Fidelity-Naturalness Evaluation of Single Image Super Resolution
This work addresses evaluation challenges in super-resolution research, offering an incremental improvement over existing methods.
The paper tackles the problem of evaluating super-resolution methods by proposing a dual approach using both fidelity (difference from original images) and naturalness (human visual perception), with a new metric for fidelity and human labeling for naturalness. Experimental results show this method outperforms traditional single-measure evaluation, aiding future research.
We study the problem of evaluating super resolution methods. Traditional evaluation methods usually judge the quality of super resolved images based on a single measure of their difference with the original high resolution images. In this paper, we proposed to use both fidelity (the difference with original images) and naturalness (human visual perception of super resolved images) for evaluation. For fidelity evaluation, a new metric is proposed to solve the bias problem of traditional evaluation. For naturalness evaluation, we let humans label preference of super resolution results using pair-wise comparison, and test the correlation between human labeling results and image quality assessment metrics' outputs. Experimental results show that our fidelity-naturalness method is better than the traditional evaluation method for super resolution methods, which could help future research on single-image super resolution.