Recherche avancée
Médias (29)
-
#7 Ambience
16 octobre 2011, par
Mis à jour : Juin 2015
Langue : English
Type : Audio
-
#6 Teaser Music
16 octobre 2011, par
Mis à jour : Février 2013
Langue : English
Type : Audio
-
#5 End Title
16 octobre 2011, par
Mis à jour : Février 2013
Langue : English
Type : Audio
-
#3 The Safest Place
16 octobre 2011, par
Mis à jour : Février 2013
Langue : English
Type : Audio
-
#4 Emo Creates
15 octobre 2011, par
Mis à jour : Février 2013
Langue : English
Type : Audio
-
#2 Typewriter Dance
15 octobre 2011, par
Mis à jour : Février 2013
Langue : English
Type : Audio
Autres articles (55)
-
Websites made with MediaSPIP
2 mai 2011, parThis page lists some websites based on MediaSPIP.
-
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 ;
-
Creating farms of unique websites
13 avril 2011, parMediaSPIP platforms can be installed as a farm, with a single "core" hosted on a dedicated server and used by multiple websites.
This allows (among other things) : implementation costs to be shared between several different projects / individuals rapid deployment of multiple unique sites creation of groups of like-minded sites, making it possible to browse media in a more controlled and selective environment than the major "open" (...)
Sur d’autres sites (12302)
-
How to use ffmpeg on hardware acceleration with multiple inputs ?
20 mars 2019, par ColeI’m trying to speed up the rendering of a video by using the GPU instead of the CPU. This code works, but I don’t know if I’m doing it correctly.
ffmpeg -hwaccel cuvid -c:v hevc_cuvid \
-i video.mp4 \
-i logo.png \
-i text.mov \
-c:v h264_nvenc \
-filter_complex " \
[0]scale_npp=1920:1080,hwdownload,format=nv12[bg0]; \
[bg0]trim=0.00:59.460,setpts=PTS-STARTPTS[bg0]; \
[1]scale=150:-1[logo1];[bg0][logo1]overlay=(W-w)-10:(H-h)-10[bg0]; \
[2]scale=500:-1[logo2];[logo2]setpts=PTS-STARTPTS[logo2]; \
[bg0][logo2]overlay=-150:-100[bg0]; \
[bg0]fade=in:00:30,fade=out:1750:30[bg0]" \
-map "[bg0]" -preset fast -y output.mp4I feel like I need to be using hwuplaod somewhere in there, but I’m not totally sure. Any help would be appreciated.
Log from run :
ffmpeg version 4.0 Copyright (c) 2000-2018 the FFmpeg developers
built with gcc 5.4.0 (Ubuntu 5.4.0-6ubuntu1~16.04.11) 20160609
configuration: --enable-gpl --enable-libx264 --enable-cuda --enable-nvenc --enable-cuvid --enable-nonfree --enable-libnpp --extra-cflags=-I/usr/local/cuda/include/ --extra-ldflags=-L/usr/local/cuda/lib64/
libavutil 56. 14.100 / 56. 14.100
libavcodec 58. 18.100 / 58. 18.100
libavformat 58. 12.100 / 58. 12.100
libavdevice 58. 3.100 / 58. 3.100
libavfilter 7. 16.100 / 7. 16.100
libswscale 5. 1.100 / 5. 1.100
libswresample 3. 1.100 / 3. 1.100
libpostproc 55. 1.100 / 55. 1.100
[hevc @ 0x3eb04c0] vps_num_hrd_parameters -1 is invalid
[hevc @ 0x3eb04c0] VPS 0 does not exist
[hevc @ 0x3eb04c0] SPS 0 does not exist.
Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'video.mp4':
Metadata:
major_brand : isom
minor_version : 512
compatible_brands: isomiso2mp41
encoder : Lavf58.7.100
Duration: 00:00:59.46, start: 0.000000, bitrate: 5894 kb/s
Stream #0:0(und): Video: hevc (Main) (hev1 / 0x31766568), yuv420p(tv, progressive), 3840x2160 [SAR 1:1 DAR 16:9], 5891 kb/s, 29.97 fps, 29.97 tbr, 30k tbn, 29.97 tbc (default)
Metadata:
handler_name : VideoHandler
Input #1, png_pipe, from 'logo.png':
Duration: N/A, bitrate: N/A
Stream #1:0: Video: png, rgba(pc), 528x128 [SAR 11339:11339 DAR 33:8], 25 tbr, 25 tbn, 25 tbc
Input #2, mov,mp4,m4a,3gp,3g2,mj2, from 'text.mov':
Metadata:
major_brand : qt
minor_version : 512
compatible_brands: qt
encoder : Lavf57.56.100
Duration: 00:00:06.00, start: 0.000000, bitrate: 1276 kb/s
Stream #2:0(eng): Video: qtrle (rle / 0x20656C72), bgra, 1920x1080, 1274 kb/s, SAR 1:1 DAR 16:9, 25 fps, 25 tbr, 12800 tbn, 12800 tbc (default)
Metadata:
handler_name : DataHandler
Stream mapping:
Stream #0:0 (hevc_cuvid) -> scale_npp
Stream #1:0 (png) -> scale
Stream #2:0 (qtrle) -> scale
fade -> Stream #0:0 (h264_nvenc)
Press [q] to stop, [?] for help
Output #0, mp4, to 'output.mp4':
Metadata:
major_brand : isom
minor_version : 512
compatible_brands: isomiso2mp41
encoder : Lavf58.12.100
Stream #0:0: Video: h264 (h264_nvenc) (Main) (avc1 / 0x31637661), yuv420p, 1920x1080 [SAR 1:1 DAR 16:9], q=-1--1, 2000 kb/s, 29.97 fps, 30k tbn, 29.97 tbc (default)
Metadata:
encoder : Lavc58.18.100 h264_nvenc
Side data:
cpb: bitrate max/min/avg: 0/0/2000000 buffer size: 4000000 vbv_delay: -1
frame= 1783 fps=151 q=30.0 Lsize= 15985kB time=00:00:59.45 bitrate=2202.4kbits/s dup=4 drop=0 speed=5.04x
video:15977kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 0.050389%Not sure what to make of this, pretty new to ffmpeg.
-
Lost in converting hardware accelarated ffmpeg decoder to raw loopback device
5 septembre 2019, par MikeynlBefore reaching out, i googled enough, tried many different docker containers, building, compiling etc etc.
What I am looking for is a hardware accelerated conversion from a rtsp stream to a /dev/video0 device. I have a working v4l2loopback kernel module, and the following command is working :
ffmpeg -loglevel panic -hide_banner -i "rtsp://192.168.0.17/user=admin&password=&channel=1&stream=0.dsp?real_stream" -f v4l2 -pix_fmt yuv420p /dev/video0I can test the /dev/video0 device to take a screenshot :
ffmpeg -f video4linux2 -i /dev/video0 -ss 0:0:2 -frames 1 /var/www/html/out.pngAbove is working, but taking around 30 / 50% cpu usage to decode/encoding.
I have a fully working test environment with a GeForce GTX 1050. All cuda / nvidia related drivers are in place. ffmpeg is compiled with the following options :
configuration: --enable-nonfree --disable-shared --enable-nvenc --enable-cuda --enable-cuvid --enable-libnpp --extra-cflags=-I/usr/local/cuda/include --extra-cflags=-I/usr/local/include --extra-ldflags=-L/usr/local/cuda/lib64My last attempt was :
ffmpeg -hwaccel cuvid -c:v h264_cuvid -i "rtsp://192.168.0.17/user=admin&password=&channel=1&stream=0.dsp?real_stream" -c:v rawvideo -pix_fmt yuv420p -f v4l2 /dev/video0It gives me an error :
Impossible to convert between the formats supported by the filter ’Parsed_null_0’ and the filter ’auto_scaler_0’
Error reinitializing filters !
Failed to inject frame into filter network : Function not implemented
Error while processing the decoded data for stream #0:0I have at the moment absolute no idea anymore how to solve this.
Thanks in advance !
//added information
root@localhost:~# ffmpeg -i "rtsp://192.168.0.17/user=admin&password=&channel=1&stream=0.dsp?real_stream" -c copy -f v4l2 /dev/video0
ffmpeg version N-91067-g1c2e5fc Copyright (c) 2000-2018 the FFmpeg developers
built with gcc 5.4.0 (Ubuntu 5.4.0-6ubuntu1~16.04.9) 20160609
configuration: --enable-nonfree --disable-shared --enable-nvenc --enable-cuda --enable-cuvid --enable-libnpp --extra-cflags=-I/usr/local/cuda/include --extra-cflags=-I/usr/local/include --extra-ldflags=-L/usr/local/cuda/lib64
libavutil 56. 18.102 / 56. 18.102
libavcodec 58. 19.101 / 58. 19.101
libavformat 58. 13.102 / 58. 13.102
libavdevice 58. 4.100 / 58. 4.100
libavfilter 7. 22.100 / 7. 22.100
libswscale 5. 2.100 / 5. 2.100
libswresample 3. 2.100 / 3. 2.100
Input #0, rtsp, from 'rtsp://192.168.0.17/user=admin&password=&channel=1&stream=0.dsp?real_stream':
Metadata:
title : RTSP Session
Duration: N/A, start: 2.300000, bitrate: N/A
Stream #0:0: Video: h264 (Main), yuvj420p(pc, bt709, progressive), 1920x1080, 10 fps, 10 tbr, 90k tbn, 20 tbc
[v4l2 @ 0x31a7f00] V4L2 output device supports only a single raw video stream
Could not write header for output file #0 (incorrect codec parameters ?): Invalid argument
Stream mapping:
Stream #0:0 -> #0:0 (copy)
Last message repeated 1 times
root@localhost:~#
root@localhost:~# ffmpeg -i "rtsp://192.168.0.17/user=admin&password=&channel=1&stream=0.dsp?real_stream" -f v4l2 -pix_fmt yuv420p /dev/video0 ffmpeg version N-91067-g1c2e5fc Copyright (c) 2000-2018 the FFmpeg developers
built with gcc 5.4.0 (Ubuntu 5.4.0-6ubuntu1~16.04.9) 20160609
configuration: --enable-nonfree --disable-shared --enable-nvenc --enable-cuda --enable-cuvid --enable-libnpp --extra-cflags=-I/usr/local/cuda/include --extra-cflags=-I/usr/local/include --extra-ldflags=-L/usr/local/cuda/lib64
libavutil 56. 18.102 / 56. 18.102
libavcodec 58. 19.101 / 58. 19.101
libavformat 58. 13.102 / 58. 13.102
libavdevice 58. 4.100 / 58. 4.100
libavfilter 7. 22.100 / 7. 22.100
libswscale 5. 2.100 / 5. 2.100
libswresample 3. 2.100 / 3. 2.100
Input #0, rtsp, from 'rtsp://192.168.0.17/user=admin&password=&channel=1&stream=0.dsp?real_stream':
Metadata:
title : RTSP Session
Duration: N/A, start: 2.300000, bitrate: N/A
Stream #0:0: Video: h264 (Main), yuvj420p(pc, bt709, progressive), 1920x1080, 10 fps, 10 tbr, 90k tbn, 20 tbc
Stream mapping:
Stream #0:0 -> #0:0 (h264 (native) -> rawvideo (native))
Press [q] to stop, [?] for help
[swscaler @ 0x2821240] deprecated pixel format used, make sure you did set range correctly
Output #0, v4l2, to '/dev/video0':
Metadata:
title : RTSP Session
encoder : Lavf58.13.102
Stream #0:0: Video: rawvideo (I420 / 0x30323449), yuv420p, 1920x1080, q=2-31, 248832 kb/s, 10 fps, 10 tbn, 10 tbc
Metadata:
encoder : Lavc58.19.101 rawvideo
frame= 85 fps= 13 q=-0.0 Lsize=N/A time=00:00:08.50 bitrate=N/A dup=0 drop=4 speed=1.31x
video:258188kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: unknown -
Hardware accelerated decoding with FFmpeg falls back to software decoding
9 février 2024, par iexavSo I have followed the FFmpeg example for hardware accelerated decoding exactly as it is (I am referring to this example).


https://github.com/FFmpeg/FFmpeg/blob/master/doc/examples/hw_decode.c#L76


But I still seem to be decoding with the software decoder. When I open the task manager on windows the GPU isn't getting used. Before I make a call to av_hwframe_transfer_data() I check whether the frame is in the relevant hw_pix_fmt format and it is. Everything works, no errors, nothing, except the GPU is doing nothing. Now as an example I tried decoding a video that uses vp9 as a codec. If I specify the hardware accelerated codec I want by name it actually does work.


vidCodec=avcodec_find_decoder_by_name("vp9_cuvid"); 


When I do this and look at the task manager I can see that the CPU does much less work and my GPU actually does Video Decode work. Having to specify the hardware accelerated decoder for every single video I am decoding is ridiculous though.


Edit : as per user4581301's answer, here are the pieces of relevant code. (It's actually in java because I am using the java FFmpeg wrapper but it's basically just making a bunch of calls to FFmpeg functions.)



 ArrayList<string> deviceTypes = new ArrayList<>();
 int type = AV_HWDEVICE_TYPE_NONE;
 while ((type = av_hwdevice_iterate_types(type)) != AV_HWDEVICE_TYPE_NONE) {
 BytePointer p = av_hwdevice_get_type_name(type);

 deviceTypes.add(CString(p));
 }
 boolean flag=false;

 for(int j=0;j* Allocate a codec context for the decoder */
 if ((video_c = avcodec_alloc_context3(vidCodec)) == null) {
 throw new Exception("avcodec_alloc_context3() error: Could not allocate video decoding context.");
 }


 /* copy the stream parameters from the muxer */

 if ((ret = avcodec_parameters_to_context(video_c, video_st.codecpar())) < 0) {
 releaseUnsafe();
 throw new Exception("avcodec_parameters_to_context() error " + ret + ": Could not copy the video stream parameters.");
 }
 

 video_c.get_format(AvFormatGetter.getInstance());
 AVBufferRef hardwareDeviceContext =av_hwdevice_ctx_alloc(type);

 if ((ret = av_hwdevice_ctx_create(hardwareDeviceContext, type,(String) null, null, 0)) < 0) {
 System.err.println("Failed to create specified HW device. error " + ret);

 }else{
 video_c.hw_device_ctx(av_buffer_ref(hardwareDeviceContext));

 }
 


 
//The function that gets called for get_format
@Override
 public int call(AVCodecContext context, IntPointer format) {
 int p;


 for (int i=0;;i++) {
 if ((p=format.get(i)) == hw_pix_fmt) {
 return p;
 }
 if(p==-1){
 break;
 }
 }

 System.out.println(hw_pix_fmt +" is not found in the codec context");
 // Error

 return AV_PIX_FMT_NONE;
 }
 
//The method that's used for decoding video frames
 public Optional<boolean> decodeVideoFrame(AVPacket pkt, boolean readPacket, boolean keyFrames) throws Exception {

 int ret;
 // Decode video frame
 if (readPacket) {
 ret = avcodec_send_packet(video_c, pkt);
 
 if (ret < 0) {
 System.out.println("error during decoding");
 return Optional.empty();
 }

 if (pkt.data() == null && pkt.size() == 0) {
 pkt.stream_index(-1);
 }
 
 }

 // Did we get a video frame?
 while (true) {
 ret = avcodec_receive_frame(video_c, picture_hw);

 if (ret == AVERROR_EAGAIN() || ret == AVERROR_EOF()) {
 if (pkt.data() == null && pkt.size() == 0) {
 return Optional.empty();
 } else {

 return Optional.of(true);

 }
 } else if (ret < 0) {

 // Ignore errors to emulate the behavior of the old API
 // throw new Exception("avcodec_receive_frame() error " + ret + ": Error during video decoding.");
 return Optional.of(true);

 }

 if (!keyFrames || picture.pict_type() == AV_PICTURE_TYPE_I) {
 
 if(picture_hw.format()==hw_pix_fmt){
 if (av_hwframe_transfer_data(
 picture, // The frame that will contain the usable data.
 picture_hw, // Frame returned by avcodec_receive_frame()
 0) < 0) {
 throw new Exception("Could not transfer data from gpu to cpu. ");

 }
 }
 //... The rest of the method here
 return Optional.of(false);

 }
 }
 }
</boolean></string>

Syndiquer tout le site
