO)
{
o_video_stream_index = i;
out_codec = avcodec_find_encoder(ofmt_ctx->streams[i]->codec->codec_id);
o_avctx = avcodec_alloc_context3(out_codec);
o_avctx->height = avctx->height;
o_avctx->width = avctx->width;
o_avctx->sample_aspect_ratio = avctx->sample_aspect_ratio;
if (out_codec->pix_fmts)
o_avctx->pix_fmt = out_codec->pix_fmts[0];
else
o_avctx->pix_fmt = avctx->pix_fmt;
o_avctx->time_base = avctx->time_base;
avcodec_open2(o_avctx, out_codec, NULL);
}
}
Encode and write:
// Main loop
while (1)
{
...
if (pkt.stream_index == video_stream_index)
{
avcodec_decode_video2(avctx, pFrame, &frameFinished, &pkt);
if (frameFinished)
{
...
// --------------------------------------------
// Something must be here
// --------------------------------------------
int got_packet = 0;
AVPacket enc_pkt = { 0 };
av_init_packet(&enc_pkt);
avcodec_encode_video2(o_avctx, &enc_pkt, pFrame, &got_packet);
av_interleaved_write_frame(ofmt_ctx, &enc_pkt);
....
}
}
you should assign processed frame's packets information to your Original packets then pass it to av_interleaved_write_frame

Extracting the h264 part of a video file (demuxing)

I am trying to demux a video file into the video part (h264, mpeg4, h265, vp8, etc) and the audio part (mp3, aac, ac3, etc) and the subtitle part (srt) using ffmpeg in c++.
The audio part came out alright and played on all the media players I have, so also did the subtitle part. The video part however came out WITHOUT error and saved into a .h264 file but when I use ffprobe to check it or ffplay to play it, it always give the error "Invalid data found when processing input".
The code below
/* Separate a media file into audio, video and subtitle files (demuxing, complex) */
//TODO: mute error when subtitle is not present
#define __STDC_CONSTANT_MACROS
extern "C"
{
#include "libavformat/avformat.h"
}
int main()
{
//Input AVFormatContext and Output AVFormatContext
AVOutputFormat *ofmt_a = NULL, *ofmt_v = NULL, *ofmt_s = NULL;
AVFormatContext *ifmt_ctx = NULL, *ofmt_ctx_a = NULL, *ofmt_ctx_v = NULL, *ofmt_ctx_s = NULL;
AVPacket pkt;
int ret, i;
int videoindex=-1, audioindex=-1, srtindex=-1;
int frame_index=0;
//Input file URL
const char *in_filename = "sample.mp4";
//Output file URL
const char *out_filename_v = "sample.h264";
const char *out_filename_a = "sample.mp3";
const char *out_filename_s = "sample.srt";
av_register_all();
//Input
if ((ret = avformat_open_input(&ifmt_ctx, in_filename, 0, 0)) < 0) {
printf( "Could not open input file.");
goto end;
}
if ((ret = avformat_find_stream_info(ifmt_ctx, 0)) < 0) {
printf( "Failed to retrieve input stream information");
goto end;
}
//Output
avformat_alloc_output_context2(&ofmt_ctx_v, NULL, NULL, out_filename_v);
if (!ofmt_ctx_v) {
printf( "Could not create output context\n");
ret = AVERROR_UNKNOWN;
goto end;
}
ofmt_v = ofmt_ctx_v->oformat;
avformat_alloc_output_context2(&ofmt_ctx_a, NULL, NULL, out_filename_a);
if (!ofmt_ctx_a) {
printf( "Could not create output context\n");
ret = AVERROR_UNKNOWN;
goto end;
}
ofmt_a = ofmt_ctx_a->oformat;
avformat_alloc_output_context2(&ofmt_ctx_s, NULL, NULL, out_filename_s);
if (!ofmt_ctx_a) {
printf( "Could not create output context\n");
ret = AVERROR_UNKNOWN;
goto end;
}
ofmt_s = ofmt_ctx_s->oformat;
for (i = 0; i < ifmt_ctx->nb_streams; i++) {
//Create output AVStream according to input AVStream
AVFormatContext *ofmt_ctx;
AVStream *in_stream = ifmt_ctx->streams[i];
AVStream *out_stream = NULL;
if(ifmt_ctx->streams[i]->codec->codec_type==AVMEDIA_TYPE_VIDEO){
videoindex=i;
out_stream=avformat_new_stream(ofmt_ctx_v, in_stream->codec->codec);
ofmt_ctx=ofmt_ctx_v;
}
else if(ifmt_ctx->streams[i]->codec->codec_type==AVMEDIA_TYPE_AUDIO){
audioindex=i;
out_stream=avformat_new_stream(ofmt_ctx_a, in_stream->codec->codec);
ofmt_ctx=ofmt_ctx_a;
}
else if(ifmt_ctx->streams[i]->codec->codec_type==AVMEDIA_TYPE_SUBTITLE){
srtindex=i;
out_stream=avformat_new_stream(ofmt_ctx_s, in_stream->codec->codec);
ofmt_ctx=ofmt_ctx_s;
}
else{
break;
}
if (!out_stream) {
printf( "Failed allocating output stream\n");
ret = AVERROR_UNKNOWN;
goto end;
}
//Copy the settings of AVCodecContext
if (avcodec_copy_context(out_stream->codec, in_stream->codec) < 0) {
printf( "Failed to copy context from input to output stream codec context\n");
goto end;
}
out_stream->codec->codec_tag = 0;
if (ofmt_ctx->oformat->flags & AVFMT_GLOBALHEADER)
out_stream->codec->flags |= CODEC_FLAG_GLOBAL_HEADER;
}
//Dump Format------------------
printf("\n==============Input Video=============\n");
av_dump_format(ifmt_ctx, 0, in_filename, 0);
printf("\n==============Output Video============\n");
av_dump_format(ofmt_ctx_v, 0, out_filename_v, 1);
printf("\n==============Output Audio============\n");
av_dump_format(ofmt_ctx_a, 0, out_filename_a, 1);
/*printf("\n==============Output Subtitle============\n");
av_dump_format(ofmt_ctx_s, 0, out_filename_s, 1);*/
printf("\n======================================\n");
//Open output file
if (!(ofmt_v->flags & AVFMT_NOFILE)) {
if (avio_open(&ofmt_ctx_v->pb, out_filename_v, AVIO_FLAG_WRITE) < 0) {
printf( "Could not open output file '%s'", out_filename_v);
goto end;
}
}
if (!(ofmt_a->flags & AVFMT_NOFILE)) {
if (avio_open(&ofmt_ctx_a->pb, out_filename_a, AVIO_FLAG_WRITE) < 0) {
printf( "Could not open output file '%s'", out_filename_a);
goto end;
}
}
if (!(ofmt_a->flags & AVFMT_NOFILE)) {
if (avio_open(&ofmt_ctx_s->pb, out_filename_s, AVIO_FLAG_WRITE) < 0) {
printf( "Could not open output file '%s'", out_filename_s);
goto end;
}
}
//Write file header
if (avformat_write_header(ofmt_ctx_v, NULL) < 0) {
printf( "Error occurred when opening video output file\n");
goto end;
}
system("pause");
if (avformat_write_header(ofmt_ctx_a, NULL) < 0) {
printf( "Error occurred when opening audio output file\n");
goto end;
}
if (avformat_write_header(ofmt_ctx_s, NULL) < 0) {
printf( "Error occurred when opening audio output file\n");
goto end;
}
AVBitStreamFilterContext* h264bsfc = av_bitstream_filter_init("h264_mp4toannexb");
while (1) {
AVFormatContext *ofmt_ctx;
AVStream *in_stream, *out_stream;
//Get an AVPacket
if (av_read_frame(ifmt_ctx, &pkt) < 0)
break;
in_stream = ifmt_ctx->streams[pkt.stream_index];
if(pkt.stream_index==videoindex){
out_stream = ofmt_ctx_v->streams[0];
ofmt_ctx=ofmt_ctx_v;
printf("Write Video Packet. size:%d\tpts:%lld\n",pkt.size,pkt.pts);
av_bitstream_filter_filter(h264bsfc, in_stream->codec, NULL, &pkt.data, &pkt.size, pkt.data, pkt.size, 0);
}else if(pkt.stream_index==audioindex){
out_stream = ofmt_ctx_a->streams[0];
ofmt_ctx=ofmt_ctx_a;
printf("Write Audio Packet. size:%d\tpts:%lld\n",pkt.size,pkt.pts);
}
else if(pkt.stream_index==srtindex){
out_stream = ofmt_ctx_s->streams[0];
ofmt_ctx=ofmt_ctx_s;
printf("Write Subtitle Packet. size:%d\tpts:%lld\n",pkt.size,pkt.pts);
}
else{
continue;
}
//Convert PTS/DTS
pkt.pts = av_rescale_q_rnd(pkt.pts, in_stream->time_base, out_stream->time_base, (AVRounding)(AV_ROUND_NEAR_INF|AV_ROUND_PASS_MINMAX));
pkt.dts = av_rescale_q_rnd(pkt.dts, in_stream->time_base, out_stream->time_base, (AVRounding)(AV_ROUND_NEAR_INF|AV_ROUND_PASS_MINMAX));
pkt.duration = av_rescale_q(pkt.duration, in_stream->time_base, out_stream->time_base);
pkt.pos = -1;
pkt.stream_index=0;
//Write
if (av_interleaved_write_frame(ofmt_ctx, &pkt) < 0) {
printf( "Error muxing packet\n");
break;
}
//printf("Write %8d frames to output file\n",frame_index);
av_free_packet(&pkt);
frame_index++;
}
av_bitstream_filter_close(h264bsfc);
//Write file trailer
av_write_trailer(ofmt_ctx_a);
av_write_trailer(ofmt_ctx_v);
av_write_trailer(ofmt_ctx_s);
end:
avformat_close_input(&ifmt_ctx);
/* close output */
if (ofmt_ctx_a && !(ofmt_a->flags & AVFMT_NOFILE))
avio_close(ofmt_ctx_a->pb);
if (ofmt_ctx_v && !(ofmt_v->flags & AVFMT_NOFILE))
avio_close(ofmt_ctx_v->pb);
if (ofmt_ctx_s && !(ofmt_s->flags & AVFMT_NOFILE))
avio_close(ofmt_ctx_s->pb);
avformat_free_context(ofmt_ctx_a);
avformat_free_context(ofmt_ctx_v);
avformat_free_context(ofmt_ctx_s);
system("pause");
if (ret < 0 && ret != AVERROR_EOF) {
printf( "Error occurred.\n");
return -1;
}
return 0;
}
EDIT 1
Screen shot of resultant h264 file
EDIT 2
I think the "error" has to do with FFMPEG's "Using AVStream.codec.time_base as a timebase hint to the muxer is deprecated. Set AVStream.time_base instead" error.
I revert to an older version of FFMPEG and with the same code, the resultant h264 file was ok!
You need to convert an H.264 bitstream from length prefixed mode to start code prefixed mode.This is required by some streaming formats, typically the MPEG-2 transport stream format ("mpegts").
Take a look at https://www.ffmpeg.org/ffmpeg-bitstream-filters.html#h264_005fmp4toannexb
Look at lines from 402 to 424 and from 842 to 843.
https://www.ffmpeg.org/doxygen/0.7/crystalhd_8c-source.html
I used it, to extract h264 from mp4.
//Use this filter on your first h264 input AVPacket
AVFormatContext *ifmt_ctx = NULL;
//...
//... //init input
//...
AVPacket *firstPacket;
//...
//... //get packet from stream
//...
uint8_t *dummy_p;
int dummy_int;
AVBitStreamFilterContext *filter = v_bitstream_filter_init("h264_mp4toannexb");
if (!filter)
{
printf("Can't open filter\n");
exit(1);
}
ret = av_bitstream_filter_filter(filter, ifmt_ctx->streams[videoindex]->codec, NULL,
&dummy_p, &dummy_int,
firstPacket->data, firstPacket->size, 0);
if( ret < 0 )
{
printf("Can't filter\n");
exit(1);
}
// use dummy_p to write to file, as first packet