Scale Equivariance Regularization and Feature Lifting in High Dynamic Range Modulo Imaging

2026-01-30

AI Summary

This work addresses the challenge of high dynamic range (HDR) reconstruction from modulo imaging, where natural image edges are often indistinguishable from artificial wrap-around discontinuities. To tackle this issue, the authors propose a learning-based HDR reconstruction framework that incorporates scale-equivariant regularization to enforce consistency across exposures. The method introduces a feature-enhanced input architecture that effectively fuses the raw modulo image, wrapped finite differences, and a closed-form initialization, thereby significantly improving the network’s ability to discriminate between genuine image structures and wrapping artifacts. Extensive experiments demonstrate that the proposed approach achieves state-of-the-art performance on both perceptual and linear HDR quality metrics, yielding substantially improved reconstruction accuracy.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Search and Optimization: Non-convex Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Modulo imaging enables high dynamic range (HDR) acquisition by cyclically wrapping saturated intensities, but accurate reconstruction remains challenging due to ambiguities between natural image edges and artificial wrap discontinuities. This work proposes a learning-based HDR restoration framework that incorporates two key strategies: (i) a scale-equivariant regularization that enforces consistency under exposure variations, and (ii) a feature lifting input design combining the raw modulo image, wrapped finite differences, and a closed-form initialization. Together, these components enhance the network's ability to distinguish true structure from wrapping artifacts, yielding state-of-the-art performance across perceptual and linear HDR quality metrics.

Problem

• HDR reconstruction
• modulo imaging
• wrap ambiguity
• scale equivariance
• feature lifting

Innovation

• scale equivariance
• modulo imaging
• HDR reconstruction
• feature lifting
• wrap artifact disambiguation