fastiqa 0.6.1

fastiqa makes deep learning for image quality assessment faster and easier

ArXiv | Website | Setup | Document

PaQ-2-PiQ

Code for our paper "From Patches to Pictures (PaQ-2-PiQ): Mapping the Perceptual Space of Picture Quality"

@article{ying2019patches,
  title={From Patches to Pictures (PaQ-2-PiQ): Mapping the Perceptual Space of Picture Quality},
  author={Ying, Zhenqi
  ang and Niu, Haoran and Gupta, Praful and Mahajan, Dhruv and Ghadiyaram, Deepti and Bovik, Alan},
  journal={arXiv preprint arXiv:1912.10088},
  year={2019}
}

Features

• support cpu-only, just install pytorch-cpu and followed by fastai

Setup

• python 3.6/3.7 python --version

• install prerequisites by pip install -r requirements.txt

• Download the pretrained models and put them under a folder named models

  • RoIPoolModel-fit.10.bs.120.pth

• Open a Jupyter notebook and run demo.ipynb

:arrow_up: back to top