English

Perceptual Quality Optimization of Image Super-Resolution

Image and Video Processing 2026-02-26 v1 Computer Vision and Pattern Recognition Multimedia

Abstract

Single-image super-resolution (SR) has achieved remarkable progress with deep learning, yet most approaches rely on distortion-oriented losses or heuristic perceptual priors, which often lead to a trade-off between fidelity and visual quality. To address this issue, we propose an \textit{Efficient Perceptual Bi-directional Attention Network (Efficient-PBAN)} that explicitly optimizes SR towards human-preferred quality. Unlike patch-based quality models, Efficient-PBAN avoids extensive patch sampling and enables efficient image-level perception. The proposed framework is trained on our self-constructed SR quality dataset that covers a wide range of state-of-the-art SR methods with corresponding human opinion scores. Using this dataset, Efficient-PBAN learns to predict perceptual quality in a way that correlates strongly with subjective judgments. The learned metric is further integrated into SR training as a differentiable perceptual loss, enabling closed-loop alignment between reconstruction and perceptual assessment. Extensive experiments demonstrate that our approach delivers superior perceptual quality. Code is publicly available at https://github.com/Lighting-YXLI/Efficient-PBAN.

Keywords

Cite

@article{arxiv.2602.21482,
  title  = {Perceptual Quality Optimization of Image Super-Resolution},
  author = {Wei Zhou and Yixiao Li and Hadi Amirpour and Xiaoshuai Hao and Jiang Liu and Peng Wang and Hantao Liu},
  journal= {arXiv preprint arXiv:2602.21482},
  year   = {2026}
}

Comments

6 pages, 2 figures, accepted in ICASSP 26

R2 v1 2026-07-01T10:50:56.327Z