A Flexible Framework for Universal Computational Aberration Correction via Automatic Lens Library Generation and Domain Adaptation

2024-09-09

arXiv.org

AI Summary

Existing general-purpose computational aberration correction (CAC) methods rely on training with limited lens libraries (LensLib), resulting in poor generalization; fine-tuning further requires known lens specifications, hindering adaptation to unknown lenses. This paper proposes OmniLens, the first framework enabling zero-shot, few-shot, and lens-parameter-free domain-adaptive aberration correction. Its core contributions are: (1) constructing AODLib—a high-fidelity, broadly representative lens library—via evolutionary automated optical design (EAOD); (2) incorporating a high-quality codebook prior to enhance model generalization; and (3) introducing a dark-channel-statistics-based unsupervised regularization term for ground-truth-free domain adaptation. Evaluated on four heterogeneous low-quality lenses, OmniLens achieves 97% of the performance of dedicated models under zero-shot settings, significantly overcoming conventional CAC’s dependence on lens priors and retraining.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Search and Optimization: Learning to Search

Application Category

• Responsible Web: Algorithmic accountability and transparency on the web
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Emerging universal Computational Aberration Correction (CAC) paradigms provide an inspiring solution to light-weight and high-quality imaging without repeated data preparation and model training to accommodate new lens designs. However, the training databases in these approaches, i.e., the lens libraries (LensLibs), suffer from their limited coverage of real-world aberration behaviors. In this work, we set up an OmniLens framework for universal CAC, considering both the generalization ability and flexibility. OmniLens extends the idea of universal CAC to a broader concept, where a base model is trained for three cases, including zero-shot CAC with the pre-trained model, few-shot CAC with a little lens-specific data for fine-tuning, and domain adaptive CAC using domain adaptation for lens-descriptions-unknown lens. In terms of OmniLens's data foundation, we first propose an Evolution-based Automatic Optical Design (EAOD) pipeline to construct LensLib automatically, coined AODLib, whose diversity is enriched by an evolution framework, with comprehensive constraints and a hybrid optimization strategy for achieving realistic aberration behaviors. For network design, we introduce the guidance of high-quality codebook priors to facilitate zero-shot CAC and few-shot CAC, which enhances the model's generalization ability, while also boosting its convergence in a few-shot case. Furthermore, based on the statistical observation of dark channel priors in optical degradation, we design an unsupervised regularization term to adapt the base model to the target descriptions-unknown lens using its aberration images without ground truth. We validate OmniLens on 4 manually designed low-end lenses with various structures and aberration behaviors. Remarkably, the base model trained on AODLib exhibits strong generalization capabilities, achieving 97% of the lens-specific performance in a zero-shot setting.

Problem

• Developing universal lens aberration correction with limited lens library coverage
• Adapting pre-trained models to unseen lenses without known descriptions
• Generating realistic lens samples for robust computational aberration correction

Innovation

• Evolution-based Automatic Optical Design for diverse lens samples
• Unsupervised regularization using dark channel priors
• LensLib-to-specific domain adaptation for unseen lenses