Recherche avancée

Sur d’autres sites (4366)

• Resizing AVIF images with transparency with FFmpeg [closed]

  4 octobre 2024, par Calebmer

  I'm trying to resize an image with transparency with FFmpeg, however the output looks to only be a resized version of the alpha layer.

  


  When I try to do a noop transform of the AVIF image with an alpha layer :

  


  ffmpeg -i input.avif output.avif


  


  output.avif appears to be the alpha layer with black representing alpha 0 and white representing alpha 1.

  


  ffprobe input.avif gives me :

  


  ffprobe version 7.0.2 Copyright (c) 2007-2024 the FFmpeg developers
  built with Apple clang version 15.0.0 (clang-1500.3.9.4)
  configuration: --prefix=/opt/homebrew/Cellar/ffmpeg/7.0.2 --enable-shared --enable-pthreads --enable-version3 --cc=clang --host-cflags= --host-ldflags='-Wl,-ld_classic' --enable-ffplay --enable-gnutls --enable-gpl --enable-libaom --enable-libaribb24 --enable-libbluray --enable-libdav1d --enable-libharfbuzz --enable-libjxl --enable-libmp3lame --enable-libopus --enable-librav1e --enable-librist --enable-librubberband --enable-libsnappy --enable-libsrt --enable-libssh --enable-libsvtav1 --enable-libtesseract --enable-libtheora --enable-libvidstab --enable-libvmaf --enable-libvorbis --enable-libvpx --enable-libwebp --enable-libx264 --enable-libx265 --enable-libxml2 --enable-libxvid --enable-lzma --enable-libfontconfig --enable-libfreetype --enable-frei0r --enable-libass --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libopenjpeg --enable-libspeex --enable-libsoxr --enable-libzmq --enable-libzimg --disable-libjack --disable-indev=jack --enable-videotoolbox --enable-audiotoolbox --enable-neon
  libavutil      59.  8.100 / 59.  8.100
  libavcodec     61.  3.100 / 61.  3.100
  libavformat    61.  1.100 / 61.  1.100
  libavdevice    61.  1.100 / 61.  1.100
  libavfilter    10.  1.100 / 10.  1.100
  libswscale      8.  1.100 /  8.  1.100
  libswresample   5.  1.100 /  5.  1.100
  libpostproc    58.  1.100 / 58.  1.100
Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'input.avif':
  Metadata:
    major_brand     : avif
    minor_version   : 0
    compatible_brands: avifmif1miaf
  Duration: N/A, start: 0.000000, bitrate: N/A
  Stream #0:0[0x1]: Video: av1 (libdav1d) (Main) (av01 / 0x31307661), gray(pc), 336x252 [SAR 1:1 DAR 4:3], 1 fps, 1 tbr, 1 tbn (default)
  Stream #0:1[0x2]: Video: av1 (libdav1d) (High) (av01 / 0x31307661), yuv444p(pc, smpte170m/bt709/iec61966-2-1), 336x252 [SAR 1:1 DAR 4:3], 1 fps, 1 tbr, 1 tbn


  


  Seeing there are two streams (the first stream being gray(pc), probably the alpha layer) I next tried :

  


  ffmpeg -i input.avif -map 0:v:1 output.avif


  


  To see the second stream and it gave me the image without any alpha channel. Transparent pixels were black.

  


  Ultimately I want to resize the AVIF file with ffmpeg -i input.avif -vf "scale=iw/2:-1" output.avif but that appears to only resize the greyscale alpha channel. Furthermore, this will be part of a script that operates on some AVIF files without an alpha channel and some AVIF files with an alpha channel and I don't know which files have an alpha channel ahead of time. ffmpeg -i input.avif -vf "scale=iw/2:-1" output.avif works for files without an alpha channel.

  


• ffmpeg minimal linux build with only one filter [closed]

  31 juillet 2024, par at8993

  I am aiming to compile a minimal build of ffmpeg for Ubuntu 22.04. The application requires only the concat filter for .mp4s using H.264 video encoder.

  


  I have followed this guide and used the following commands to include libx264 only :

  


  sudo apt-get update -qq && sudo apt-get -y install \
  autoconf \
  automake \
  build-essential \
  cmake \
  git-core \
  libass-dev \
  libfreetype6-dev \
  libgnutls28-dev \
  libmp3lame-dev \
  libsdl2-dev \
  libtool \
  libva-dev \
  libvdpau-dev \
  libvorbis-dev \
  libxcb1-dev \
  libxcb-shm0-dev \
  libxcb-xfixes0-dev \
  meson \
  ninja-build \
  pkg-config \
  texinfo \
  wget \
  yasm \
  zlib1g-dev

mkdir -p ~/ffmpeg_sources ~/bin

sudo apt-get install libx264-dev


  


  and for the configure command I have used the following options with a view to only enable what's necessary for concat.

  


  cd ~/ffmpeg_sources && \
wget -O ffmpeg-snapshot.tar.bz2 https://ffmpeg.org/releases/ffmpeg-snapshot.tar.bz2 && \
tar xjvf ffmpeg-snapshot.tar.bz2 && \
cd ffmpeg && \

PATH="$HOME/bin:$PATH" PKG_CONFIG_PATH="$HOME/ffmpeg_build/lib/pkgconfig" ./configure \
  --prefix="$HOME/ffmpeg_build" \
  --pkg-config-flags="--static" \
  --extra-cflags="-I$HOME/ffmpeg_build/include" \
  --extra-ldflags="-L$HOME/ffmpeg_build/lib" \
  --extra-libs="-lpthread -lm" \
  --ld="g++" \
  --bindir="$HOME/bin" \
  --disable-everything \
  --enable-filter=concat \
  --enable-avformat \
  --enable-avdevice \
  --enable-avcodec \
  --enable-decoder=h264 
  --enable-libx264 
  --enable-muxer=mp4 
  --enable-gpl

PATH="$HOME/bin:$PATH" make && \
make install && \
hash -r


  


  However running

  


  ffmpeg -f concat -i files_to_merge.txt -c copy output.mp4


  


  returns error

  


  ffmpeg -f concat -i files_to_merge.txt -c copy output.mp4
ffmpeg version N-116445-gb1d410716b Copyright (c) 2000-2024 the FFmpeg developers
  built with gcc 11 (Ubuntu 11.4.0-1ubuntu1~22.04)
  configuration: --prefix=/home/user/ffmpeg_build 
--extra-cflags=-I/home/user/ffmpeg_build/include --extra-ldflags=-L/home/user/ffmpeg_build/lib 
--extra-libs='-lpthread -lm' 
--ld=g++ --bindir=/home/user/bin --disable-everything 
--enable-filter=concat --enable-avformat 
--enable-avdevice --enable-avcodec --enable-decoder=h264
 --enable-libx264 --enable-muxer=mp4 
--enable-gpl
  libavutil      59. 30.100 / 59. 30.100
  libavcodec     61. 10.100 / 61. 10.100
  libavformat    61.  5.101 / 61.  5.101
  libavdevice    61.  2.100 / 61.  2.100
  libavfilter    10.  2.102 / 10.  2.102
  libswscale      8.  2.100 /  8.  2.100
  libswresample   5.  2.100 /  5.  2.100
  libpostproc    58.  2.100 / 58.  2.100
[in#0 @ 0x555605fab640] Unknown input format: 'concat'


  


  I assume this is due to the omission of an enable option in the configure command.

  


  Thanks !

  


• How to reinsert edited metadata stream information from the FFMETADATAFILE file ? [closed]

  6 septembre 2024, par SENYCH

  I'm working on simplifying and speeding up the process of editing video metadata for user convenience. I've successfully edited metadata streams using console commands, such as :

  


  ffmpeg -i INPUT.mp4 -map 0 -metadata:s:0 "handler_name=An other video" -metadata:s:1 "handler_name=An other audio recording in russian" -metadata:s:2 "handler_name=An other audio recording in english" -metadata:s:3 "handler_name=An other audio recording in japanese" -c copy OUTPUT.mp4


  


  However, I'd like to accomplish this through a ffmetadata file. Here's the approach I've taken :

  


  ffmpeg -t 0 -i INPUT.mp4 -map 0 -c copy -f ffmetadata ffmetadata.txt -hide_banner


  


  Original ffmetadata.txt is :

  


  ;FFMETADATA1
major_brand=isom
minor_version=512
compatible_brands=isomiso2avc1mp41
encoder=Lavf61.5.101
[STREAM]
language=und
handler_name=The best video
vendor_id=[0][0][0][0]
[STREAM]
language=rus
handler_name=The best russian language
vendor_id=[0][0][0][0]
[STREAM]
language=eng
handler_name=The best english language
vendor_id=[0][0][0][0]
[STREAM]
language=jpn
handler_name=The best japanese language
vendor_id=[0][0][0][0]


  


  Edit the ffmetadata.txt file to update the handler_name values :

  


  ;FFMETADATA1
major_brand=isom
minor_version=512
compatible_brands=isomiso2avc1mp41
encoder=Lavf61.5.101
[STREAM]
language=und
handler_name=An other video
vendor_id=[0][0][0][0]
[STREAM]
language=rus
handler_name=An other audio recording in russian
vendor_id=[0][0][0][0]
[STREAM]
language=eng
handler_name=An other audio recording in english
vendor_id=[0][0][0][0]
[STREAM]
language=jpn
handler_name=An other audio recording in japanese
vendor_id=[0][0][0][0]


  


  Attempt to apply the updated metadata from ffmetadata2.txt :

  


  C:\Users\Alexander\Videos>ffmpeg -i INPUT.mp4 -i ffmetadata2.txt -map 0:v -map 0:a -map_metadata 1 -c copy OUTPUT2.mp4 -hide_banner


  


  Despite these steps, I've noticed that only the global metadata is updated, while the metadata for each stream remains unchanged. The console output shows that metadata for each stream is not updated as expected.

  


  What am I missing ? How can I ensure that the stream-specific metadata is also updated correctly when using a ffmetadata file ?

  


  Additional Information :

  


  

  • FFmpeg version : 2024-08-26-git-98610fe95f-full_build

  


  • The ffmetadata file format and the approach I've used should be correct according to the FFmpeg documentation.

  


  


  I would greatly appreciate any recommendations or suggestions on how to solve this problem !

  


  I found a bad solution for my problem, but it still isn't ideal as it requires specifying -map_metadata:s:N 1:s:N for each stream individually, which is quite cumbersome. Is there a way to simplify this process and avoid having to set metadata for each stream separately ?

  


  The command I’m using is :

  


  C:\Users\Alexander\Videos>ffmpeg -i INPUT.mp4 -i ffmetadata2.txt -map 0 -map_metadata:s:0 1:s:0 -map_metadata:s:1 1:s:1 -map_metadata:s:2 1:s:2 -map_metadata:s:3 1:s:3 -c copy OUTPUT2.mp4 -hide_banner


  


  This works, but having to specify -map_metadata:s:N for each stream creates extra work, especially as the number of streams increases. Is there a more efficient way to handle this ?