VMAF
Netflix's perceptual video quality metric, with a command-line tool and library for scoring encoded video.
Alternatives
About VMAF
VMAF (Video Multi-Method Assessment Fusion) is an Emmy-winning perceptual video quality assessment algorithm developed by Netflix. It rates how good an encoded or compressed video looks, which helps when comparing encoder settings or codecs.
The package contains the standalone C library libvmaf and its wrappers, Python and MATLAB code, pretrained models, and Dockerfiles, including one for FFmpeg. It suits video engineers and people tuning encodes who want a repeatable quality score.
Strengths
- Widely used, Emmy-winning quality metric from Netflix
- Includes pretrained models and the libvmaf C library
- Docker setup for use with FFmpeg
- Actively maintained
Limitations
- Command-line and library only, with no graphical interface
- Scores need some background knowledge to read well
Details
- Pricing
- FreeFree and open source under the BSD-2-Clause-Patent licence.
- License
- BSD-2-Clause-Patent
- Developer
- Netflix
- Platforms
- Windows, Command line
- How it runs
- Downloadable app
- Account
- Not required
- Works offline
- Yes
- Best suited for
- Video engineers comparing encoder settings with an objective quality score
- Categories
- Video editing, CLI tools
- Last verified
- Added
- Sources