Recherche avancée

Sur d’autres sites (10048)

• How can I remove every nth frame from raw video using ffmpeg ?

  15 juin 2024, par DaveB44

  Question moved to SuperUser, please reply there

  


  I have many cine films that have been digitally converted. They have been converted as a 25 fps with 2 frames in every 23 duplicated. I need to remove the duplicated frames using a bitstream filter so there is no decoding/encoding.

  


  After removing the frames I will change to the original cine frame rate of 18 fps. I will then change the frame rate to 25 fps using duplicated frames. I don't want to interpolate as I want to preserve the jerky format of the cine. Finally I will video editing software that will do the final encoding.

  


  All the additional steps will cause several stages of decoding/encoding, that I need to avoid.

  


  First I used a combination of the answers to FFmpeg remove video frames at specific intervals and FFmpeg remove every 6th frame starting from the 3rd frame, on an mp4 file to check it works. I modified it to remove every 4th and 17th frame in 25.

  


  ffmpeg -loglevel warning -i cine.mp4 -an -vf "select='if((mod(n-4,25)),(mod(n-17,25)))',setpts='N/FRAME_RATE/TB'" cine-23.mp4


  


  This works fine, but I end up with a file a quarter the size.

  


  I now used Gyan's answer in Using ffmpeg to change framerate to extract the raw bitstream.

  


  ffmpeg -loglevel warning -i cine.mp4 -c copy -f h264 cine.h264


  


  This created the h264 file as expected, surprisingly it was 16 kB smaller than the original (67 GB file size).

  


  I then modified the code to use the .h264 file.

  


  ffmpeg -loglevel warning -i cine.h264 -vf "select='if((mod(n-4,25)),(mod(n-17,25)))',setpts='N/FRAME_RATE/TB'" cine-23.h264


  


  This gave the following error, but created the cine-23.h264 file although it was the same size as cine-23.mp4 in the test above

  


  [h264 @ 00000245ec0bfb80] non-existing SPS 0 referenced in buffering period
    Last message repeated 1 times


  


  I then checked the ffmpeg bitstream filter documentation and found the bitstream filter setts. I changed my code to the following.

  


  ffmpeg -loglevel warning -i cine.h264 -bsf:v "select='if((mod(n-4,25)),(mod(n-17,25)))',setts=pts='N/FRAME_RATE/TB'" cine-23.h264


  


  Which resulted in the following error.

  


  [vost#0:0/libx264 @ 000002916cf173c0] Error parsing bitstream filter sequence 'select='if((mod(n-4,25)),(mod(n-17,25)))',setts=pts='N/FRAME_RATE/TB'': Bitstream filter not found
Error opening output file cine-23.h264.
Error opening output files: Bitstream filter not found


  


  I'm assuming the error is because setts does not support select. Is there another way to achieve what I am looking for ?

  


  I could use mpdecimate but as that has to compare each frame it is much slower than defining it only needs to delete frames 4 and 17 in every second.

  


  This is the output of ffprobe on my original file.

  


  ffprobe version 2023-11-28-git-47e214245b-full_build-www.gyan.dev Copyright (c) 2007-2023 the FFmpeg developers
  built with gcc 12.2.0 (Rev10, Built by MSYS2 project)
  configuration: --enable-gpl --enable-version3 --enable-static --pkg-config=pkgconf --disable-w32threads --disable-autodetect --enable-fontconfig --enable-iconv --enable-gnutls --enable-libxml2 --enable-gmp --enable-bzlib --enable-lzma --enable-libsnappy --enable-zlib --enable-librist --enable-libsrt --enable-libssh --enable-libzmq --enable-avisynth --enable-libbluray --enable-libcaca --enable-sdl2 --enable-libaribb24 --enable-libaribcaption --enable-libdav1d --enable-libdavs2 --enable-libuavs3d --enable-libzvbi --enable-librav1e --enable-libsvtav1 --enable-libwebp --enable-libx264 --enable-libx265 --enable-libxavs2 --enable-libxvid --enable-libaom --enable-libjxl --enable-libopenjpeg --enable-libvpx --enable-mediafoundation --enable-libass --enable-frei0r --enable-libfreetype --enable-libfribidi --enable-libharfbuzz --enable-liblensfun --enable-libvidstab --enable-libvmaf --enable-libzimg --enable-amf --enable-cuda-llvm --enable-cuvid --enable-ffnvcodec --enable-nvdec --enable-nvenc --enable-dxva2 --enable-d3d11va --enable-libvpl --enable-libshaderc --enable-vulkan --enable-libplacebo --enable-opencl --enable-libcdio --enable-libgme --enable-libmodplug --enable-libopenmpt --enable-libopencore-amrwb --enable-libmp3lame --enable-libshine --enable-libtheora --enable-libtwolame --enable-libvo-amrwbenc --enable-libcodec2 --enable-libilbc --enable-libgsm --enable-libopencore-amrnb --enable-libopus --enable-libspeex --enable-libvorbis --enable-ladspa --enable-libbs2b --enable-libflite --enable-libmysofa --enable-librubberband --enable-libsoxr --enable-chromaprint
  libavutil      58. 32.100 / 58. 32.100
  libavcodec     60. 35.100 / 60. 35.100
  libavformat    60. 18.100 / 60. 18.100
  libavdevice    60.  4.100 / 60.  4.100
  libavfilter     9. 14.100 /  9. 14.100
  libswscale      7.  6.100 /  7.  6.100
  libswresample   4. 13.100 /  4. 13.100
  libpostproc    57.  4.100 / 57.  4.100
Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'cine.mp4':
  Metadata:
    major_brand     : mp42
    minor_version   : 19529864
    compatible_brands: mp42isom
    creation_time   : 2024-02-19T21:01:10.000000Z
  Duration: 00:01:00.00, start: 0.000000, bitrate: 9245 kb/s
  Stream #0:0[0x1](eng): Video: h264 (High) (avc1 / 0x31637661), yuv420p(progressive), 720x576 [SAR 35:32 DAR 175:128], 9243 kb/s, 25 fps, 25 tbr, 25 tbn (default)
    Metadata:
      creation_time   : 2024-02-19T21:01:10.000000Z
      handler_name    : Video Media Handler
      vendor_id       : [0][0][0][0]
      encoder         : AVC Coding


  


• FFMPEG conversion failed at random times on live stream to youtube

  3 avril 2022, par good karma

  I am running a bash script using ubuntu server on arm64 that live streams video and audio to youtube, heres the script below

  


  FFMPEG is the lateset version available in the ubuntu repo

  


  #! /bin/bash

VBR="8000k"
FPS="24"
QUAL="superfast"

YOUTUBE_URL="rtmp://a.rtmp.youtube.com/live2"
KEY="****"

VIDEO_SOURCE="/mnt/disk1/test/****"

AUDIO_SOURCE="****"

ffmpeg \
    -re -f lavfi -i "movie=filename=$VIDEO_SOURCE:loop=0, setpts=N/(FRAME_RATE*TB)" \
    -thread_queue_size 512 -i "$AUDIO_SOURCE" \
    -map 0:v:0 -map 1:a:0 \
    -map_metadata:g 1:g \
    -vcodec libx264 -pix_fmt yuv420p -preset $QUAL -r $FPS -g $(($FPS * 2)) -b:v $VBR \
    -acodec libmp3lame -ar 44100 -threads 4 -qscale:v 3 -b:a 320000 -bufsize 512k \
    -report -loglevel fatal -nostats  \
    -f flv "$YOUTUBE_URL/$KEY"


  


  The stream works fine, however after a seemingly random period of time, I get a conversion failed error

  


  first after 64 hours of streaming
second after 12 hours of streaming
third after 8 hours of streaming

  


  The server itsef is fine cpu and memory usage are extremly low and there are no spikes

  


  from what i can guess the error has something to do with libx264 judging by the error logs, although I do not really know. After some guidance on what to do try next. Here is the tail end of the ffmpeg log report

  


  [libx264 @ 0xaaab23ad3e30] frame=1351975 QP=16.89 NAL=2 Slice:P Poc:14  I:6461 P:1224 SKIP:475  size=26602 bytes
av_interleaved_write_frame(): Broken pipe
No more output streams to write to, finishing.
[libx264 @ 0xaaab23ad3e30] frame=1351976 QP=17.90 NAL=2 Slice:B Poc:10  I:2395 P:3322 SKIP:2443 size=20127 bytes
[libx264 @ 0xaaab23ad3e30] frame=1351977 QP=17.89 NAL=0 Slice:B Poc:12  I:1848 P:3235 SKIP:3077 size=17313 bytes
[libx264 @ 0xaaab23ad3e30] frame=1351978 QP=15.65 NAL=2 Slice:P Poc:22  I:6458 P:1171 SKIP:531  size=26073 bytes
[libx264 @ 0xaaab23ad3e30] frame=1351979 QP=17.19 NAL=2 Slice:B Poc:18  I:2766 P:3514 SKIP:1880 size=19633 bytes
[libx264 @ 0xaaab23ad3e30] frame=1351980 QP=18.35 NAL=0 Slice:B Poc:16  I:1233 P:2876 SKIP:4051 size=11769 bytes
[libx264 @ 0xaaab23ad3e30] frame=1351981 QP=17.28 NAL=0 Slice:B Poc:20  I:822  P:3197 SKIP:4141 size=10699 bytes
[libx264 @ 0xaaab23ad3e30] frame=1351982 QP=17.08 NAL=2 Slice:P Poc:30  I:6351 P:1443 SKIP:366  size=29120 bytes
[libx264 @ 0xaaab23ad3e30] frame=1351983 QP=17.29 NAL=2 Slice:B Poc:26  I:2060 P:4211 SKIP:1889 size=18465 bytes
[libx264 @ 0xaaab23ad3e30] frame=1351984 QP=18.00 NAL=0 Slice:B Poc:24  I:472  P:3225 SKIP:4463 size=8893 bytes
[libx264 @ 0xaaab23ad3e30] frame=1351985 QP=17.98 NAL=0 Slice:B Poc:28  I:1339 P:3681 SKIP:3140 size=13378 bytes
[libx264 @ 0xaaab23ad3e30] frame=1351986 QP=15.57 NAL=2 Slice:P Poc:32  I:6270 P:1584 SKIP:306  size=33703 bytes
av_interleaved_write_frame(): Broken pipe
[libmp3lame @ 0xaaab23ac06f0] Trying to remove 47 more samples than there are in the queue
[flv @ 0xaaab23ad2ff0] Failed to update header with correct duration.
[flv @ 0xaaab23ad2ff0] Failed to update header with correct filesize.
Error writing trailer of rtmp://a.rtmp.youtube.com/live2/****: Broken pipe
frame=1351987 fps= 24 q=22.0 Lsize=37296521kB time=15:38:52.22 bitrate=5423.8kbits/s dup=0 drop=56332 speed=   1x    
video:35035969kB audio:2200479kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 0.161329%
Input file #0 (movie=filename=/mnt/disk1/test/****.mp4:loop=0, setpts=N/(FRAME_RATE*TB)):
  Input stream #0:0 (video): 1408319 packets read (4380435417600 bytes); 1408319 frames decoded; 
  Total: 1408319 packets (4380435417600 bytes) demuxed
Input file #1 (https://****.mp3):
  Input stream #1:0 (audio): 2156470 packets read (901316441 bytes); 2156470 frames decoded (2484253440 samples); 
  Total: 2156470 packets (901316441 bytes) demuxed
Output file #0 (rtmp://a.rtmp.youtube.com/live2/****):
  Output stream #0:0 (video): 1351987 frames encoded; 1351976 packets muxed (35876832486 bytes); 
  Output stream #0:1 (audio): 2156470 frames encoded (2484253440 samples); 2156469 packets muxed (2253290057 bytes); 
  Total: 3508445 packets (38130122543 bytes) muxed
3564789 frames successfully decoded, 0 decoding errors
[AVIOContext @ 0xaaab23ac4ef0] Statistics: 0 seeks, 3823215 writeouts
[rtmp @ 0xaaab23dab6a0] UnPublishing stream...
[rtmp @ 0xaaab23dab6a0] Deleting stream...
[libx264 @ 0xaaab23ad3e30] frame I:53166 Avg QP:16.31  size: 39514
[libx264 @ 0xaaab23ad3e30] frame P:500115 Avg QP:17.79  size: 32831
[libx264 @ 0xaaab23ad3e30] frame B:798706 Avg QP:19.67  size: 21732
[libx264 @ 0xaaab23ad3e30] consecutive B-frames: 13.9% 15.2% 20.2% 50.7%
[libx264 @ 0xaaab23ad3e30] mb I  I16..4: 45.9% 44.7%  9.4%
[libx264 @ 0xaaab23ad3e30] mb P  I16..4: 40.6% 37.1%  2.2%  P16..4: 16.5%  0.0%  0.0%  0.0%  0.0%    skip: 3.7%
[libx264 @ 0xaaab23ad3e30] mb B  I16..4: 16.3% 16.2%  0.2%  B16..8: 25.0%  0.0%  0.0%  direct:19.7%  skip:22.5%  L0:48.7% L1:48.4% BI: 2.9%
[libx264 @ 0xaaab23ad3e30] 8x8 transform intra:47.5% inter:31.9%
[libx264 @ 0xaaab23ad3e30] coded y,uvDC,uvAC intra: 15.8% 78.7% 25.0% inter: 2.0% 60.3% 1.8%
[libx264 @ 0xaaab23ad3e30] i16 v,h,dc,p: 50% 20%  9% 21%
[libx264 @ 0xaaab23ad3e30] i8 v,h,dc,ddl,ddr,vr,hd,vl,hu: 19% 16% 31%  4%  7%  5%  5%  6%  6%
[libx264 @ 0xaaab23ad3e30] i4 v,h,dc,ddl,ddr,vr,hd,vl,hu: 36% 20% 33%  2%  2%  2%  1%  2%  2%
[libx264 @ 0xaaab23ad3e30] i8c dc,h,v,p: 29% 18% 18% 35%
[libx264 @ 0xaaab23ad3e30] Weighted P-Frames: Y:3.0% UV:2.2%
[libx264 @ 0xaaab23ad3e30] kb/s:5095.01
[AVIOContext @ 0xaaab22460eb0] Statistics: 42281852862 bytes read, 2731 seeks
[AVIOContext @ 0xaaab23ac4de0] Statistics: 901390389 bytes read, 0 seeks
Conversion failed!


  


• Method For Crawling Google

  28 mai 2011, par Multimedia Mike — Big Data

  I wanted to crawl Google in order to harvest a large corpus of certain types of data as yielded by a certain search term (we’ll call it “term” for this exercise). Google doesn’t appear to offer any API to automatically harvest their search results (why would they ?). So I sat down and thought about how to do it. This is the solution I came up with.

  
  
  

  FAQ
  Q : Is this legal / ethical / compliant with Google’s terms of service ?
  A : Does it look like I care ? Moving right along…

  Manual Crawling Process
  For this exercise, I essentially automated the task that would be performed by a human. It goes something like this :

  1. Search for “term”
  2. On the first page of results, download each of the 10 results returned
  3. Click on the next page of results
  4. Go to step 2, until Google doesn’t return anymore pages of search results

  Google returns up to 1000 results for a given search term. Fetching them 10 at a time is less than efficient. Fortunately, the search URL can easily be tweaked to return up to 100 results per page.

  Expanding Reach
  Problem : 1000 results for the “term” search isn’t that many. I need a way to expand the search. I’m not aiming for relevancy ; I’m just searching for random examples of some data that occurs around the internet.

  My solution for this is to refine the search using the “site” wildcard. For example, you can ask Google to search for “term” at all Canadian domains using “site :.ca”. So, the manual process now involves harvesting up to 1000 results for every single internet top level domain (TLD). But many TLDs can be more granular than that. For example, there are 50 sub-domains under .us, one for each state (e.g., .ca.us, .ny.us). Those all need to be searched independently. Same for all the sub-domains under TLDs which don’t allow domains under the main TLD, such as .uk (search under .co.uk, .ac.uk, etc.).

  Another extension is to combine “term” searches with other terms that are likely to have a rich correlation with “term”. For example, if “term” is relevant to various scientific fields, search for “term” in conjunction with various scientific disciplines.

  Algorithmically
  My solution is to create an SQLite database that contains a table of search seeds. Each seed is essentially a “site :” string combined with a starting index.

  Each TLD and sub-TLD is inserted as a searchseed record with a starting index of 0.

  A script performs the following crawling algorithm :

  • Fetch the next record from the searchseed table which has not been crawled
  • Fetch search result page from Google
  • Scrape URLs from page and insert each into URL table
  • Mark the searchseed record as having been crawled
  • If the results page indicates there are more results for this search, insert a new searchseed for the same seed but with a starting index 100 higher

  Digging Into Sites
  Sometimes, Google notes that certain sites are particularly rich sources of “term” and offers to let you search that site for “term”. This basically links to another search for ‘term site:somesite”. That site gets its own search seed and the program might harvest up to 1000 URLs from that site alone.

  Harvesting the Data
  Armed with a database of URLs, employ the following algorithm :

  • Fetch a random URL from the database which has yet to be downloaded
  • Try to download it
  • For goodness sake, have a mechanism in place to detect whether the download process has stalled and automatically kill it after a certain period of time
  • Store the data and update the database, noting where the information was stored and that it is already downloaded

  This step is easy to parallelize by simply executing multiple copies of the script. It is useful to update the URL table to indicate that one process is already trying to download a URL so multiple processes don’t duplicate work.

  Acting Human
  A few factors here :

  • Google allegedly doesn’t like automated programs crawling its search results. Thus, at the very least, don’t let your script advertise itself as an automated program. At a basic level, this means forging the User-Agent : HTTP header. By default, Python’s urllib2 will identify itself as a programming language. Change this to a well-known browser string.
  • Be patient ; don’t fire off these search requests as quickly as possible. My crawling algorithm inserts a random delay of a few seconds in between each request. This can still yield hundreds of useful URLs per minute.
  • On harvesting the data : Even though you can parallelize this and download data as quickly as your connection can handle, it’s a good idea to randomize the URLs. If you hypothetically had 4 download processes running at once and they got to a point in the URL table which had many URLs from a single site, the server might be configured to reject too many simultaneous requests from a single client.

  Conclusion
  Anyway, that’s just the way I would (and did) do it. What did I do with all the data ? That’s a subject for a different post.

  Adorable spider drawing from here.