Interest Entanglement: The Hidden Barrier to Blind Super-Resolution Optimization

2026-06-21

AI Summary

This work addresses the inherent trade-off between fidelity and perceptual quality in image super-resolution, which conventional multi-objective optimization approaches fail to balance effectively due to their neglect of the entangled interests between these objectives at the feature level. The study formally identifies and characterizes this “interest entanglement” problem for the first time, and proposes a Shared Feature Representation (SFR) framework that decouples the learning processes of regression and perceptual objectives through frequency-domain analysis. To further enhance synergy between the two goals, an InfoSqueeze module is introduced to compress redundant information in the shared representation. Extensive experiments on five standard benchmarks demonstrate that SFR consistently achieves superior performance over state-of-the-art methods in both quantitative metrics and visual quality.

Abstract

Fidelity and perceptual quality are two inherently competing and conflicting objectives in the image super-resolution (SR) task. Different loss functions focus on these objectives to varying extents. Regression losses enhance the model's fidelity but lack sufficient attention to high-frequency details, resulting in a loss of fine details. In contrast, perception losses improve the model's visual quality but may introduce undesirable artifacts. Balancing these two optimization goals can be viewed as a Multi-Objective Optimization problem. Existing methods are limited to cautiously adjusting weight parameters between these losses, overlooking the underlying Interest Entanglement problem. To address this problem, we explore the inherent frequency-domain conflict between the regression objective and the perceptual objective, and analyze the causes of Interest Entanglement in SR tasks. According to our findings, we propose the Shared-Feature-Representation based Super-Resolution framework (SFR), which decouples the learning process of different optimization objectives, allowing the model to explore a common optimization direction for both goals and achieve an effective balance between them. To better leverage shared features, we also proposed the InfoSqueeze module, which filters redundant information through a dimensionality reduction and expansion process, effectively transforming features into a consistent space. Quantitative and qualitative experiments across five representative datasets affirm the superiority of SFR.

Problem

• Super-Resolution
• Fidelity
• Perceptual Quality
• Interest Entanglement
• Multi-Objective Optimization

Innovation

• Interest Entanglement
• Super-Resolution
• Multi-Objective Optimization
• Shared-Feature Representation
• InfoSqueeze