Recherche avancée
Médias (2)
-
Granite de l’Aber Ildut
9 septembre 2011, par
Mis à jour : Septembre 2011
Langue : français
Type : Texte
-
Géodiversité
9 septembre 2011, par ,
Mis à jour : Août 2018
Langue : français
Type : Texte
Autres articles (40)
-
Des sites réalisés avec MediaSPIP
2 mai 2011, parCette page présente quelques-uns des sites fonctionnant sous MediaSPIP.
Vous pouvez bien entendu ajouter le votre grâce au formulaire en bas de page. -
Ajouter notes et légendes aux images
7 février 2011, parPour pouvoir ajouter notes et légendes aux images, la première étape est d’installer le plugin "Légendes".
Une fois le plugin activé, vous pouvez le configurer dans l’espace de configuration afin de modifier les droits de création / modification et de suppression des notes. Par défaut seuls les administrateurs du site peuvent ajouter des notes aux images.
Modification lors de l’ajout d’un média
Lors de l’ajout d’un média de type "image" un nouveau bouton apparait au dessus de la prévisualisation (...) -
Support audio et vidéo HTML5
10 avril 2011MediaSPIP utilise les balises HTML5 video et audio pour la lecture de documents multimedia en profitant des dernières innovations du W3C supportées par les navigateurs modernes.
Pour les navigateurs plus anciens, le lecteur flash Flowplayer est utilisé.
Le lecteur HTML5 utilisé a été spécifiquement créé pour MediaSPIP : il est complètement modifiable graphiquement pour correspondre à un thème choisi.
Ces technologies permettent de distribuer vidéo et son à la fois sur des ordinateurs conventionnels (...)
Sur d’autres sites (3486)
-
Fast movie creation using MATLAB and ffmpeg
24 février 2018, par hyiltizI have some time series data that I would like to create into movies. The data could be 2D (about 500x10000) or 3D (500x500x10000). For 2D data, the movie frames are simply line plot using
plot, and for 3D data, we can usesurf,imagesc,contouretc. Then we create a video file using these frames in MATLAB, then compress the video file usingffmpeg.To do it fast, one would try not to render all the images to display, nor save the data to disk then read it back again during the process. Usually, one would use
getframeorVideoWriterto create movie in MATLAB, but they seem to easily get tricky if one tries not to display the figures to screen. Some even suggest plotting in hidden figures, then saving them as images to disk as.pngfiles, then compress them usingffmpeg(e.g. withx265encoder into.mp4). However, saving the output ofimagescin my iMac took 3.5s the first time, then 0.5s after. I also find it not fast enough to save so many files to disk only to askffmpegto read them again. One couldhardcopythe data as this suggests, but I am not sure whether it works regardless of the plotting method (e.g.plot,surfetc.), and how one would transfer data over toffmpegwith minimal disk access.This is similiar to this, but
immovieis too slow. This post 3 is similar, but advocates writing images to disk then reading them (slow IO). -
Extracting frames from a video does not work correctly [closed]
13 avril 2024, par Al TilmidhUsing the libraries (libav) and (ffmpeg), I try to extract frames as
.jpgfiles from avideo.mp4, the problem is that my program crashes when I use theCV_8UC3parameter, but by changing this parameter toCV_8UC1, the extracted images end up without color (grayscale), I don't really know what I missed, here is a minimal code to reproduce the two situations :

#include <opencv2></opencv2>opencv.hpp>

extern "C"
{
#include <libavformat></libavformat>avformat.h>
#include <libavcodec></libavcodec>avcodec.h>
}

int main()
{
 AVFormatContext *formatContext = nullptr;

 if (avformat_open_input(&formatContext, "video.mp4", nullptr, nullptr) != 0)
 {
 return -1;
 }

 if (avformat_find_stream_info(formatContext, nullptr) < 0)
 {
 return -1;
 }

 AVPacket packet;
 const AVCodec *codec = nullptr;
 AVCodecContext *codecContext = nullptr;

 int videoStreamIndex = av_find_best_stream(formatContext, AVMEDIA_TYPE_VIDEO, -1, -1, &codec, 0);
 if (videoStreamIndex < 0)
 {
 return -1;
 }

 codecContext = avcodec_alloc_context3(codec);
 avcodec_parameters_to_context(codecContext, formatContext->streams[videoStreamIndex]->codecpar);

 if (avcodec_open2(codecContext, codec, nullptr) < 0)
 {
 return -1;
 }

 AVFrame *frame = av_frame_alloc();

 while (av_read_frame(formatContext, &packet) >= 0)
 {
 if (packet.stream_index == videoStreamIndex)
 {
 int response = avcodec_send_packet(codecContext, &packet);
 
 if (response < 0)
 {
 break;
 }

 while (response >= 0)
 {
 response = avcodec_receive_frame(codecContext, frame);
 if (response == AVERROR(EAGAIN))
 {
 // NO FRAMES
 break;
 }

 else if (response == AVERROR_EOF)
 {
 // END OF FILE
 break;
 }

 else if (response < 0)
 {
 break;
 }

 //cv::Mat frameMat(frame->height, frame->width, CV_8UC3, frame->data[0]); // CV_8UC3 → THE PROGRAM CRASHES
 cv::Mat frameMat(frame->height, frame->width, CV_8UC1, frame->data[0]); // CV_8UC1 → WORK BUT IMAGES ARE IN GRAYSCALE
 cv::imwrite("frame_" + std::to_string(frame->pts) + ".jpg", frameMat);
 }
 }

 av_packet_unref(&packet);
 }

 av_frame_free(&frame);
 avcodec_free_context(&codecContext);
 avformat_close_input(&formatContext);

 return 0;
}


-
How to extract motion vectors from h264 without a full decode on the CPU
25 septembre 2020, par Adrian MayI'm trying to use my nose as a pointing device. The plan is to encode the video stream from a webcam pointed at my face as h264 or the like, get the motion vectors, cook the numbers a bit and chuck them into /dev/uinput to make the mouse pointer move about. The uinput bit was easy.


This has to work with zero discernable latency. This, for instance :


#!/bin/bash
[ -p pipe.mkv ] || mkfifo pipe.mkv
ffmpeg -y -rtbufsize 1M -s 640x360 -vcodec mjpeg -i /dev/video0 -c h264_nvenc pipe.mkv &
ffplay -flags2 +export_mvs -vf codecview=mv=pf+bf+bb pipe.mkv


shows that the vectors are there but with a latency of several seconds which is unusable in a mouse. I know that the first ffmpeg step is working very fast by using the GPU, so either the pipe or the h264 decode in the second step is introducing the latency.


I tried MV Tractus (same as mpegflow I think) in a similar pipe arrangement and it was also very slow. They do a full h264 decode on the CPU and I think that's the problem cos I can see them imposing a lot of load on one CPU. If the pipe had caused the delay by buffering badly then the CPU wouldn't have been loaded. I guess ffplay also did the decoding on the CPU and I couldn't persuade it not to, but it only wants to draw arrows which are no use to me.


I think there are several approaches, and I'd like advice on which would be best, or if there's something even better I don't know about. I could :


- 

- Decode in hardware and get the motion vectors. So far this has failed. I tried combining ffmpeg's
extract_mvs.candhw_decode.csamples but no motion vectors turn up. vdpau is the only decoder I got working on my linux box. I have a nvidia gpu. - Do a minimal parse of the h264 to fish out the motion vectors only, ignoring all the other data. I think this would mean putting some kind of "motion only" option in libav's parser, but I'm not at all familiar with that code.
- Find some other h264 parsing library that has said option and also unpacks the container.
- Forget about hardware accelerated encoding and use a stripped down encoder to make only the motion vectors on either CPU or GPU. I suspect this would be slow cos I think calculating the motion vectors is the hardest part of the algorithm.










I'm tending towards the second option but I need some help figuring out where in the libav code to do it.


- Decode in hardware and get the motion vectors. So far this has failed. I tried combining ffmpeg's
Syndiquer tout le site
