FFMPEG (C++) convert & compress a single image out of buffer - c++

i try to encode (with compression) and decode (without compression) a image with ffmpeg. But if i want to get the sent image back with avcodec_receive_packet i get only the error AVERROR(EAGAIN).
It doesnt matter what i change ... allways AVERROR(EAGAIN) is the outcome. Is it maybe a problem of sending just one single frame to the encoder? And if yes, how to fix it?
Code (only relevant stuff shown):
avcodec_register_all();
/* ------ init codec ------------------*/
AVCodec *codec;
codec = avcodec_find_decoder(AV_CODEC_ID_H264);
if (!codec)
{
print("compressH264, could not find decoder:\"AV_CODEC_ID_H264\"!!!");
return false;
}
AVCodec *nVidiaCodec = avcodec_find_encoder_by_name("h264_nvenc");
if (!nVidiaCodec)
{
print("err");
}
/* ------ ------------ ------------------*/
/* ------ init context ------------------*/
AVCodecContext* av_codec_context_ = NULL;
av_codec_context_ = avcodec_alloc_context3(nVidiaCodec);
if (!av_codec_context_)
{
print("compressH264, avcodec_alloc_context3 failed!!!");
return false;
}
int w = imgSrc.width();
int h = imgSrc.height();
if ((w % 2) != 0)
{
++w;
}
if ((h % 2) != 0)
{
++h;
}
av_codec_context_->width = w;
av_codec_context_->height = h;
av_codec_context_->pix_fmt = AV_PIX_FMT_YUV420P;
av_codec_context_->gop_size = 1;
av_codec_context_->max_b_frames = 1;
av_codec_context_->bit_rate = 400000;
av_codec_context_->time_base.den = 1;
av_codec_context_->time_base.num = 1;
av_opt_set(av_codec_context_->priv_data, "preset", "slow", 0);
int ret = avcodec_open2(av_codec_context_, nVidiaCodec, NULL);
if (0 > ret)
{
print("compressH264, could not open codec context for decoder:\"AV_CODEC_ID_H264\"!!!");
return false;
}
AVFrame *picture = av_frame_alloc();
picture->format = AV_PIX_FMT_RGB24;
picture->width = w;
picture->height = h;
ret = avpicture_fill((AVPicture *)picture, imgSrc.bits(), AV_PIX_FMT_RGB24, w, h);
if (0 > ret)
{
print("compressH264, avpicture_fill - failed!!!");
return false;
}
AVFrame *tmp_picture = av_frame_alloc();
tmp_picture->format = AV_PIX_FMT_YUV420P;
tmp_picture->width = w;
tmp_picture->height = h;
ret = av_frame_get_buffer(tmp_picture, 32);
SwsContext *img_convert_ctx = sws_getContext(av_codec_context_->width, av_codec_context_->height, AV_PIX_FMT_RGB24, av_codec_context_->width, av_codec_context_->height, av_codec_context_->pix_fmt, SWS_BICUBIC, NULL, NULL, NULL);
ret = sws_scale(img_convert_ctx, picture->data, picture->linesize, 0, av_codec_context_->height, tmp_picture->data, tmp_picture->linesize);
int h264Size = avpicture_get_size(AV_PIX_FMT_YUV420P, w, h);
ret = avcodec_send_frame(av_codec_context_, tmp_picture);
if (0 > ret)
{
char err[AV_ERROR_MAX_STRING_SIZE];
av_make_error_string(err, AV_ERROR_MAX_STRING_SIZE, ret);
print("compressH264, avcodec_send_frame: %s", err);
}
AVPacket *pkt = av_packet_alloc();
while (ret >= 0)
{
ret = avcodec_receive_packet(av_codec_context_, pkt);
if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF)
{
break;
}
else if (ret < 0)
{
fprintf(stderr, "Error during encoding\n");
exit(1);
}
av_packet_unref(pkt);
}
print("success");
Everything works well until:
- avcodec_receive_packet ... i get all time the error AVERROR(EAGAIN).
I can start decoding just if i have the compressed image.
Thanks for your help guys.
Edit:
If i do now the following code, i get a packet and ret == 0, but i have to send 46 times the same image ... for me this makes no sence.
do
{
ret = avcodec_receive_packet(av_codec_context_, &pkt);
if (ret == 0)
{
break;
}
else if ((ret < 0) && (ret != AVERROR(EAGAIN)))
{
coutF("error");
}
else if (ret == AVERROR(EAGAIN))
{
ret = avcodec_send_frame(av_codec_context_, tmp_picture);
if (0 > ret)
{
char err[AV_ERROR_MAX_STRING_SIZE];
av_make_error_string(err, AV_ERROR_MAX_STRING_SIZE, ret);
coutFRed("compressH264, avcodec_send_frame: %s", err);
}
coutF("cnt:%d", ++cnt);
}
} while (ret == 0);
Edit:
Good morning,
after more invest, i got the issue. I have to send the same frame a lot of time, because of the keyframe stuff for h264. The question now is, if it is possible to remove the h264 standart stuff from the encoder and just let FFMPEG convert one single frame.

I am not sure but following an ffmpeg example it seems that that just means it is done and you should return like they do in this code snippet:
/* if no more frames for output - returns AVERROR(EAGAIN)
* if flushed and no more frames for output - returns AVERROR_EOF
* rewrite retcode to 0 to show it as normal procedure completion
*/
if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF)
ret = 0;
In their comments they seem to imply that it signals "normal procedure completion"

the answer for this issue is this:
do
{
ret = avcodec_receive_packet(av_codec_context_, &pkt);
if (ret == 0)
{
break;
}
else if ((ret < 0) && (ret != AVERROR(EAGAIN)))
{
return false;
}
else if (ret == AVERROR(EAGAIN))
{
ret = avcodec_send_frame(av_codec_context_, NULL);
if (0 > ret)
{
return false;
}
}
} while (ret == 0);
The NULL frame will flush the buffer and we are able to get the encoded frame.

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!

How to use FFmpeg API overlay filter in C / C++

I have a C++ project which creates 7/24 WebTV like RTMP stream and allows operations like changing current content on runtime, seeking content, looping through a playlist which is constructed by a json array, also supports changing whole playlist on runtime.
Currently i am reading H264 and AAC encoded packets from mp4 files then sending them to destination RTMP server after adjusting their PTS & DTS values without any encoding or decoding.
But i want to apply overlay images to raw frames using FFmpeg "overlay" filter after decoding H264 packets. I looked at sample which came with FFmpeg examples ;
#define _XOPEN_SOURCE 600 /* for usleep */
#include <unistd.h>
#include <stdio.h>
#include <stdlib.h>
#include <libavcodec/avcodec.h>
#include <libavformat/avformat.h>
#include <libavfilter/buffersink.h>
#include <libavfilter/buffersrc.h>
#include <libavutil/opt.h>
const char *filter_descr = "scale=78:24,transpose=cclock";
/* other way:
scale=78:24 [scl]; [scl] transpose=cclock // assumes "[in]" and "[out]" to be input output pads respectively
*/
static AVFormatContext *fmt_ctx;
static AVCodecContext *dec_ctx;
AVFilterContext *buffersink_ctx;
AVFilterContext *buffersrc_ctx;
AVFilterGraph *filter_graph;
static int video_stream_index = -1;
static int64_t last_pts = AV_NOPTS_VALUE;
static int open_input_file(const char *filename)
{
int ret;
AVCodec *dec;
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);
/* 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;
}
/*
* Set the endpoints for the filter graph. The filter_graph will
* be linked to the graph described by filters_descr.
*/
/*
* The buffer source output must be connected to the input pad of
* the first filter described by filters_descr; since the first
* filter input label is not specified, it is set to "in" by
* default.
*/
outputs->name = av_strdup("in");
outputs->filter_ctx = buffersrc_ctx;
outputs->pad_idx = 0;
outputs->next = NULL;
/*
* The buffer sink input must be connected to the output pad of
* the last filter described by filters_descr; since the last
* filter output label is not specified, it is set to "out" by
* default.
*/
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. */
delay = av_rescale_q(frame->pts - last_pts,
time_base, AV_TIME_BASE_Q);
if (delay > 0 && delay < 1000000)
usleep(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 main(int argc, char **argv)
{
int ret;
AVPacket packet;
AVFrame *frame;
AVFrame *filt_frame;
if (argc != 2) {
fprintf(stderr, "Usage: %s file\n", argv[0]);
exit(1);
}
frame = av_frame_alloc();
filt_frame = av_frame_alloc();
if (!frame || !filt_frame) {
perror("Could not allocate frame");
exit(1);
}
if ((ret = open_input_file(argv[1])) < 0)
goto end;
if ((ret = init_filters(filter_descr)) < 0)
goto end;
/* read all packets */
while (1) {
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);
}
av_frame_unref(frame);
}
}
av_packet_unref(&packet);
}
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);
if (ret < 0 && ret != AVERROR_EOF) {
fprintf(stderr, "Error occurred: %s\n", av_err2str(ret));
exit(1);
}
exit(0);
}
That sample uses these filters ;
"scale=78:24,transpose=cclock"
I compiled and run it with a sample video file but it just outputs fancy characters to console, the code block given below is responsible for this ;
/* 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);
I have no issues with Encoding & Decoding, i just don't know how to apply "overlay" filter. Are there any tutorials out there demonstrate how to use "overlay" filter?
Just like in the example, except you use "overlay".
snprintf(args, sizeof(args), args here...);
avfilter_graph_create_filter(sink, avfilter_get_by_name("overlay"), nullptr, nullptr, arg, graph);
Then you need TWO create two source pads. i.e.
avfilter_graph_create_filter(sourceX, avfilter_get_by_name("buffer"), nullptr, args, nullptr, m_graph);
and one sink pad. Then feed one source with the video frame and other second with the image to overlay
Following code fragments will be helpful..
char args[512];
int ret = 0;
const AVFilter *bufferSrc = avfilter_get_by_name("buffer");
const AVFilter *bufferOvr = avfilter_get_by_name("buffer");
const AVFilter *bufferSink = avfilter_get_by_name("buffersink");
const AVFilter *ovrFilter = avfilter_get_by_name("overlay");
const AVFilter *colorFilter = avfilter_get_by_name("colorchannelmixer");
enum AVPixelFormat pix_fmts[] = { AV_PIX_FMT_YUV420P, AV_PIX_FMT_NONE };
fFilterGraph = avfilter_graph_alloc();
if (!fFilterGraph) {
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",
decCtx->width, decCtx->height, decCtx->pix_fmt,
fTimeBase.num, fTimeBase.den,
decCtx->sample_aspect_ratio.num, decCtx->sample_aspect_ratio.den);
ret = avfilter_graph_create_filter(&fBufSrc0Ctx, bufferSrc, "in0",
args, NULL, fFilterGraph);
if (ret < 0)
goto end;
/* buffer video overlay source: the overlayed frame from the file will be inserted here. */
snprintf(args, sizeof(args),
"video_size=%dx%d:pix_fmt=%d:time_base=%d/%d:pixel_aspect=%d/%d",
ovrCtx->width, ovrCtx->height, ovrCtx->pix_fmt,
fTimeBase.num, fTimeBase.den,
ovrCtx->sample_aspect_ratio.num, ovrCtx->sample_aspect_ratio.den);
ret = avfilter_graph_create_filter(&fBufSrc1Ctx, bufferOvr, "in1",
args, NULL, fFilterGraph);
if (ret < 0)
goto end;
/* color filter */
snprintf(args, sizeof(args), "aa=%f", (float)fWatermarkOpacity / 10.0);
ret = avfilter_graph_create_filter(&fColorFilterCtx, colorFilter, "colorFilter",
args, NULL, fFilterGraph);
if (ret < 0)
goto end;
/* overlay filter */
switch (fWatermarkPos) {
case 0:
/* Top left */
snprintf(args, sizeof(args), "x=%d:y=%d:repeatlast=1",
fWatermarkOffset, fWatermarkOffset);
break;
case 1:
/* Top right */
snprintf(args, sizeof(args), "x=W-w-%d:y=%d:repeatlast=1",
fWatermarkOffset, fWatermarkOffset);
break;
case 3:
/* Bottom left */
snprintf(args, sizeof(args), "x=%d:y=H-h-%d:repeatlast=1",
fWatermarkOffset, fWatermarkOffset);
break;
case 4:
/* Bottom right */
snprintf(args, sizeof(args), "x=W-w-%d:y=H-h-%d:repeatlast=1",
fWatermarkOffset, fWatermarkOffset);
break;
case 2:
default:
/* Center */
snprintf(args, sizeof(args), "x=(W-w)/2:y=(H-h)/2:repeatlast=1");
break;
}
ret = avfilter_graph_create_filter(&fOvrFilterCtx, ovrFilter, "overlay",
args, NULL, fFilterGraph);
if (ret < 0)
goto end;
/* buffer sink - destination of the final video */
ret = avfilter_graph_create_filter(&fBufSinkCtx, bufferSink, "out",
NULL, NULL, fFilterGraph);
if (ret < 0)
goto end;
ret = av_opt_set_int_list(fBufSinkCtx, "pix_fmts", pix_fmts,
AV_PIX_FMT_NONE, AV_OPT_SEARCH_CHILDREN);
if (ret < 0)
goto end;
/*
* Link all filters..
*/
avfilter_link(fBufSrc0Ctx, 0, fOvrFilterCtx, 0);
avfilter_link(fBufSrc1Ctx, 0, fColorFilterCtx, 0);
avfilter_link(fColorFilterCtx, 0, fOvrFilterCtx, 1);
avfilter_link(fOvrFilterCtx, 0, fBufSinkCtx, 0);
if ((ret = avfilter_graph_config(fFilterGraph, NULL)) < 0)
goto end;
end:

Error in avcodec_receive_packet (gdi screenshot + ffmpeg)

I try to capture windows screen with ffmpeg.
All's working, but avcodec_receive_packet return error AVERROR(EAGAIN)
and I can't undestand why it's happenning.
Can anybody give advice?
SwsContext* convertContext = sws_getContext(c->width, c->height, AV_PIX_FMT_BGRA, c->width, c->height, c->pix_fmt, SWS_FAST_BILINEAR, NULL, NULL, NULL);
for (int i = 0; i < fps*seconds; i++)
{
fflush(stdout);
/* make sure the frame data is writable */
ret = av_frame_make_writable(frame);
if (ret < 0)
exit(1);
gdi->MakeScreenshoot();
OutFrame->pts = i;
int ret = av_image_fill_arrays(GDIFrame->data, GDIFrame->linesize, gdi->m_bufferGDIBits, AV_PIX_FMT_BGRA, c->width, c->height, 1);
ret = av_image_fill_arrays(OutFrame->data, OutFrame->linesize, outbuffer, c->pix_fmt, c->width, c->height, 1);
GDIFrame->data[0] += GDIFrame->linesize[0] * (c->height - 1); // flipping frame
GDIFrame->linesize[0] *= -1;
int hslice = sws_scale(convertContext, GDIFrame->data, GDIFrame->linesize, 0, c->height,OutFrame->data, OutFrame->linesize);
/* encode the image */
encode(c, OutFrame, pkt, f);
}
static void encode(AVCodecContext *enc_ctx, AVFrame *frame, AVPacket *pkt, FILE *outfile)
{
int ret = -1;
/* send the frame to the encoder */
/*if (frame)
printf("Send frame %3"PRId64"\n", frame->pts);*/
ret = avcodec_send_frame(enc_ctx, frame);
if (ret < 0)
{
fprintf(stderr, "Error sending a frame for encoding\n");
exit(1);
}
while (ret >= 0)
{
ret = avcodec_receive_packet(enc_ctx, pkt);
if (ret == AVERROR_EOF)
return;
if (ret == AVERROR(EAGAIN))
return;
else
if (ret < 0)
{
fprintf(stderr, "Error during encoding\n");
exit(1);
}
//printf("Write packet %3"PRId64" (size=%5d)\n", pkt->pts, pkt->size);
fwrite(pkt->data, 1, pkt->size, outfile);
av_packet_unref(pkt);
}
}
If I comment
if (ret == AVERROR_EOF)
return;
if (ret == AVERROR(EAGAIN))
return;
else
if (ret < 0)
{
fprintf(stderr, "Error during encoding\n");
exit(1);
}
in encode function - all is fine, file will be written and will be correct, but I want solve the problem.
Bit late to the party.
I am also trying this out with encode_video.c
Seems that avcodec_receive_packet() will return these "errors" when encoder does not have a complete packet ready... possibly next call to avcodec_send_frame() will result in a full packet.
So just keep going.
The final call to avcodec_send_frame() with NULL frame will flush the final packet.

FFMPEG H264 encode each single image

i encode currently a QImage from RGB888 to H264, but i want to encode each image (even if this is not the perfect way) by itself.
Im able to encode the image, but its needed to send the same image 46 times. And i dont know what i do wrong (probably wrong config of the encode, but i cannot find the issue there).
Afterwards i decode this image and then convert it back to a QImage. I do this only for testing some other code.
avcodec_register_all();
AVCodec *nVidiaCodec = avcodec_find_encoder_by_name("h264_nvenc");
if (!nVidiaCodec)
{
return false;
}
AVCodecContext* av_codec_context_ = NULL;
av_codec_context_ = avcodec_alloc_context3(nVidiaCodec);
if (!av_codec_context_)
{
return false;
}
av_codec_context_->width = dst->width;
av_codec_context_->height = dst->height;
av_codec_context_->pix_fmt = AV_PIX_FMT_YUV420P;
av_codec_context_->gop_size = 1;
av_codec_context_->keyint_min = 0;
av_codec_context_->scenechange_threshold = 0;
av_codec_context_->bit_rate = 8000000;
av_codec_context_->time_base.den = 1;
av_codec_context_->time_base.num = 1;
av_codec_context_->refs = 0;
av_codec_context_->qmin = 1;
av_codec_context_->qmax = 1;
av_codec_context_->b_frame_strategy = 0;
av_codec_context_->max_b_frames = 0;
av_codec_context_->thread_count = 1;
av_opt_set(av_codec_context_, "preset", "slow", 0);
av_opt_set(av_codec_context_, "tune", "zerolatency", 0);
int ret = avcodec_open2(av_codec_context_, nVidiaCodec, NULL);
if (0 > ret)
{
return false;
}
AVFrame *picture = av_frame_alloc();
picture->format = AV_PIX_FMT_RGB24;
picture->width = dst->width;
picture->height = dst->height;
ret = avpicture_fill((AVPicture *)picture, imgSrc.bits(), AV_PIX_FMT_RGB24, dst->width, dst->height);
if (0 > ret)
{
return false;
}
AVFrame *tmp_picture = av_frame_alloc();
tmp_picture->format = AV_PIX_FMT_YUV420P;
tmp_picture->width = dst->width;
tmp_picture->height = dst->height;
ret = av_frame_get_buffer(tmp_picture, 32);
SwsContext *img_convert_ctx = sws_getContext(av_codec_context_->width, av_codec_context_->height, AV_PIX_FMT_RGB24, av_codec_context_->width, av_codec_context_->height, av_codec_context_->pix_fmt, SWS_BICUBIC, NULL, NULL, NULL);
if (!img_convert_ctx)
{
return false;
}
ret = sws_scale(img_convert_ctx, picture->data, picture->linesize, 0, av_codec_context_->height, tmp_picture->data, tmp_picture->linesize);
if (0 > ret)
{
return false;
}
ret = avcodec_send_frame(av_codec_context_, tmp_picture);
if (0 > ret)
{
return false;
}
AVPacket pkt;
av_init_packet(&pkt);
pkt.data = NULL;
pkt.size = 0;
do
{
ret = avcodec_receive_packet(av_codec_context_, &pkt);
if (ret == 0)
{
break;
}
else if ((ret < 0) && (ret != AVERROR(EAGAIN)))
{
return false;
}
else if (ret == AVERROR(EAGAIN))
{
ret = avcodec_send_frame(av_codec_context_, tmp_picture);
if (0 > ret)
{
return false;
}
}
} while (ret == 0);
// the do while is called 46 times, then i get the packet, but i want to get the packet at the first call
It would be very nice if you can help me.
Thanks guys.
I assume you just want to encode a single frame. You need to flush the encoder after you have sent your single uncompressed frame by sending NULL instead of a valid buffer.
int result = 0;
// encoder init
// send one uncompressed frame
result = avcodec_send_frame(av_codec_context_, tmp_picture);
if (result < 0) return false;
// send NULL to indicate flushing
result = avcodec_send_frame(av_codec_context_, NULL);
if (result < 0) return false;
while (result != AVERROR_EOF)
{
result = avcodec_receive_packet(av_codec_context_, &pkt);
if (!result)
{
// you should have your encoded frame; do something with it
}
}

How to use palettegen and paletteuse filters with FFmpeg in C++?

I want to create an animated gif in a QT project.
When I scale the a QImage to AV_PIX_FMT_RGB8 directly the output looks awfull with flickering artifacts and when I scale to AV_PIX_FMT_YUV420P in between the output is dithered which does not look much better.
I found out that ffmpeg is able to produce a palette file using a filter called palettegen and then convert a movie to gif using this palette.
Is there any sample c++ file out which I could use for my project or does anybody have a clue who to use these filters in code?
The filter can be describe like below
format=pix_fmts=rgb32,fps=10,scale=320:240:flags=lanczos,split [o1] [o2];[o1] palettegen [p]; [o2] fifo [o3];[o3] [p] paletteuse
You can do some fps or scale customization with the filter above.
Then we should apply this filter graph in ffmpeg.
One important note is, as the filter palettegen need the entire stream, we need to add the whole stream to the buffer source, and a end of stream to buffer source, then we can get frame from buffer sink and muxing to the output gif file.
Here is a completed example, change the micro definition in your test and be sure to use the latest version of ffmpeg.
#define INPUT_PATH "/Users/gif/Downloads/VUE_1503566813005.mp4"
#define OUTPUT_PATH "/Users/gif/Downloads/VUE.gif"
video2gif.cpp
#include <iostream>
#ifdef __cplusplus
extern "C" {
#include "libavformat/avformat.h"
#include "libavfilter/avfilter.h"
#include "libavutil/opt.h"
#include "libavfilter/buffersrc.h"
#include "libavfilter/buffersink.h"
#include "libswscale/swscale.h"
#include "libavutil/imgutils.h"
}
#endif
#define INPUT_PATH "/Users/gif/Downloads/VUE_1503566813005.mp4"
#define OUTPUT_PATH "/Users/gif/Downloads/VUE.gif"
const char *filter_descr = "format=pix_fmts=rgb32,fps=10,scale=320:240:flags=lanczos,split [o1] [o2];[o1] palettegen [p]; [o2] fifo [o3];[o3] [p] paletteuse";
static AVFormatContext* ofmt_ctx;
static AVCodecContext* o_codec_ctx;
static AVFilterGraph* filter_graph;
static AVFilterContext* buffersrc_ctx;
static AVFilterContext* buffersink_ctx;
static int init_filters(const char* filter_desc,
AVFormatContext* ifmt_ctx,
int stream_index,
AVCodecContext* dec_ctx)
{
char args[512];
int ret = 0;
AVFilter* buffersrc = avfilter_get_by_name("buffer");
AVFilter* buffersink = avfilter_get_by_name("buffersink");
AVFilterInOut* inputs = avfilter_inout_alloc();
AVFilterInOut* outputs = avfilter_inout_alloc();
AVRational time_base = ifmt_ctx->streams[stream_index]->time_base;
enum AVPixelFormat pix_fmts[] = { AV_PIX_FMT_PAL8, AV_PIX_FMT_NONE };
filter_graph = avfilter_graph_alloc();
if (!outputs || !inputs || !filter_graph) {
ret = AVERROR(ENOMEM);
return ret;
}
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, nullptr, filter_graph);
if (ret < 0) {
av_log(nullptr, AV_LOG_ERROR, "Cannot create buffer source\n");
return ret;
}
ret = avfilter_graph_create_filter(&buffersink_ctx, buffersink, "out", nullptr, nullptr, filter_graph);
if (ret < 0) {
av_log(nullptr, AV_LOG_ERROR, "Cannot create buffer sink\n");
return 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(nullptr, AV_LOG_ERROR, "can not set output pixel format\n");
return ret;
}
outputs->name = av_strdup("in");
outputs->filter_ctx = buffersrc_ctx;
outputs->pad_idx = 0;
outputs->next = nullptr;
inputs->name = av_strdup("out");
inputs->filter_ctx = buffersink_ctx;
inputs->pad_idx = 0;
inputs->next = nullptr;
if ((ret = avfilter_graph_parse_ptr(filter_graph, filter_desc, &inputs, &outputs, nullptr)) < 0) {
av_log(nullptr, AV_LOG_ERROR, "parse filter graph error\n");
return ret;
}
if ((ret = avfilter_graph_config(filter_graph, nullptr)) < 0) {
av_log(nullptr, AV_LOG_ERROR, "config graph error\n");
}
avfilter_inout_free(&inputs);
avfilter_inout_free(&outputs);
return 0;
}
static int init_muxer()
{
int ret = 0;
AVOutputFormat* o_fmt = av_guess_format("gif", OUTPUT_PATH, "video/gif");
ret = avformat_alloc_output_context2(&ofmt_ctx, o_fmt, "gif", OUTPUT_PATH);
if (ret < 0) {
av_log(nullptr, AV_LOG_ERROR, "%s allocate output format\n", av_err2str(ret));
return -1;
}
AVCodec* codec = avcodec_find_encoder(AV_CODEC_ID_GIF);
if (!codec) {
return -1;
}
AVStream* stream = avformat_new_stream(ofmt_ctx, codec);
AVCodecParameters* codec_paramters = stream->codecpar;
codec_paramters->codec_tag = 0;
codec_paramters->codec_id = codec->id;
codec_paramters->codec_type = AVMEDIA_TYPE_VIDEO;
codec_paramters->width = 320;
codec_paramters->height = 240;
codec_paramters->format = AV_PIX_FMT_PAL8;
o_codec_ctx = avcodec_alloc_context3(codec);
avcodec_parameters_to_context(o_codec_ctx, codec_paramters);
o_codec_ctx->time_base = { 1, 10 };
if (ofmt_ctx->oformat->flags & AVFMT_GLOBALHEADER) {
o_codec_ctx->flags |= AV_CODEC_FLAG_GLOBAL_HEADER;
}
ret = avcodec_open2(o_codec_ctx, codec, NULL);
if (ret < 0) {
av_log(nullptr, AV_LOG_ERROR, "%s open output codec\n", av_err2str(ret));
return ret;
}
ret = avio_open(&ofmt_ctx->pb, OUTPUT_PATH, AVIO_FLAG_WRITE);
if (ret < 0) {
av_log(nullptr, AV_LOG_ERROR, "%s avio open error\n", av_err2str(ret));
return ret;
}
ret = avformat_write_header(ofmt_ctx, NULL);
if (ret < 0) {
av_log(nullptr, AV_LOG_ERROR, "%s write header\n", av_err2str(ret));
return ret;
}
av_dump_format(ofmt_ctx, -1, OUTPUT_PATH, 1);
return 0;
}
static void destroy_muxer()
{
avformat_free_context(ofmt_ctx);
avcodec_close(o_codec_ctx);
avcodec_free_context(&o_codec_ctx);
}
static void destroy_filter()
{
avfilter_free(buffersrc_ctx);
avfilter_free(buffersink_ctx);
avfilter_graph_free(&filter_graph);
}
static void muxing_one_frame(AVFrame* frame)
{
int ret = avcodec_send_frame(o_codec_ctx, frame);
AVPacket *pkt = av_packet_alloc();
av_init_packet(pkt);
while (ret >= 0) {
ret = avcodec_receive_packet(o_codec_ctx, pkt);
if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF) {
break;
}
av_write_frame(ofmt_ctx, pkt);
}
av_packet_unref(pkt);
}
int main(int argc, const char * argv[]) {
av_register_all();
avcodec_register_all();
avfilter_register_all();
int ret = 0;
AVFormatContext* fmt_ctx = avformat_alloc_context();
AVCodecContext* codec_ctx = nullptr;
AVCodec* codec = nullptr;
int video_index = -1;
AVPacket *pkt = av_packet_alloc();
av_init_packet(pkt);
pkt->size = 0;
pkt->data = nullptr;
ret = avformat_open_input(&fmt_ctx, INPUT_PATH, NULL, NULL);
if (ret != 0) {
std::cerr << "error open input" << std::endl;
goto die;
}
avformat_find_stream_info(fmt_ctx, NULL);
av_dump_format(fmt_ctx, -1, INPUT_PATH, 0);
ret = av_find_best_stream(fmt_ctx, AVMEDIA_TYPE_VIDEO, -1, -1, &codec, 0);
if (ret == AVERROR_STREAM_NOT_FOUND) {
std::cerr << "error no video stream found" << std::endl;
goto die;
}
for (unsigned i = 0; i < fmt_ctx->nb_streams; i++) {
if (fmt_ctx->streams[i]->codecpar->codec_type == AVMEDIA_TYPE_VIDEO) {
video_index = i;
if (ret == AVERROR_DECODER_NOT_FOUND) {
codec = avcodec_find_decoder(fmt_ctx->streams[i]->codecpar->codec_id);
}
break;
}
}
if (!codec) {
std::cerr << "could not find the decoder" << std::endl;
goto die;
}
codec_ctx = avcodec_alloc_context3(codec);
avcodec_parameters_to_context(codec_ctx, fmt_ctx->streams[video_index]->codecpar);
ret = avcodec_open2(codec_ctx, codec, NULL);
if (ret < 0) {
std::cerr << "open codec error" << std::endl;
goto die;
}
if (init_muxer() < 0) {
av_log(nullptr, AV_LOG_ERROR, "could not init muxer\n");
goto die;
}
if ((ret = init_filters(filter_descr, fmt_ctx, video_index, codec_ctx)) < 0) {
av_log(nullptr, AV_LOG_ERROR, "could not init filters %s\n", av_err2str(ret));
goto die;
}
// it's time to decode
while (av_read_frame(fmt_ctx, pkt) == 0) {
if (pkt->stream_index == video_index) {
ret = avcodec_send_packet(codec_ctx, pkt);
while (ret >= 0) {
AVFrame* frame = av_frame_alloc();
ret = avcodec_receive_frame(codec_ctx, frame);
if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF) {
break;
}
if (ret >= 0) {
// processing one frame
frame->pts = frame->best_effort_timestamp;
// palettegen need a whole stream, just add frame to buffer and don't get frame
ret = av_buffersrc_add_frame_flags(buffersrc_ctx, frame, AV_BUFFERSRC_FLAG_KEEP_REF);
if (ret < 0) {
av_log(nullptr, AV_LOG_ERROR, "error add frame to buffer source %s\n", av_err2str(ret));
}
}
av_frame_free(&frame);
}
}
}
// end of buffer
if ((ret = av_buffersrc_add_frame_flags(buffersrc_ctx, nullptr, AV_BUFFERSRC_FLAG_KEEP_REF)) >= 0) {
do {
AVFrame* filter_frame = av_frame_alloc();
ret = av_buffersink_get_frame(buffersink_ctx, filter_frame);
if (ret == AVERROR(EAGAIN) || ret == AVERROR_EOF) {
// av_log(nullptr, AV_LOG_ERROR, "error get frame from buffer sink %s\n", av_err2str(ret));
av_frame_unref(filter_frame);
break;
}
// write the filter frame to output file
muxing_one_frame(filter_frame);
av_log(nullptr, AV_LOG_INFO, "muxing one frame\n");
av_frame_unref(filter_frame);
} while (ret >= 0);
} else {
av_log(nullptr, AV_LOG_ERROR, "error add frame to buffer source %s\n", av_err2str(ret));
}
av_packet_free(&pkt);
av_write_trailer(ofmt_ctx);
destroy_muxer();
destroy_filter();
die:
avformat_free_context(fmt_ctx);
avcodec_close(codec_ctx);
avcodec_free_context(&codec_ctx);
return 0;
}