Recherche avancée
Autres articles (85)
-
Organiser par catégorie
17 mai 2013, parDans MédiaSPIP, une rubrique a 2 noms : catégorie et rubrique.
Les différents documents stockés dans MédiaSPIP peuvent être rangés dans différentes catégories. On peut créer une catégorie en cliquant sur "publier une catégorie" dans le menu publier en haut à droite ( après authentification ). Une catégorie peut être rangée dans une autre catégorie aussi ce qui fait qu’on peut construire une arborescence de catégories.
Lors de la publication prochaine d’un document, la nouvelle catégorie créée sera proposée (...) -
Récupération d’informations sur le site maître à l’installation d’une instance
26 novembre 2010, parUtilité
Sur le site principal, une instance de mutualisation est définie par plusieurs choses : Les données dans la table spip_mutus ; Son logo ; Son auteur principal (id_admin dans la table spip_mutus correspondant à un id_auteur de la table spip_auteurs)qui sera le seul à pouvoir créer définitivement l’instance de mutualisation ;
Il peut donc être tout à fait judicieux de vouloir récupérer certaines de ces informations afin de compléter l’installation d’une instance pour, par exemple : récupérer le (...) -
Websites made with MediaSPIP
2 mai 2011, parThis page lists some websites based on MediaSPIP.
Sur d’autres sites (7579)
-
ffmpeg record screen and save video file to disk as .mpg
7 janvier 2015, par musimbateI want to record the screen of my pc (using gdigrab on my windows machine) and store the saved video file on my disk as an mp4 or mpg file .I have found an example piece of code that grabs the screen and shows it in an SDL window here :http://xwk.iteye.com/blog/2125720 (The code is on the bottom of the page and has an english version) and the ffmpeg muxing example https://ffmpeg.org/doxygen/trunk/muxing_8c-source.html seems to be able to help encode audio and video into a desired output video file.
I have tried to combine these two by having a format context for grabbing the screen (AVFormatContext *pFormatCtx ; in my code ) and a separate format context to write the desired video file (AVFormatContext *outFormatContextEncoded ;).Within the loop to read packets from the input stream( screen grab stream) I directly encode write packets to the output file as shown in my code.I have kept the SDL code so I can see what I am recording.Below is my code with my modified write_video_frame() function .
The code builds OK but the output video can’t be played by vlc. When I run the command
ffmpeg -i filename.mpgI get this output
[mpeg @ 003fed20] probed stream 0 failed
[mpeg @ 003fed20] Stream #0: not enough frames to estimate rate; consider increasing probesize
[mpeg @ 003fed20] Could not find codec parameters for stream 0 (Video: none): unknown codec
Consider increasing the value for the 'analyzeduration' and 'probesize' options
karamage.mpg: could not find codec parameters
Input #0, mpeg, from 'karamage.mpg':
Duration: 19:30:09.25, start: 37545.438756, bitrate: 2 kb/s
Stream #0:0[0x1e0]: Video: none, 90k tbr, 90k tbn
At least one output file must be specifiedAm I doing something wrong here ? I am new to ffmpeg and any guidance on this is highly appreciated.Thank you for your time.
int main(int argc, char* argv[])
{
AVFormatContext *pFormatCtx;
int i, videoindex;
AVCodecContext *pCodecCtx;
AVCodec *pCodec;
av_register_all();
avformat_network_init();
//Localy defined structure.
OutputStream outVideoStream = { 0 };
const char *filename;
AVOutputFormat *outFormatEncoded;
AVFormatContext *outFormatContextEncoded;
AVCodec *videoCodec;
filename="karamage.mpg";
int ret1;
int have_video = 0, have_audio = 0;
int encode_video = 0, encode_audio = 0;
AVDictionary *opt = NULL;
//ASSIGN STH TO THE FORMAT CONTEXT.
pFormatCtx = avformat_alloc_context();
//
//Use this when opening a local file.
//char filepath[]="src01_480x272_22.h265";
//avformat_open_input(&pFormatCtx,filepath,NULL,NULL)
//Register Device
avdevice_register_all();
//Use gdigrab
AVDictionary* options = NULL;
//Set some options
//grabbing frame rate
//av_dict_set(&options,"framerate","5",0);
//The distance from the left edge of the screen or desktop
//av_dict_set(&options,"offset_x","20",0);
//The distance from the top edge of the screen or desktop
//av_dict_set(&options,"offset_y","40",0);
//Video frame size. The default is to capture the full screen
//av_dict_set(&options,"video_size","640x480",0);
AVInputFormat *ifmt=av_find_input_format("gdigrab");
if(avformat_open_input(&pFormatCtx,"desktop",ifmt,&options)!=0){
printf("Couldn't open input stream.\n");
return -1;
}
if(avformat_find_stream_info(pFormatCtx,NULL)<0)
{
printf("Couldn't find stream information.\n");
return -1;
}
videoindex=-1;
for(i=0; inb_streams; i++)
if(pFormatCtx->streams[i]->codec->codec_type==AVMEDIA_TYPE_VIDEO)
{
videoindex=i;
break;
}
if(videoindex==-1)
{
printf("Didn't find a video stream.\n");
return -1;
}
pCodecCtx=pFormatCtx->streams[videoindex]->codec;
pCodec=avcodec_find_decoder(pCodecCtx->codec_id);
if(pCodec==NULL)
{
printf("Codec not found.\n");
return -1;
}
if(avcodec_open2(pCodecCtx, pCodec,NULL)<0)
{
printf("Could not open codec.\n");
return -1;
}
AVFrame *pFrame,*pFrameYUV;
pFrame=avcodec_alloc_frame();
pFrameYUV=avcodec_alloc_frame();
//PIX_FMT_YUV420P WHAT DOES THIS SAY ABOUT THE FORMAT??
uint8_t *out_buffer=(uint8_t *)av_malloc(avpicture_get_size(PIX_FMT_YUV420P, pCodecCtx->width, pCodecCtx->height));
avpicture_fill((AVPicture *)pFrameYUV, out_buffer, PIX_FMT_YUV420P, pCodecCtx->width, pCodecCtx->height);
//<<<<<<<<<<<-------PREP WORK TO WRITE ENCODED VIDEO FILES-----
avformat_alloc_output_context2(&outFormatContextEncoded, NULL, NULL, filename);
if (!outFormatContextEncoded) {
printf("Could not deduce output format from file extension: using MPEG.\n");
avformat_alloc_output_context2(&outFormatContextEncoded, NULL, "mpeg", filename);
}
if (!outFormatContextEncoded)
return 1;
outFormatEncoded=outFormatContextEncoded->oformat;
//THIS CREATES THE STREAMS(AUDIO AND VIDEO) ADDED TO OUR OUTPUT STREAM
if (outFormatEncoded->video_codec != AV_CODEC_ID_NONE) {
//YOUR VIDEO AND AUDIO PROPS ARE SET HERE.
add_stream(&outVideoStream, outFormatContextEncoded, &videoCodec, outFormatEncoded->video_codec);
have_video = 1;
encode_video = 1;
}
// Now that all the parameters are set, we can open the audio and
// video codecs and allocate the necessary encode buffers.
if (have_video)
open_video(outFormatContextEncoded, videoCodec, &outVideoStream, opt);
av_dump_format(outFormatContextEncoded, 0, filename, 1);
/* open the output file, if needed */
if (!(outFormatEncoded->flags & AVFMT_NOFILE)) {
ret1 = avio_open(&outFormatContextEncoded->pb, filename, AVIO_FLAG_WRITE);
if (ret1 < 0) {
//fprintf(stderr, "Could not open '%s': %s\n", filename,
// av_err2str(ret));
fprintf(stderr, "Could not open your dumb file.\n");
return 1;
}
}
/* Write the stream header, if any. */
ret1 = avformat_write_header(outFormatContextEncoded, &opt);
if (ret1 < 0) {
//fprintf(stderr, "Error occurred when opening output file: %s\n",
// av_err2str(ret));
fprintf(stderr, "Error occurred when opening output file\n");
return 1;
}
//<<<<<<<<<<<-------PREP WORK TO WRITE ENCODED VIDEO FILES-----
//SDL----------------------------
if(SDL_Init(SDL_INIT_VIDEO | SDL_INIT_AUDIO | SDL_INIT_TIMER)) {
printf( "Could not initialize SDL - %s\n", SDL_GetError());
return -1;
}
int screen_w=640,screen_h=360;
const SDL_VideoInfo *vi = SDL_GetVideoInfo();
//Half of the Desktop's width and height.
screen_w = vi->current_w/2;
screen_h = vi->current_h/2;
SDL_Surface *screen;
screen = SDL_SetVideoMode(screen_w, screen_h, 0,0);
if(!screen) {
printf("SDL: could not set video mode - exiting:%s\n",SDL_GetError());
return -1;
}
SDL_Overlay *bmp;
bmp = SDL_CreateYUVOverlay(pCodecCtx->width, pCodecCtx->height,SDL_YV12_OVERLAY, screen);
SDL_Rect rect;
//SDL End------------------------
int ret, got_picture;
AVPacket *packet=(AVPacket *)av_malloc(sizeof(AVPacket));
//TRY TO INIT THE PACKET HERE
av_init_packet(packet);
//Output Information-----------------------------
printf("File Information---------------------\n");
av_dump_format(pFormatCtx,0,NULL,0);
printf("-------------------------------------------------\n");
struct SwsContext *img_convert_ctx;
img_convert_ctx = sws_getContext(pCodecCtx->width, pCodecCtx->height, pCodecCtx->pix_fmt, pCodecCtx->width, pCodecCtx->height, PIX_FMT_YUV420P, SWS_BICUBIC, NULL, NULL, NULL);
//------------------------------
//
while(av_read_frame(pFormatCtx, packet)>=0)
{
if(packet->stream_index==videoindex)
{
//HERE WE DECODE THE PACKET INTO THE FRAME
ret = avcodec_decode_video2(pCodecCtx, pFrame, &got_picture, packet);
if(ret < 0)
{
printf("Decode Error.\n");
return -1;
}
if(got_picture)
{
//THIS IS WHERE WE DO STH WITH THE FRAME WE JUST GOT FROM THE STREAM
//FREE AREA--START
//IN HERE YOU CAN WORK WITH THE FRAME OF THE PACKET.
write_video_frame(outFormatContextEncoded, &outVideoStream,packet);
//FREE AREA--END
sws_scale(img_convert_ctx, (const uint8_t* const*)pFrame->data, pFrame->linesize, 0, pCodecCtx->height, pFrameYUV->data, pFrameYUV->linesize);
SDL_LockYUVOverlay(bmp);
bmp->pixels[0]=pFrameYUV->data[0];
bmp->pixels[2]=pFrameYUV->data[1];
bmp->pixels[1]=pFrameYUV->data[2];
bmp->pitches[0]=pFrameYUV->linesize[0];
bmp->pitches[2]=pFrameYUV->linesize[1];
bmp->pitches[1]=pFrameYUV->linesize[2];
SDL_UnlockYUVOverlay(bmp);
rect.x = 0;
rect.y = 0;
rect.w = screen_w;
rect.h = screen_h;
SDL_DisplayYUVOverlay(bmp, &rect);
//Delay 40ms----WHY THIS DELAY????
SDL_Delay(40);
}
}
av_free_packet(packet);
}//THE LOOP TO PULL PACKETS FROM THE FORMAT CONTEXT ENDS HERE.
//AFTER THE WHILE LOOP WE DO SOME CLEANING
//av_read_pause(context);
av_write_trailer(outFormatContextEncoded);
close_stream(outFormatContextEncoded, &outVideoStream);
if (!(outFormatContextEncoded->flags & AVFMT_NOFILE))
/* Close the output file. */
avio_close(outFormatContextEncoded->pb);
/* free the stream */
avformat_free_context(outFormatContextEncoded);
//STOP DOING YOUR CLEANING
sws_freeContext(img_convert_ctx);
SDL_Quit();
av_free(out_buffer);
av_free(pFrameYUV);
avcodec_close(pCodecCtx);
avformat_close_input(&pFormatCtx);
return 0;
}
/*
* encode one video frame and send it to the muxer
* return 1 when encoding is finished, 0 otherwise
*/
static int write_video_frame(AVFormatContext *oc, OutputStream *ost,AVPacket * pkt11)
{
int ret;
AVCodecContext *c;
AVFrame *frame;
int got_packet = 0;
c = ost->st->codec;
//DO NOT NEED THIS FRAME.
//frame = get_video_frame(ost);
if (oc->oformat->flags & AVFMT_RAWPICTURE) {
//IGNORE THIS FOR A MOMENT
/* a hack to avoid data copy with some raw video muxers */
AVPacket pkt;
av_init_packet(&pkt);
if (!frame)
return 1;
pkt.flags |= AV_PKT_FLAG_KEY;
pkt.stream_index = ost->st->index;
pkt.data = (uint8_t *)frame;
pkt.size = sizeof(AVPicture);
pkt.pts = pkt.dts = frame->pts;
av_packet_rescale_ts(&pkt, c->time_base, ost->st->time_base);
ret = av_interleaved_write_frame(oc, &pkt);
} else {
ret = write_frame(oc, &c->time_base, ost->st, pkt11);
}
if (ret < 0) {
fprintf(stderr, "Error while writing video frame: %s\n");
exit(1);
}
return 1;
} -
ffmpeg record screen and save video file to disk as .mp4 or .mpg
5 janvier 2015, par musimbateI want to record the screen of my pc (using gdigrab on my windows machine) and store the saved video file on my disk as an mp4 or mpg file .I have found an example piece of code that grabs the screen and shows it in an SDL window here :http://xwk.iteye.com/blog/2125720 (The code is on the bottom of the page and has an english version) and the ffmpeg muxing example https://ffmpeg.org/doxygen/trunk/muxing_8c-source.html seems to be able to help encode audio and video into a desired output video file.
I have tried to combine these two by having a format context for grabbing the screen (AVFormatContext *pFormatCtx ; in my code ) and a separate format context to write the desired video file (AVFormatContext *outFormatContextEncoded ;).Within the loop to read packets from the input stream( screen grab stream) I directly encode write packets to the output file as shown in my code.I have kept the SDL code so I can see what I am recording.Below is my code with my modified write_video_frame() function .
The code builds OK but the output video can’t be played by vlc. When I run the command
ffmpeg -i filename.mpgI get this output
[mpeg @ 003fed20] probed stream 0 failed
[mpeg @ 003fed20] Stream #0: not enough frames to estimate rate; consider increasing probesize
[mpeg @ 003fed20] Could not find codec parameters for stream 0 (Video: none): unknown codec
Consider increasing the value for the 'analyzeduration' and 'probesize' options
karamage.mpg: could not find codec parameters
Input #0, mpeg, from 'karamage.mpg':
Duration: 19:30:09.25, start: 37545.438756, bitrate: 2 kb/s
Stream #0:0[0x1e0]: Video: none, 90k tbr, 90k tbn
At least one output file must be specifiedAm I doing something wrong here ? I am new to ffmpeg and any guidance on this is highly appreciated.Thank you for your time.
int main(int argc, char* argv[])
{
AVFormatContext *pFormatCtx;
int i, videoindex;
AVCodecContext *pCodecCtx;
AVCodec *pCodec;
av_register_all();
avformat_network_init();
//Localy defined structure.
OutputStream outVideoStream = { 0 };
const char *filename;
AVOutputFormat *outFormatEncoded;
AVFormatContext *outFormatContextEncoded;
AVCodec *videoCodec;
filename="karamage.mpg";
int ret1;
int have_video = 0, have_audio = 0;
int encode_video = 0, encode_audio = 0;
AVDictionary *opt = NULL;
//ASSIGN STH TO THE FORMAT CONTEXT.
pFormatCtx = avformat_alloc_context();
//
//Use this when opening a local file.
//char filepath[]="src01_480x272_22.h265";
//avformat_open_input(&pFormatCtx,filepath,NULL,NULL)
//Register Device
avdevice_register_all();
//Use gdigrab
AVDictionary* options = NULL;
//Set some options
//grabbing frame rate
//av_dict_set(&options,"framerate","5",0);
//The distance from the left edge of the screen or desktop
//av_dict_set(&options,"offset_x","20",0);
//The distance from the top edge of the screen or desktop
//av_dict_set(&options,"offset_y","40",0);
//Video frame size. The default is to capture the full screen
//av_dict_set(&options,"video_size","640x480",0);
AVInputFormat *ifmt=av_find_input_format("gdigrab");
if(avformat_open_input(&pFormatCtx,"desktop",ifmt,&options)!=0){
printf("Couldn't open input stream.\n");
return -1;
}
if(avformat_find_stream_info(pFormatCtx,NULL)<0)
{
printf("Couldn't find stream information.\n");
return -1;
}
videoindex=-1;
for(i=0; inb_streams; i++)
if(pFormatCtx->streams[i]->codec->codec_type==AVMEDIA_TYPE_VIDEO)
{
videoindex=i;
break;
}
if(videoindex==-1)
{
printf("Didn't find a video stream.\n");
return -1;
}
pCodecCtx=pFormatCtx->streams[videoindex]->codec;
pCodec=avcodec_find_decoder(pCodecCtx->codec_id);
if(pCodec==NULL)
{
printf("Codec not found.\n");
return -1;
}
if(avcodec_open2(pCodecCtx, pCodec,NULL)<0)
{
printf("Could not open codec.\n");
return -1;
}
AVFrame *pFrame,*pFrameYUV;
pFrame=avcodec_alloc_frame();
pFrameYUV=avcodec_alloc_frame();
//PIX_FMT_YUV420P WHAT DOES THIS SAY ABOUT THE FORMAT??
uint8_t *out_buffer=(uint8_t *)av_malloc(avpicture_get_size(PIX_FMT_YUV420P, pCodecCtx->width, pCodecCtx->height));
avpicture_fill((AVPicture *)pFrameYUV, out_buffer, PIX_FMT_YUV420P, pCodecCtx->width, pCodecCtx->height);
//<<<<<<<<<<<-------PREP WORK TO WRITE ENCODED VIDEO FILES-----
avformat_alloc_output_context2(&outFormatContextEncoded, NULL, NULL, filename);
if (!outFormatContextEncoded) {
printf("Could not deduce output format from file extension: using MPEG.\n");
avformat_alloc_output_context2(&outFormatContextEncoded, NULL, "mpeg", filename);
}
if (!outFormatContextEncoded)
return 1;
outFormatEncoded=outFormatContextEncoded->oformat;
//THIS CREATES THE STREAMS(AUDIO AND VIDEO) ADDED TO OUR OUTPUT STREAM
if (outFormatEncoded->video_codec != AV_CODEC_ID_NONE) {
//YOUR VIDEO AND AUDIO PROPS ARE SET HERE.
add_stream(&outVideoStream, outFormatContextEncoded, &videoCodec, outFormatEncoded->video_codec);
have_video = 1;
encode_video = 1;
}
// Now that all the parameters are set, we can open the audio and
// video codecs and allocate the necessary encode buffers.
if (have_video)
open_video(outFormatContextEncoded, videoCodec, &outVideoStream, opt);
av_dump_format(outFormatContextEncoded, 0, filename, 1);
/* open the output file, if needed */
if (!(outFormatEncoded->flags & AVFMT_NOFILE)) {
ret1 = avio_open(&outFormatContextEncoded->pb, filename, AVIO_FLAG_WRITE);
if (ret1 < 0) {
//fprintf(stderr, "Could not open '%s': %s\n", filename,
// av_err2str(ret));
fprintf(stderr, "Could not open your dumb file.\n");
return 1;
}
}
/* Write the stream header, if any. */
ret1 = avformat_write_header(outFormatContextEncoded, &opt);
if (ret1 < 0) {
//fprintf(stderr, "Error occurred when opening output file: %s\n",
// av_err2str(ret));
fprintf(stderr, "Error occurred when opening output file\n");
return 1;
}
//<<<<<<<<<<<-------PREP WORK TO WRITE ENCODED VIDEO FILES-----
//SDL----------------------------
if(SDL_Init(SDL_INIT_VIDEO | SDL_INIT_AUDIO | SDL_INIT_TIMER)) {
printf( "Could not initialize SDL - %s\n", SDL_GetError());
return -1;
}
int screen_w=640,screen_h=360;
const SDL_VideoInfo *vi = SDL_GetVideoInfo();
//Half of the Desktop's width and height.
screen_w = vi->current_w/2;
screen_h = vi->current_h/2;
SDL_Surface *screen;
screen = SDL_SetVideoMode(screen_w, screen_h, 0,0);
if(!screen) {
printf("SDL: could not set video mode - exiting:%s\n",SDL_GetError());
return -1;
}
SDL_Overlay *bmp;
bmp = SDL_CreateYUVOverlay(pCodecCtx->width, pCodecCtx->height,SDL_YV12_OVERLAY, screen);
SDL_Rect rect;
//SDL End------------------------
int ret, got_picture;
AVPacket *packet=(AVPacket *)av_malloc(sizeof(AVPacket));
//TRY TO INIT THE PACKET HERE
av_init_packet(packet);
//Output Information-----------------------------
printf("File Information---------------------\n");
av_dump_format(pFormatCtx,0,NULL,0);
printf("-------------------------------------------------\n");
struct SwsContext *img_convert_ctx;
img_convert_ctx = sws_getContext(pCodecCtx->width, pCodecCtx->height, pCodecCtx->pix_fmt, pCodecCtx->width, pCodecCtx->height, PIX_FMT_YUV420P, SWS_BICUBIC, NULL, NULL, NULL);
//------------------------------
//
while(av_read_frame(pFormatCtx, packet)>=0)
{
if(packet->stream_index==videoindex)
{
//HERE WE DECODE THE PACKET INTO THE FRAME
ret = avcodec_decode_video2(pCodecCtx, pFrame, &got_picture, packet);
if(ret < 0)
{
printf("Decode Error.\n");
return -1;
}
if(got_picture)
{
//THIS IS WHERE WE DO STH WITH THE FRAME WE JUST GOT FROM THE STREAM
//FREE AREA--START
//IN HERE YOU CAN WORK WITH THE FRAME OF THE PACKET.
write_video_frame(outFormatContextEncoded, &outVideoStream,packet);
//FREE AREA--END
sws_scale(img_convert_ctx, (const uint8_t* const*)pFrame->data, pFrame->linesize, 0, pCodecCtx->height, pFrameYUV->data, pFrameYUV->linesize);
SDL_LockYUVOverlay(bmp);
bmp->pixels[0]=pFrameYUV->data[0];
bmp->pixels[2]=pFrameYUV->data[1];
bmp->pixels[1]=pFrameYUV->data[2];
bmp->pitches[0]=pFrameYUV->linesize[0];
bmp->pitches[2]=pFrameYUV->linesize[1];
bmp->pitches[1]=pFrameYUV->linesize[2];
SDL_UnlockYUVOverlay(bmp);
rect.x = 0;
rect.y = 0;
rect.w = screen_w;
rect.h = screen_h;
SDL_DisplayYUVOverlay(bmp, &rect);
//Delay 40ms----WHY THIS DELAY????
SDL_Delay(40);
}
}
av_free_packet(packet);
}//THE LOOP TO PULL PACKETS FROM THE FORMAT CONTEXT ENDS HERE.
//AFTER THE WHILE LOOP WE DO SOME CLEANING
//av_read_pause(context);
av_write_trailer(outFormatContextEncoded);
close_stream(outFormatContextEncoded, &outVideoStream);
if (!(outFormatContextEncoded->flags & AVFMT_NOFILE))
/* Close the output file. */
avio_close(outFormatContextEncoded->pb);
/* free the stream */
avformat_free_context(outFormatContextEncoded);
//STOP DOING YOUR CLEANING
sws_freeContext(img_convert_ctx);
SDL_Quit();
av_free(out_buffer);
av_free(pFrameYUV);
avcodec_close(pCodecCtx);
avformat_close_input(&pFormatCtx);
return 0;
}
/*
* encode one video frame and send it to the muxer
* return 1 when encoding is finished, 0 otherwise
*/
static int write_video_frame(AVFormatContext *oc, OutputStream *ost,AVPacket * pkt11)
{
int ret;
AVCodecContext *c;
AVFrame *frame;
int got_packet = 0;
c = ost->st->codec;
//DO NOT NEED THIS FRAME.
//frame = get_video_frame(ost);
if (oc->oformat->flags & AVFMT_RAWPICTURE) {
//IGNORE THIS FOR A MOMENT
/* a hack to avoid data copy with some raw video muxers */
AVPacket pkt;
av_init_packet(&pkt);
if (!frame)
return 1;
pkt.flags |= AV_PKT_FLAG_KEY;
pkt.stream_index = ost->st->index;
pkt.data = (uint8_t *)frame;
pkt.size = sizeof(AVPicture);
pkt.pts = pkt.dts = frame->pts;
av_packet_rescale_ts(&pkt, c->time_base, ost->st->time_base);
ret = av_interleaved_write_frame(oc, &pkt);
} else {
ret = write_frame(oc, &c->time_base, ost->st, pkt11);
}
if (ret < 0) {
fprintf(stderr, "Error while writing video frame: %s\n");
exit(1);
}
return 1;
} -
ffmpeg record video plays too fast
29 avril 2023, par Kris XiaI'm a college student and I am studying FFmpeg now.



I have wrote a software that can record desktops and audio('virtual-audio-capturer') with FFmpeg.And I am now writing Audio and Video Synchronization.
I met some problems that video recording plays too fast.



When I look for audio and video synchronization help on the Internet，I find a formula for calculating PTS ：



pts = n * (（1 / timbase）/ fps)



When I use this formula,I find a phenomenon.



1.The higher frame rate is,the faster the video playback speed.



2.The slower the frame rate, the faster the video playback.



Also I find while the framerate is 10，the video playback speed will be right.



Why has this situation happened ?



I have thought this question for three days. I really hope someone can help me solve this problem.



I really appreciate the help.



#include "stdafx.h"

#ifdef __cplusplus
extern "C"
{
#endif
#include "libavcodec/avcodec.h"
#include "libavformat/avformat.h"
#include "libswscale/swscale.h"
#include "libavdevice/avdevice.h"
#include "libavutil/audio_fifo.h"

#include "libavfilter/buffersink.h"
#include "libavfilter/buffersrc.h"
#include "libavutil/imgutils.h"
#include "libavutil/mathematics.h"
#include "libavutil/samplefmt.h"
#include "libavutil/time.h"
#include "libavutil/opt.h"
#include "libavutil/pixdesc.h"
#include "libavutil/file.h"
#include "libavutil/mem.h"
#include "libavutil/frame.h"
#include "libavfilter/avfilter.h"
#include "libswresample/swresample.h"

#pragma comment(lib, "avcodec.lib")
#pragma comment(lib, "avformat.lib")
#pragma comment(lib, "avutil.lib")
#pragma comment(lib, "avdevice.lib")
#pragma comment(lib, "avfilter.lib")

#pragma comment(lib, "avfilter.lib")
#pragma comment(lib, "postproc.lib")
#pragma comment(lib, "swresample.lib")
#pragma comment(lib, "swscale.lib")
#ifdef __cplusplus
};
#endif

AVFormatContext *pFormatCtx_Video = NULL, *pFormatCtx_Audio = NULL, *pFormatCtx_Out = NULL;

AVCodecContext *outVideoCodecCtx = NULL;
AVCodecContext *outAudioCodecCtx = NULL;

AVStream *pVideoStream = NULL, *pAudioStream = NULL;

AVCodec *outAVCodec;
AVCodec *outAudioCodec;

AVCodecContext *pCodecCtx_Video;
AVCodec *pCodec_Video;
AVFifoBuffer *fifo_video = NULL;
AVAudioFifo *fifo_audio = NULL;
int VideoIndex, AudioIndex;
int codec_id;

CRITICAL_SECTION AudioSection, VideoSection;



SwsContext *img_convert_ctx;
int frame_size = 0;

uint8_t *picture_buf = NULL, *frame_buf = NULL;

bool bCap = true;

DWORD WINAPI ScreenCapThreadProc( LPVOID lpParam );
DWORD WINAPI AudioCapThreadProc( LPVOID lpParam );

int OpenVideoCapture()
{
 AVInputFormat *ifmt=av_find_input_format("gdigrab");
 AVDictionary *options = NULL;
 av_dict_set(&options, "framerate", "60", NULL);
 if(avformat_open_input(&pFormatCtx_Video, "desktop", ifmt, &options)!=0)
 {
 printf("Couldn't open input stream.（无法打开视频输入流）\n");
 return -1;
 }
 if(avformat_find_stream_info(pFormatCtx_Video,NULL)<0)
 {
 printf("Couldn't find stream information.（无法获取视频流信息）\n");
 return -1;
 }
 if (pFormatCtx_Video->streams[0]->codec->codec_type != AVMEDIA_TYPE_VIDEO)
 {
 printf("Couldn't find video stream information.（无法获取视频流信息）\n");
 return -1;
 }
 pCodecCtx_Video = pFormatCtx_Video->streams[0]->codec;
 pCodec_Video = avcodec_find_decoder(pCodecCtx_Video->codec_id);
 if(pCodec_Video == NULL)
 {
 printf("Codec not found.（没有找到解码器）\n");
 return -1;
 }
 if(avcodec_open2(pCodecCtx_Video, pCodec_Video, NULL) < 0)
 {
 printf("Could not open codec.（无法打开解码器）\n");
 return -1;
 }

 av_dump_format(pFormatCtx_Video, 0, NULL, 0);

 img_convert_ctx = sws_getContext(pCodecCtx_Video->width, pCodecCtx_Video->height, pCodecCtx_Video->pix_fmt, 
 pCodecCtx_Video->width, pCodecCtx_Video->height, PIX_FMT_YUV420P, SWS_BICUBIC, NULL, NULL, NULL); 

 frame_size = avpicture_get_size(pCodecCtx_Video->pix_fmt, pCodecCtx_Video->width, pCodecCtx_Video->height);
 fifo_video = av_fifo_alloc(30 * avpicture_get_size(AV_PIX_FMT_YUV420P, pCodecCtx_Video->width, pCodecCtx_Video->height));

 return 0;
}

static char *dup_wchar_to_utf8(wchar_t *w)
{
 char *s = NULL;
 int l = WideCharToMultiByte(CP_UTF8, 0, w, -1, 0, 0, 0, 0);
 s = (char *) av_malloc(l);
 if (s)
 WideCharToMultiByte(CP_UTF8, 0, w, -1, s, l, 0, 0);
 return s;
}

int OpenAudioCapture()
{
 AVInputFormat *pAudioInputFmt = av_find_input_format("dshow");
 char * psDevName = dup_wchar_to_utf8(L"audio=virtual-audio-capturer");

 if (avformat_open_input(&pFormatCtx_Audio, psDevName, pAudioInputFmt,NULL) < 0)
 {
 printf("Couldn't open input stream.（无法打开音频输入流）\n");
 return -1;
 }

 if(avformat_find_stream_info(pFormatCtx_Audio,NULL)<0) 
 return -1; 

 if(pFormatCtx_Audio->streams[0]->codec->codec_type != AVMEDIA_TYPE_AUDIO)
 {
 printf("Couldn't find video stream information.（无法获取音频流信息）\n");
 return -1;
 }

 AVCodec *tmpCodec = avcodec_find_decoder(pFormatCtx_Audio->streams[0]->codec->codec_id);
 if(0 > avcodec_open2(pFormatCtx_Audio->streams[0]->codec, tmpCodec, NULL))
 {
 printf("can not find or open audio decoder!\n");
 }

 av_dump_format(pFormatCtx_Audio, 0, NULL, 0);

 return 0;
}

int OpenOutPut()
{
 AVStream *pVideoStream = NULL, *pAudioStream = NULL;
 const char *outFileName = "test.mp4";
 avformat_alloc_output_context2(&pFormatCtx_Out, NULL, NULL, outFileName);

 if (pFormatCtx_Video->streams[0]->codec->codec_type == AVMEDIA_TYPE_VIDEO)
 {
 VideoIndex = 0;
 pVideoStream = avformat_new_stream(pFormatCtx_Out, NULL);
 if (!pVideoStream)
 {
 printf("can not new stream for output!\n");
 return -1;
 }

 outVideoCodecCtx = avcodec_alloc_context3(outAVCodec);
 if ( !outVideoCodecCtx )
 {
 printf("Error : avcodec_alloc_context3()\n");
 return -1;
 }

 //set codec context param
 outVideoCodecCtx = pVideoStream->codec;
 outVideoCodecCtx->codec_id = AV_CODEC_ID_MPEG4;
 outVideoCodecCtx->width = pFormatCtx_Video->streams[0]->codec->width;
 outVideoCodecCtx->height = pFormatCtx_Video->streams[0]->codec->height;
 outVideoCodecCtx->time_base = pFormatCtx_Video->streams[0]->codec->time_base;
 outVideoCodecCtx->pix_fmt = AV_PIX_FMT_YUV420P;
 outVideoCodecCtx->codec_type = AVMEDIA_TYPE_VIDEO;

 if (codec_id == AV_CODEC_ID_H264)
 {
 av_opt_set(outVideoCodecCtx->priv_data, "preset", "slow", 0);
 }

 outAVCodec = avcodec_find_encoder(AV_CODEC_ID_MPEG4);
 if( !outAVCodec )
 {
 printf("\n\nError : avcodec_find_encoder()");
 return -1;
 }
 if (pFormatCtx_Out->oformat->flags & AVFMT_GLOBALHEADER)
 outVideoCodecCtx->flags |=CODEC_FLAG_GLOBAL_HEADER;

 if ((avcodec_open2(outVideoCodecCtx,outAVCodec, NULL)) < 0)
 {
 printf("can not open the encoder\n");
 return -1;
 }
 }

 if(pFormatCtx_Audio->streams[0]->codec->codec_type == AVMEDIA_TYPE_AUDIO)
 {
 AVCodecContext *pOutputCodecCtx;
 AudioIndex = 1;
 pAudioStream = avformat_new_stream(pFormatCtx_Out, NULL);

 pAudioStream->codec->codec = avcodec_find_encoder(pFormatCtx_Out->oformat->audio_codec);

 pOutputCodecCtx = pAudioStream->codec;

 pOutputCodecCtx->sample_rate = pFormatCtx_Audio->streams[0]->codec->sample_rate;
 pOutputCodecCtx->channel_layout = pFormatCtx_Out->streams[0]->codec->channel_layout;
 pOutputCodecCtx->channels = av_get_channel_layout_nb_channels(pAudioStream->codec->channel_layout);
 if(pOutputCodecCtx->channel_layout == 0)
 {
 pOutputCodecCtx->channel_layout = AV_CH_LAYOUT_STEREO;
 pOutputCodecCtx->channels = av_get_channel_layout_nb_channels(pOutputCodecCtx->channel_layout);

 }
 pOutputCodecCtx->sample_fmt = pAudioStream->codec->codec->sample_fmts[0];
 AVRational time_base={1, pAudioStream->codec->sample_rate};
 pAudioStream->time_base = time_base;
 //audioCodecCtx->time_base = time_base;

 pOutputCodecCtx->codec_tag = 0; 
 if (pFormatCtx_Out->oformat->flags & AVFMT_GLOBALHEADER) 
 pOutputCodecCtx->flags |= CODEC_FLAG_GLOBAL_HEADER;

 if (avcodec_open2(pOutputCodecCtx, pOutputCodecCtx->codec, 0) < 0)
 {
 printf("编码器打开失败，退出程序\n");
 return -1;
 }
 }

 if (!(pFormatCtx_Out->oformat->flags & AVFMT_NOFILE))
 {
 if(avio_open(&pFormatCtx_Out->pb, outFileName, AVIO_FLAG_WRITE) < 0)
 {
 printf("can not open output file handle!\n");
 return -1;
 }
 }

 if(avformat_write_header(pFormatCtx_Out, NULL) < 0)
 {
 printf("can not write the header of the output file!\n");
 return -1;
 }

 return 0;
}

int _tmain(int argc, _TCHAR* argv[])
{
 av_register_all();
 avdevice_register_all();
 if (OpenVideoCapture() < 0)
 {
 return -1;
 }
 if (OpenAudioCapture() < 0)
 {
 return -1;
 }
 if (OpenOutPut() < 0)
 {
 return -1;
 }
// int fps;
 /*printf("输入帧率：");
 scanf_s("%d",&fps);
 if ( NULL == fps)
 {
 fps = 10;
 }*/

 InitializeCriticalSection(&VideoSection);
 InitializeCriticalSection(&AudioSection);

 AVFrame *picture = av_frame_alloc();
 int size = avpicture_get_size(pFormatCtx_Out->streams[VideoIndex]->codec->pix_fmt, 
 pFormatCtx_Out->streams[VideoIndex]->codec->width, pFormatCtx_Out->streams[VideoIndex]->codec->height);
 picture_buf = new uint8_t[size];

 avpicture_fill((AVPicture *)picture, picture_buf, 
 pFormatCtx_Out->streams[VideoIndex]->codec->pix_fmt, 
 pFormatCtx_Out->streams[VideoIndex]->codec->width, 
 pFormatCtx_Out->streams[VideoIndex]->codec->height);



 //star cap screen thread
 CreateThread( NULL, 0, ScreenCapThreadProc, 0, 0, NULL);
 //star cap audio thread
 CreateThread( NULL, 0, AudioCapThreadProc, 0, 0, NULL);
 int64_t cur_pts_v=0,cur_pts_a=0;
 int VideoFrameIndex = 0, AudioFrameIndex = 0;

 while(1)
 {
 if (_kbhit() != 0 && bCap)
 {
 bCap = false;
 Sleep(2000);
 }
 if (fifo_audio && fifo_video)
 {
 int sizeAudio = av_audio_fifo_size(fifo_audio);
 int sizeVideo = av_fifo_size(fifo_video);
 //缓存数据写完就结束循环
 if (av_audio_fifo_size(fifo_audio) <= pFormatCtx_Out->streams[AudioIndex]->codec->frame_size && 
 av_fifo_size(fifo_video) <= frame_size && !bCap)
 {
 break;
 }
 }

 if(av_compare_ts(cur_pts_v, pFormatCtx_Out->streams[VideoIndex]->time_base, 
 cur_pts_a,pFormatCtx_Out->streams[AudioIndex]->time_base) <= 0)
 {
 if (av_fifo_size(fifo_video) < frame_size && !bCap)
 {
 cur_pts_v = 0x7fffffffffffffff;
 }
 if(av_fifo_size(fifo_video) >= size)
 {
 EnterCriticalSection(&VideoSection);
 av_fifo_generic_read(fifo_video, picture_buf, size, NULL); //将数据从avfifobuffer馈送到用户提供的回调。
 LeaveCriticalSection(&VideoSection);

 avpicture_fill((AVPicture *)picture, picture_buf,
 pFormatCtx_Out->streams[VideoIndex]->codec->pix_fmt,
 pFormatCtx_Out->streams[VideoIndex]->codec->width,
 pFormatCtx_Out->streams[VideoIndex]->codec->height); //根据指定的图像参数和提供的图像数据缓冲区设置图片字段。

 //pts = n * (（1 / timbase）/ fps);
 //picture->pts = VideoFrameIndex * ((pFormatCtx_Video->streams[0]->time_base.den / pFormatCtx_Video->streams[0]->time_base.num) / 24);
 picture->pts = VideoFrameIndex * ((outVideoCodecCtx->time_base.den * 100000 / outVideoCodecCtx->time_base.num) / 180);

 int got_picture = 0;
 AVPacket pkt;
 av_init_packet(&pkt);

 pkt.data = NULL;
 pkt.size = 0;
 //从帧中获取输入的原始视频数据
 int ret = avcodec_encode_video2(pFormatCtx_Out->streams[VideoIndex]->codec, &pkt, picture, &got_picture);
 if(ret < 0)
 {
 continue;
 }

 if (got_picture==1)
 {
 pkt.stream_index = VideoIndex;
 /*int count = 1;
 pkt.pts = pkt.dts = count * ((pFormatCtx_Video->streams[0]->time_base.den / pFormatCtx_Video->streams[0]->time_base.num) / 15);
 count++;*/

 //x = pts * (timebase1.num / timebase1.den )* (timebase2.den / timebase2.num);

 pkt.pts = av_rescale_q_rnd(pkt.pts, pFormatCtx_Video->streams[0]->time_base, 
 pFormatCtx_Out->streams[VideoIndex]->time_base, (AVRounding)(AV_ROUND_NEAR_INF|AV_ROUND_PASS_MINMAX)); 
 pkt.dts = av_rescale_q_rnd(pkt.dts, pFormatCtx_Video->streams[0]->time_base, 
 pFormatCtx_Out->streams[VideoIndex]->time_base, (AVRounding)(AV_ROUND_NEAR_INF|AV_ROUND_PASS_MINMAX)); 


 pkt.duration = ((pFormatCtx_Out->streams[0]->time_base.den / pFormatCtx_Out->streams[0]->time_base.num) / 60);
 //pkt.duration = 1000/60;
 //pkt.pts = pkt.dts = Count * (ofmt_ctx->streams[stream_index]->time_base.den) /ofmt_ctx->streams[stream_index]->time_base.num / 10;

 //Count++;


 cur_pts_v = pkt.pts;

 ret = av_interleaved_write_frame(pFormatCtx_Out, &pkt);
 //delete[] pkt.data;
 av_free_packet(&pkt);
 }
 VideoFrameIndex++;
 }
 }
 else
 {
 if (NULL == fifo_audio)
 {
 continue;//还未初始化fifo
 }
 if (av_audio_fifo_size(fifo_audio) < pFormatCtx_Out->streams[AudioIndex]->codec->frame_size && !bCap)
 {
 cur_pts_a = 0x7fffffffffffffff;
 }
 if(av_audio_fifo_size(fifo_audio) >= 
 (pFormatCtx_Out->streams[AudioIndex]->codec->frame_size > 0 ? pFormatCtx_Out->streams[AudioIndex]->codec->frame_size : 1024))
 {
 AVFrame *frame;
 frame = av_frame_alloc();
 frame->nb_samples = pFormatCtx_Out->streams[AudioIndex]->codec->frame_size>0 ? pFormatCtx_Out->streams[AudioIndex]->codec->frame_size: 1024;
 frame->channel_layout = pFormatCtx_Out->streams[AudioIndex]->codec->channel_layout;
 frame->format = pFormatCtx_Out->streams[AudioIndex]->codec->sample_fmt;
 frame->sample_rate = pFormatCtx_Out->streams[AudioIndex]->codec->sample_rate;
 av_frame_get_buffer(frame, 0);

 EnterCriticalSection(&AudioSection);
 av_audio_fifo_read(fifo_audio, (void **)frame->data, 
 (pFormatCtx_Out->streams[AudioIndex]->codec->frame_size > 0 ? pFormatCtx_Out->streams[AudioIndex]->codec->frame_size : 1024));
 LeaveCriticalSection(&AudioSection);

 AVPacket pkt_out;
 av_init_packet(&pkt_out);
 int got_picture = -1;
 pkt_out.data = NULL;
 pkt_out.size = 0;

 frame->pts = AudioFrameIndex * pFormatCtx_Out->streams[AudioIndex]->codec->frame_size;
 if (avcodec_encode_audio2(pFormatCtx_Out->streams[AudioIndex]->codec, &pkt_out, frame, &got_picture) < 0)
 {
 printf("can not decoder a frame");
 }
 av_frame_free(&frame);
 if (got_picture) 
 {
 pkt_out.stream_index = AudioIndex;
 pkt_out.pts = AudioFrameIndex * pFormatCtx_Out->streams[AudioIndex]->codec->frame_size;
 pkt_out.dts = AudioFrameIndex * pFormatCtx_Out->streams[AudioIndex]->codec->frame_size;
 pkt_out.duration = pFormatCtx_Out->streams[AudioIndex]->codec->frame_size;

 cur_pts_a = pkt_out.pts;

 int ret = av_interleaved_write_frame(pFormatCtx_Out, &pkt_out);
 av_free_packet(&pkt_out);
 }
 AudioFrameIndex++;
 }
 }
 }

 delete[] picture_buf;

 av_fifo_free(fifo_video);
 av_audio_fifo_free(fifo_audio);

 av_write_trailer(pFormatCtx_Out);

 avio_close(pFormatCtx_Out->pb);
 avformat_free_context(pFormatCtx_Out);

 if (pFormatCtx_Video != NULL)
 {
 avformat_close_input(&pFormatCtx_Video);
 pFormatCtx_Video = NULL;
 }
 if (pFormatCtx_Audio != NULL)
 {
 avformat_close_input(&pFormatCtx_Audio);
 pFormatCtx_Audio = NULL;
 }

 return 0;
}

DWORD WINAPI ScreenCapThreadProc( LPVOID lpParam )
{
 AVPacket packet;
 int got_picture;
 AVFrame *pFrame;
 pFrame=av_frame_alloc();

 AVFrame *picture = av_frame_alloc();
 int size = avpicture_get_size(pFormatCtx_Out->streams[VideoIndex]->codec->pix_fmt, 
 pFormatCtx_Out->streams[VideoIndex]->codec->width, 
 pFormatCtx_Out->streams[VideoIndex]->codec->height);

 avpicture_fill((AVPicture *)picture, picture_buf, 
 pFormatCtx_Out->streams[VideoIndex]->codec->pix_fmt, 
 pFormatCtx_Out->streams[VideoIndex]->codec->width, 
 pFormatCtx_Out->streams[VideoIndex]->codec->height);

 FILE *p = NULL;
 p = fopen("proc_test.yuv", "wb+");
 av_init_packet(&packet);
 int height = pFormatCtx_Out->streams[VideoIndex]->codec->height;
 int width = pFormatCtx_Out->streams[VideoIndex]->codec->width;
 int y_size=height*width;
 while(bCap)
 {
 packet.data = NULL;
 packet.size = 0;
 if (av_read_frame(pFormatCtx_Video, &packet) < 0)
 {
 continue;
 }
 if(packet.stream_index == 0)
 {
 if (avcodec_decode_video2(pCodecCtx_Video, pFrame, &got_picture, &packet) < 0)
 {
 printf("Decode Error.（解码错误）\n");
 continue;
 }
 if (got_picture)
 {
 sws_scale(img_convert_ctx, 
 (const uint8_t* const*)pFrame->data,
 pFrame->linesize, 
 0, 
 pFormatCtx_Out->streams[VideoIndex]->codec->height,
 picture->data,
 picture->linesize);

 if (av_fifo_space(fifo_video) >= size)
 {
 EnterCriticalSection(&VideoSection); 
 av_fifo_generic_write(fifo_video, picture->data[0], y_size, NULL);
 av_fifo_generic_write(fifo_video, picture->data[1], y_size/4, NULL);
 av_fifo_generic_write(fifo_video, picture->data[2], y_size/4, NULL);
 LeaveCriticalSection(&VideoSection);
 }
 }
 }
 av_free_packet(&packet);
 }
 av_frame_free(&pFrame);
 av_frame_free(&picture);
 return 0;
}

DWORD WINAPI AudioCapThreadProc( LPVOID lpParam )
{
 AVPacket pkt;
 AVFrame *frame;
 frame = av_frame_alloc();
 int gotframe;
 while(bCap)
 {
 pkt.data = NULL;
 pkt.size = 0;
 if(av_read_frame(pFormatCtx_Audio,&pkt) < 0)
 {
 continue;
 }

 if (avcodec_decode_audio4(pFormatCtx_Audio->streams[0]->codec, frame, &gotframe, &pkt) < 0)
 {
 av_frame_free(&frame);
 printf("can not decoder a frame");
 break;
 }
 av_free_packet(&pkt);

 if (!gotframe)
 {
 printf("没有获取到数据，继续下一次");
 continue;
 }

 if (NULL == fifo_audio)
 {
 fifo_audio = av_audio_fifo_alloc(pFormatCtx_Audio->streams[0]->codec->sample_fmt, 
 pFormatCtx_Audio->streams[0]->codec->channels, 30 * frame->nb_samples);
 }

 int buf_space = av_audio_fifo_space(fifo_audio);
 if (av_audio_fifo_space(fifo_audio) >= frame->nb_samples)
 {
 EnterCriticalSection(&AudioSection);
 av_audio_fifo_write(fifo_audio, (void **)frame->data, frame->nb_samples);
 LeaveCriticalSection(&AudioSection);
 }
 }
 av_frame_free(&frame);
 return 0;
}



Maybe there is another way to calculate PTS and DTS



I hope whatever the frame rate is,video playback speed is right.Not too fast or too slow.


Syndiquer tout le site
