CVIVNov 28, 2024

Auto-Encoded Supervision for Perceptual Image Super-Resolution

arXiv:2412.00124v24 citationsh-index: 9CVPR
Originality Incremental advance
AI Analysis

This work addresses a specific bottleneck in perceptual image super-resolution for computer vision applications, offering an incremental improvement by modifying loss functions in existing frameworks.

The paper tackles the trade-off between fidelity and perceptual quality in GAN-based super-resolution by identifying and removing the blur-inducing component of pixel-level loss, proposing a new loss function (L_AESOP) that measures distance in an auto-encoder space to improve results without compromising perceptual quality.

This work tackles the fidelity objective in the perceptual super-resolution~(SR). Specifically, we address the shortcomings of pixel-level $L_\text{p}$ loss ($\mathcal{L}_\text{pix}$) in the GAN-based SR framework. Since $L_\text{pix}$ is known to have a trade-off relationship against perceptual quality, prior methods often multiply a small scale factor or utilize low-pass filters. However, this work shows that these circumventions fail to address the fundamental factor that induces blurring. Accordingly, we focus on two points: 1) precisely discriminating the subcomponent of $L_\text{pix}$ that contributes to blurring, and 2) only guiding based on the factor that is free from this trade-off relationship. We show that they can be achieved in a surprisingly simple manner, with an Auto-Encoder (AE) pretrained with $L_\text{pix}$. Accordingly, we propose the Auto-Encoded Supervision for Optimal Penalization loss ($L_\text{AESOP}$), a novel loss function that measures distance in the AE space, instead of the raw pixel space. Note that the AE space indicates the space after the decoder, not the bottleneck. By simply substituting $L_\text{pix}$ with $L_\text{AESOP}$, we can provide effective reconstruction guidance without compromising perceptual quality. Designed for simplicity, our method enables easy integration into existing SR frameworks. Experimental results verify that AESOP can lead to favorable results in the perceptual SR task.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes