av1an-vmaf
is a command-line tool that serves as a wrapper for running
VMAF (Video Multimethod Assessment Fusion) on video files. It utilizes
the av1an_core
library for VMAF calculations and provides a convenient
interface for users to assess the quality of distorted videos compared
to reference videos.
- Rust: Make sure you have Rust installed on your system. You can install it by following the instructions at https://www.rust-lang.org/learn/get-started.
- ffmpeg: Ensure that ffmpeg is installed on your system. The tool relies on ffmpeg for video processing.
av1an-vmaf -d <distorted_file> -r <reference_file> -o <output_file> [OPTIONS]
-d, --distorted <FILE>
: Specifies the distorted video file (required).-r, --reference <FILE>
: Specifies the reference video file (required).-o, --output <FILE>
: Specifies the output JSON or SVG plot file (required).-m, --model <MODEL>
: Specifies the VMAF model to use (optional).--res <RES>
: Specifies the resolution to run VMAF with (default: 1920x1080).-s, --scaler <SCALER>
: Specifies the scaler to use (default: bicubic).-t, --threads <THREADS>
: Specifies the number of threads to use (default: 0, auto-detect).
av1an-vmaf -d distorted_video.mp4 -r reference_video.mp4 -o output.svg -m /usr/share/model/vmaf_4k_v0.6.1.json --res 3840x2160 -s bilinear -t 4096
av1an_core
: The core library for VMAF calculations.clap
: Command-line argument parsing.env_logger
: Environment-based logger for outputting log messages.
To build av1an
from source, navigate to the project directory and run:
cargo build --release
This will generate the executable in the target/release
directory.
This project is licensed under the AGPLv3 License - see the LICENSE file for details.