Metadata-Version: 2.1
Name: robust-vmaf
Version: 0.1.0.dev3
Summary: robust modification of VMAF
Home-page: https://github.com/Hrebtovmaksim1/robust_vmaf
Author: Mg_det
Author-email: hrebtovmaksim1@gmail.com
License: UNKNOWN
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.6
Description-Content-Type: text/markdown
License-File: LICENSE


# Robust VMAF

Robust modification of Video Multimethod Assessment Fusion (VMAF)

## Installation
You should install ffmpeg with VMAF support (guide can be found here https://github.com/netflix/vmaf/blob/master/resource/doc/ffmpeg.md)

## Usage

Example code:
```
from robust_vmaf import RobustVMAF

if __name__ == "__main__":
    rvmaf = RobustVMAF()
    print(rvmaf.predict("example/reference.mp4", "example/DragoToneMap.mp4"))
    print(rvmaf.predict("example/reference.mp4", "example/DragoToneMap.mp4", return_vmaf=False, return_vmaf_neg=False))
```

Output:
```
{'model': np.float32(95.81248), 'vmaf': 112.4708638, 'vmaf_neg': 91.1492206}
95.81248
```



