Resolving Blind Inverse Problems under Dynamic Range Compression via Structured Forward Operator Modeling

2026-03-02

AI Summary

This work addresses the challenge of recovering radiometric fidelity in dynamic range compression tasks—such as low-light enhancement and HDR reconstruction—where unknown forward models and irreversible information loss hinder accurate inversion. To this end, the authors propose the Cascaded Monotonic Bernstein (CaMB) operator, which for the first time embeds monotonicity as a hard structural prior directly into the parameterization of the forward model. Integrated within a plug-and-play diffusion framework, CaMB enables zero-shot blind inverse problem solving without requiring paired training data. Extensive experiments demonstrate that the method significantly outperforms existing approaches across diverse applications including low-light image enhancement, low-field MRI reconstruction, and HDR imaging, achieving state-of-the-art performance in both signal fidelity and physical consistency.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• 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
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

Recovering radiometric fidelity from unknown dynamic range compression (UDRC), such as low-light enhancement and HDR reconstruction, is a challenging blind inverse problem, due to the unknown forward model and irreversible information loss introduced by compression. To address this challenge, we first identify monotonicity as the fundamental physical invariant shared across UDRC tasks. Leveraging this insight, we introduce the \textbf{cascaded monotonic Bernstein} (CaMB) operator to parameterize the unknown forward model. CaMB enforces monotonicity as a hard architectural inductive bias, constraining optimization to physically consistent mappings and enabling robust and stable operator estimation. We further integrate CaMB with a plug-and-play diffusion framework, proposing \textbf{CaMB-Diff}. Within this framework, the diffusion model serves as a powerful geometric prior for structural and semantic recovery, while CaMB explicitly models and corrects radiometric distortions through a physically grounded forward operator. Extensive experiments on a variety of zero-shot UDRC tasks, including low-light enhancement, low-field MRI enhancement, and HDR reconstruction, demonstrate that CaMB-Diff significantly outperforms state-of-the-art zero-shot baselines in terms of both signal fidelity and physical consistency. Moreover, we empirically validate the effectiveness of the proposed CaMB parameterization in accurately modeling the unknown forward operator.

Problem

• blind inverse problem
• dynamic range compression
• radiometric fidelity
• unknown forward model
• information loss

Innovation

• monotonicity
• structured forward operator
• blind inverse problems
• dynamic range compression
• diffusion prior