Cross-platform command-line AV1 / VP9 / HEVC / H264 encoding framework with per scene quality encoding
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
 
 
Josh Holmer 32cebb8808
Scene detect changes (#383)
2 days ago
.cargo Fix x264/x265, Disable multiplexing (#290) 3 months ago
.devcontainer Add probe slow, add devcontainer, update readme (#289) 4 months ago
.github/workflows Scene detect changes (#383) 2 days ago
av1an-cli Changes for rav1e's hybrid scene detection update (#381) 1 week ago
av1an-core Scene detect changes (#383) 2 days ago
docs Set log level in FFmpeg + small fixes (#376) 2 weeks ago
src Refactor into `Input` type for handling vapoursynth input (#374) 3 weeks ago
.codeclimate.yml Disable file-lines and method-lines check for codeclimate, enable fixme plugin to mark TODOs/FIXMEs 1 year ago
.gitignore More idiomatic Rust, add Cargo.lock to the repo (#313) 3 months ago
CHANGELOG.md Removed per frame target quaity 5 months ago
Cargo.lock fix building 1 week ago
Cargo.toml Scene detect changes (#383) 2 days ago
Dockerfile Use a different method for ffmpeg frame count (#368) 3 weeks ago
LICENSE.md License change 1 year ago
MANIFEST.in wip 5 months ago
README.md Correct readme.md 2 months ago
appveyor.yml Fix chunking/resume, port chunk generation to Rust, remove unused Python code, update indicatif (#297) 3 months ago
rustfmt.toml formating 5 months ago

README.md


Av1an

A cross-platform framework to streamline encoding

alt text

Discord server

Easy, Fast, Efficient and Feature Rich

An easy way to start using AV1 / HEVC / H264 / VP9 / VP8 encoding. AOM, RAV1E, SVT-AV1, SVT-VP9, VPX, x265, x264 are supported.

Example with default parameters:

av1an -i input

With your own parameters:

av1an -i input -v " --cpu-used=3 --end-usage=q --cq-level=30 --threads=8 " -w 10
--target-quality 95 -a "-c:a libopus -ac 2 -b:a 192k" -log my_log -o output

Usage

-i  --input             Input file(s), or Vapoursynth (.py,.vpy) script
                        (relative or absolute path)

-o  --output-file       Name/Path for output file (Default: (input file name)_(encoder).mkv)
                        Output is `mkv` by default
                        Ouput extension can be set to: `mkv`, `webm`, `mp4`

-e  --encoder           Encoder to use
                        [default: aom] [possible values: aom, rav1e, vpx, svt-av1, x264, x265]

-v  --video-params      Encoder settings flags (If not set, will be used default parameters.)
                        Must be inside ' ' or " "

-p  --passes            Set number of passes for encoding
                        (Default: AOMENC: 2, rav1e: 1, SVT-AV1: 1, SVT-VP9: 1,
                        VPX: 2, x265: 1, x264: 1)

-w  --workers           Override number of workers.

-r  --resume            Resumes encoding.

--keep                  Doesn't delete temporary folders after encode has finished.

-q  --quiet             Do not print a progress bar to the terminal.

-l  --logging           Path to .log file(By default created in temp folder)

--temp                  Set path for the temporary folder. [default: .hash]

-c  --concat            Concatenation method to use for splits Default: ffmpeg
                        [possible values: ffmpeg, mkvmerge, ivf]

FFmpeg options

-a  --audio-params      FFmpeg audio settings (Default: copy audio from source to output)
                        Example: -a '-c:a libopus -b:a  64k'

-f  --ffmpeg            FFmpeg options video options.
                        Applied to each encoding segment individually.
                        (Warning: Cropping doesn't work with Target VMAF mode
                        without specifying it in --vmaf-filter)
                        Example:
                        --ff " -vf scale=320:240 "

--pix-format            Setting custom pixel/bit format for piping
                        (Default: 'yuv420p10le')

Segmenting

--split-method          Method used for generating splits.(Default: av-scenedetect)
                        Options: `av-scenedetect`, `none`
                        `none` -  skips scenedetection.

-m  --chunk-method      Determine the method in which chunks are made for encoding.
                        By default the best method is selected automatically.
                        [possible values: segment, select, ffms2, lsmash, hybrid]

-s  --scenes            File to save/read scenes.

-x  --extra-split       Size of chunk after which it will be split [default: 240]

--min-scene-len         Specifies the minimum number of frames in each split.

Target Quality

--target-quality        Quality value to target.
                        VMAF used as substructure for algorithms.
                        When using this mode, you must use quantizer/quality modes of enocoder.

--target-quality-method Type of algorithm for use.
                        Options: per_shot

--min-q, --max-q        Min,Max Q values limits
                        If not set by the user, the default for encoder range will be used.

--vmaf                  Calculate VMAF after encoding is done and make a plot.

--vmaf-path             Custom path to libvmaf models.
                        example: --vmaf-path "vmaf_v0.6.1.pkl"
                        Recommended to place both files in encoding folder
                        (`vmaf_v0.6.1.pkl` and `vmaf_v0.6.1.pkl.model`)
                        (Required if VMAF calculation doesn't work by default)

--vmaf-res              Resolution for VMAF calculation.
                        [default: 1920x1080]

--probes                Number of probes for target quality. [default: 4]

--probe-slow            Use probided video encoding parameters for vmaf probes.

--vmaf-filter           Filter used for VMAF calculation. The passed format is filter_complex.
                        So if crop filter used ` -ff " -vf crop=200:1000:0:0 "`
                        `--vmaf-filter` must be : ` --vmaf-filter "crop=200:1000:0:0"`

--probing-rate          Setting rate for VMAF probes. Using every N frame used in probe.
                        [default: 4]

--vmaf-threads          Limit number of threads that are used for VMAF calculation

Main Features

Splitting video by scenes for parallel encoding because AV1 encoders are currently not very good at multithreading and encoding is limited to a very limited number of threads.

  • Vapoursynth script input support.
  • Speed up video encoding.
  • Target Quality mode. Targeting end result reference visual quality. VMAF used as a substructure
  • Resuming encoding without loss of encoded progress.
  • Simple and clean console look.
  • Automatic detection of the number of workers the host can handle.
  • Both video and audio transcoding.

Install

Docker

Av1an can be run in a Docker container with the following command if you are in the current directory Linux

docker run --privileged -v "$(pwd):/videos" --user $(id -u):$(id -g) -it --rm masterofzen/av1an:latest -i S01E01.mkv {options}

Windows

docker run --privileged -v "${PWD}:/videos" -it --rm masterofzen/av1an:latest -i S01E01.mkv {options}

Docker can also be built by using

docker build -t "av1an" .

To specify a different directory to use you would replace $(pwd) with the directory

docker run --privileged -v "/c/Users/masterofzen/Videos":/videos --user $(id -u):$(id -g) -it --rm masterofzen/av1an:latest -i S01E01.mkv {options}

The --user flag is required on linux to avoid permission issues with the docker container not being able to write to the location, if you get permission issues ensure your user has access to the folder that you are using to encode.

Docker tags

The docker image has the following tags

Tag Description
latest Contains the latest stable av1an version release
master Contains the latest av1an commit to the master branch
sha-##### Contains the commit of the hash that is referenced
#.## Stable av1an version release

Support the developer

Bitcoin - 1GTRkvV4KdSaRyFDYTpZckPKQCoWbWkJV1