Recherche avancée
Médias (1)
-
Bug de détection d’ogg
22 mars 2013, par
Mis à jour : Avril 2013
Langue : français
Type : Video
Autres articles (50)
-
Use, discuss, criticize
13 avril 2011, parTalk to people directly involved in MediaSPIP’s development, or to people around you who could use MediaSPIP to share, enhance or develop their creative projects.
The bigger the community, the more MediaSPIP’s potential will be explored and the faster the software will evolve.
A discussion list is available for all exchanges between users. -
Personnaliser en ajoutant son logo, sa bannière ou son image de fond
5 septembre 2013, parCertains thèmes prennent en compte trois éléments de personnalisation : l’ajout d’un logo ; l’ajout d’une bannière l’ajout d’une image de fond ;
-
Ecrire une actualité
21 juin 2013, parPrésentez les changements dans votre MédiaSPIP ou les actualités de vos projets sur votre MédiaSPIP grâce à la rubrique actualités.
Dans le thème par défaut spipeo de MédiaSPIP, les actualités sont affichées en bas de la page principale sous les éditoriaux.
Vous pouvez personnaliser le formulaire de création d’une actualité.
Formulaire de création d’une actualité Dans le cas d’un document de type actualité, les champs proposés par défaut sont : Date de publication ( personnaliser la date de publication ) (...)
Sur d’autres sites (9929)
-
Evolution #2812 : Avertir de l’absence de SQLite sur la page de backup
12 mai 2013, par cedric -@Thierry : Les warning sur l’unicité des clés primaires sont "normaux" dans le sens où ils permettent de se protéger vis à vis de l’approximation inhérente à la reprise sur interruption. Les backups au format XML des versions précédentes avaient les mêmes défauts mais sans protection de ce côté là, conduisant à des données erronées au final.
L’absence de table meta a la connexion est aussi normale lors du backup.
Le fait que ce soit lent est lui inhérent au mode de sauvegarde où l’on lit les tables morceau par morceau pour les écrires dans un fichier (que ce soit SQLite ou XML ne change rien sur ce point).Et bien sur, mysqldump est bien plus rapide, de plusieurs ordres de grandeur. Mais le format SQL qu’il produit n’est pas portable, pose parfois des problèmes au ré-import sur un autre serveur avec une version mySQL différente, et n’est pas importable dans un SQLite par exemple.
Bref pas de solution idéale, encore une fois celle proposée dans le core essaye d’être le plus générique possible.
-
Opening the ALSA microphone port using libavformat
17 mai 2013, par Kartik AiyerI'm writing a fairly simple application to open the microphone port using the ALSA format on a linux VM. I'm trying to do this using libavformat and other ffmpeg libs ( avutil, avcodec, etc ).
I followed what ffmpeg was doing ( ffmpeg_opt.c : open_input_file) to open the actual microphone port( hw:0,0). I find the ALSA format using
av_find_input_format( "alsa" )which returns with aAVInputFormat*object.I then use this with
avformat_open_input.
I didn't precreate AVFormatContext and load it with any params before the above call ( I rely on it being created inavformat_open_input). This worked and I was able to read packets of the port. However, after a while I was not able to open the port. I added a bunch of debug in the libs and find out that :
avformat_open_input -> init_input -> ... -> avio_open2 -> ffurl_open ->ffurl_alloc
the final call to ffurl_alloc goes through all the protocols and tries to see ifhwmatches any given protocol. This fails.I was wondering if anyone has tried opening a mic with the ALSA format using libavformat and other libs, and if they have any tips on how to get this done. Specifically how does ffmpeg translate something like hw:0,0 to the actual device port ?
P.S : I tried the following line on the command line app
ffmpeg -f alsa -i hw:0,0 -ac 2 -acodec pcm_s16le output.aac
this worked for a while and then suddenly stopped working as well with the same kind of problem. Unable to open the protocol.I'm not sure what changed and why its not able to open the audio port, but any advice on how also port are opened with the rather different handle of HW:x,x would be much appreciated.
Alternatively, I realize ffmpeg is a little heavy to just read PCM packets and then encode and I could use a more direct approach by using the aac libs directly. Can anyone recommend a lib or kernel driver that I can use to open the microphone ( given that it is an ALSA card )and read PCM data out of it. ( I'm new to LInux so please forgive me if any of this sounds trivial )
Thanks
Kartik -
FFmpeg C++ api decode h264 error
29 mai 2015, par armsI’m trying to use the C++ API of FFMpeg (version 20150526) under Windows using the prebuilt binaries to decode an h264 video file (*.ts).
I’ve written a very simple code that automatically detects the required codec from the file itself (and it is AV_CODEC_ID_H264, as expected).
Then I re-open the video file in read-binary mode and I read a fixed-size buffer of bytes from it and provide the read bytes to the decoder within a while-loop until the end of file. However when I call the function avcodec_decode_video2 a large amount of errors happen like the following ones :
[h264 @ 008df020] top block unavailable for requested intro mode at 34 0
[h264 @ 008df020] error while decoding MB 34 0, bytestream 3152
[h264 @ 008df020] decode_slice_header error
Sometimes the function avcodec_decode_video2 sets the value of got_picture_ptr to 1 and hence I expect to find a good frame. Instead, though all the computations are successful, when I view the decoded frame (using OpenCV only for visualization purposes) I see a gray one with some artifacts.
If I employ the same code to decode an *.avi file it works fine.
Reading the examples of FFMpeg I did not find a solution to my problem. I’ve also implemented the solution proposed in the simlar question FFmpeg c++ H264 decoding error but it did not work.
Does anyone know where the error is ?
Thank you in advance for any reply !
The code is the following [EDIT : code updated including the parser management] :
#include <iostream>
#include <iomanip>
#include <string>
#include <sstream>
#include <opencv2></opencv2>opencv.hpp>
#ifdef __cplusplus
extern "C"
{
#endif // __cplusplus
#include <libavcodec></libavcodec>avcodec.h>
#include <libavdevice></libavdevice>avdevice.h>
#include <libavfilter></libavfilter>avfilter.h>
#include <libavformat></libavformat>avformat.h>
#include <libavformat></libavformat>avio.h>
#include <libavutil></libavutil>avutil.h>
#include <libpostproc></libpostproc>postprocess.h>
#include <libswresample></libswresample>swresample.h>
#include <libswscale></libswscale>swscale.h>
#ifdef __cplusplus
} // end extern "C".
#endif // __cplusplus
#define INBUF_SIZE 4096
void main()
{
AVCodec* l_pCodec;
AVCodecContext* l_pAVCodecContext;
SwsContext* l_pSWSContext;
AVFormatContext* l_pAVFormatContext;
AVFrame* l_pAVFrame;
AVFrame* l_pAVFrameBGR;
AVPacket l_AVPacket;
AVPacket l_AVPacket_out;
AVStream* l_pStream;
AVCodecParserContext* l_pParser;
FILE* l_pFile_in;
FILE* l_pFile_out;
std::string l_sFile;
int l_iResult;
int l_iFrameCount;
int l_iGotFrame;
int l_iBufLength;
int l_iParsedBytes;
int l_iPts;
int l_iDts;
int l_iPos;
int l_iSize;
int l_iDecodedBytes;
uint8_t l_auiInBuf[INBUF_SIZE + FF_INPUT_BUFFER_PADDING_SIZE];
uint8_t* l_pData;
cv::Mat l_cvmImage;
l_pCodec = NULL;
l_pAVCodecContext = NULL;
l_pSWSContext = NULL;
l_pAVFormatContext = NULL;
l_pAVFrame = NULL;
l_pAVFrameBGR = NULL;
l_pParser = NULL;
l_pStream = NULL;
l_pFile_in = NULL;
l_pFile_out = NULL;
l_iPts = 0;
l_iDts = 0;
l_iPos = 0;
l_pData = NULL;
l_sFile = "myvideo.ts";
avdevice_register_all();
avfilter_register_all();
avcodec_register_all();
av_register_all();
avformat_network_init();
l_pAVFormatContext = avformat_alloc_context();
l_iResult = avformat_open_input(&l_pAVFormatContext,
l_sFile.c_str(),
NULL,
NULL);
if (l_iResult >= 0)
{
l_iResult = avformat_find_stream_info(l_pAVFormatContext, NULL);
if (l_iResult >= 0)
{
for (int i=0; inb_streams; i++)
{
if (l_pAVFormatContext->streams[i]->codec->codec_type ==
AVMEDIA_TYPE_VIDEO)
{
l_pCodec = avcodec_find_decoder(
l_pAVFormatContext->streams[i]->codec->codec_id);
l_pStream = l_pAVFormatContext->streams[i];
}
}
}
}
av_init_packet(&l_AVPacket);
av_init_packet(&l_AVPacket_out);
memset(l_auiInBuf + INBUF_SIZE, 0, FF_INPUT_BUFFER_PADDING_SIZE);
if (l_pCodec)
{
l_pAVCodecContext = avcodec_alloc_context3(l_pCodec);
l_pParser = av_parser_init(l_pAVCodecContext->codec_id);
if (l_pParser)
{
av_register_codec_parser(l_pParser->parser);
}
if (l_pAVCodecContext)
{
if (l_pCodec->capabilities & CODEC_CAP_TRUNCATED)
{
l_pAVCodecContext->flags |= CODEC_FLAG_TRUNCATED;
}
l_iResult = avcodec_open2(l_pAVCodecContext, l_pCodec, NULL);
if (l_iResult >= 0)
{
l_pFile_in = fopen(l_sFile.c_str(), "rb");
if (l_pFile_in)
{
l_pAVFrame = av_frame_alloc();
l_pAVFrameBGR = av_frame_alloc();
if (l_pAVFrame)
{
l_iFrameCount = 0;
avcodec_get_frame_defaults(l_pAVFrame);
while (1)
{
l_iBufLength = fread(l_auiInBuf,
1,
INBUF_SIZE,
l_pFile_in);
if (l_iBufLength == 0)
{
break;
}
else
{
l_pData = l_auiInBuf;
l_iSize = l_iBufLength;
while (l_iSize > 0)
{
if (l_pParser)
{
l_iParsedBytes = av_parser_parse2(
l_pParser,
l_pAVCodecContext,
&l_AVPacket_out.data,
&l_AVPacket_out.size,
l_pData,
l_iSize,
l_AVPacket.pts,
l_AVPacket.dts,
AV_NOPTS_VALUE);
if (l_iParsedBytes <= 0)
{
break;
}
l_AVPacket.pts = l_AVPacket.dts = AV_NOPTS_VALUE;
l_AVPacket.pos = -1;
}
else
{
l_AVPacket_out.data = l_pData;
l_AVPacket_out.size = l_iSize;
}
l_iDecodedBytes =
avcodec_decode_video2(
l_pAVCodecContext,
l_pAVFrame,
&l_iGotFrame,
&l_AVPacket_out);
if (l_iDecodedBytes >= 0)
{
if (l_iGotFrame)
{
l_pSWSContext = sws_getContext(
l_pAVCodecContext->width,
l_pAVCodecContext->height,
l_pAVCodecContext->pix_fmt,
l_pAVCodecContext->width,
l_pAVCodecContext->height,
AV_PIX_FMT_BGR24,
SWS_BICUBIC,
NULL,
NULL,
NULL);
if (l_pSWSContext)
{
l_iResult = avpicture_alloc(
reinterpret_cast(l_pAVFrameBGR),
AV_PIX_FMT_BGR24,
l_pAVFrame->width,
l_pAVFrame->height);
l_iResult = sws_scale(
l_pSWSContext,
l_pAVFrame->data,
l_pAVFrame->linesize,
0,
l_pAVCodecContext->height,
l_pAVFrameBGR->data,
l_pAVFrameBGR->linesize);
if (l_iResult > 0)
{
l_cvmImage = cv::Mat(
l_pAVFrame->height,
l_pAVFrame->width,
CV_8UC3,
l_pAVFrameBGR->data[0],
l_pAVFrameBGR->linesize[0]);
if (l_cvmImage.empty() == false)
{
cv::imshow("image", l_cvmImage);
cv::waitKey(10);
}
}
}
l_iFrameCount++;
}
}
else
{
break;
}
l_pData += l_iParsedBytes;
l_iSize -= l_iParsedBytes;
}
}
} // end while(1).
}
fclose(l_pFile_in);
}
}
}
}
}
</sstream></string></iomanip></iostream>
EDIT : The following is the final code that solves my problem, thanks to the suggestions of Ronald.
#include <iostream>
#include <iomanip>
#include <string>
#include <sstream>
#include <opencv2></opencv2>opencv.hpp>
#ifdef __cplusplus
extern "C"
{
#endif // __cplusplus
#include <libavcodec></libavcodec>avcodec.h>
#include <libavdevice></libavdevice>avdevice.h>
#include <libavfilter></libavfilter>avfilter.h>
#include <libavformat></libavformat>avformat.h>
#include <libavformat></libavformat>avio.h>
#include <libavutil></libavutil>avutil.h>
#include <libpostproc></libpostproc>postprocess.h>
#include <libswresample></libswresample>swresample.h>
#include <libswscale></libswscale>swscale.h>
#ifdef __cplusplus
} // end extern "C".
#endif // __cplusplus
void main()
{
AVCodec* l_pCodec;
AVCodecContext* l_pAVCodecContext;
SwsContext* l_pSWSContext;
AVFormatContext* l_pAVFormatContext;
AVFrame* l_pAVFrame;
AVFrame* l_pAVFrameBGR;
AVPacket l_AVPacket;
std::string l_sFile;
uint8_t* l_puiBuffer;
int l_iResult;
int l_iFrameCount;
int l_iGotFrame;
int l_iDecodedBytes;
int l_iVideoStreamIdx;
int l_iNumBytes;
cv::Mat l_cvmImage;
l_pCodec = NULL;
l_pAVCodecContext = NULL;
l_pSWSContext = NULL;
l_pAVFormatContext = NULL;
l_pAVFrame = NULL;
l_pAVFrameBGR = NULL;
l_puiBuffer = NULL;
l_sFile = "myvideo.ts";
av_register_all();
l_iResult = avformat_open_input(&l_pAVFormatContext,
l_sFile.c_str(),
NULL,
NULL);
if (l_iResult >= 0)
{
l_iResult = avformat_find_stream_info(l_pAVFormatContext, NULL);
if (l_iResult >= 0)
{
for (int i=0; inb_streams; i++)
{
if (l_pAVFormatContext->streams[i]->codec->codec_type ==
AVMEDIA_TYPE_VIDEO)
{
l_iVideoStreamIdx = i;
l_pAVCodecContext =
l_pAVFormatContext->streams[l_iVideoStreamIdx]->codec;
if (l_pAVCodecContext)
{
l_pCodec = avcodec_find_decoder(l_pAVCodecContext->codec_id);
}
break;
}
}
}
}
if (l_pCodec && l_pAVCodecContext)
{
l_iResult = avcodec_open2(l_pAVCodecContext, l_pCodec, NULL);
if (l_iResult >= 0)
{
l_pAVFrame = av_frame_alloc();
l_pAVFrameBGR = av_frame_alloc();
l_iNumBytes = avpicture_get_size(PIX_FMT_BGR24,
l_pAVCodecContext->width,
l_pAVCodecContext->height);
l_puiBuffer = (uint8_t *)av_malloc(l_iNumBytes*sizeof(uint8_t));
avpicture_fill((AVPicture *)l_pAVFrameBGR,
l_puiBuffer,
PIX_FMT_RGB24,
l_pAVCodecContext->width,
l_pAVCodecContext->height);
l_pSWSContext = sws_getContext(
l_pAVCodecContext->width,
l_pAVCodecContext->height,
l_pAVCodecContext->pix_fmt,
l_pAVCodecContext->width,
l_pAVCodecContext->height,
AV_PIX_FMT_BGR24,
SWS_BICUBIC,
NULL,
NULL,
NULL);
while (av_read_frame(l_pAVFormatContext, &l_AVPacket) >= 0)
{
if (l_AVPacket.stream_index == l_iVideoStreamIdx)
{
l_iDecodedBytes = avcodec_decode_video2(
l_pAVCodecContext,
l_pAVFrame,
&l_iGotFrame,
&l_AVPacket);
if (l_iGotFrame)
{
if (l_pSWSContext)
{
l_iResult = sws_scale(
l_pSWSContext,
l_pAVFrame->data,
l_pAVFrame->linesize,
0,
l_pAVCodecContext->height,
l_pAVFrameBGR->data,
l_pAVFrameBGR->linesize);
if (l_iResult > 0)
{
l_cvmImage = cv::Mat(
l_pAVFrame->height,
l_pAVFrame->width,
CV_8UC3,
l_pAVFrameBGR->data[0],
l_pAVFrameBGR->linesize[0]);
if (l_cvmImage.empty() == false)
{
cv::imshow("image", l_cvmImage);
cv::waitKey(1);
}
}
}
l_iFrameCount++;
}
}
}
}
}
}
</sstream></string></iomanip></iostream>
Syndiquer tout le site
