README.md

Perceptual Assessment and Optimization of HDR Image Rendering

Introduction

This repository contains the official implementation of the paper "Perceptual Assessment and Optimization of HDR Image Rendering " by Peibei Cao, Rafal K. Mantiuk, and Kede Ma, IEEE Conference on Computer Vision and Pattern Recognition, 2024.

Prerequisites

• python>=3.6

As assessment metric

Usage:

• Please put reference HDR images in ./assessment/image/ref/, and test HDR images in ./assessment/image/test/.
• Please put the name of test HDR images in ./assessment/samples.txt.
• To evaluate the test HDR image:

python ./assessment/HDRMetric.py

Please make sure the LDR sequence is perfectly aligned with the reference sequence.

As loss function

Usage:

from HDRloss import hdrLoss
D = hdrLoss()
HDR_Loss = D(X, Y)
HDR_Loss.backward()

Please contact peibeicao2-c@my.cityu.edu.hk if you have any problem with the code.

Citation

@article{Cao2023Perceptual,
  title={Perceptual Assessment and Optimization of High Dynamic Range Image Rendering},
  author={Peibei Cao and Rafał K. Mantiuk and Kede Ma},
  journal={ArXiv},
  year={2023},
  volume={abs/2310.12877},
  url={https://api.semanticscholar.org/CorpusID:264306134}
}