Exposure Bracketing Is All You Need For A High-Quality Image

2024-01-01

AI Summary

Low-light single-image enhancement suffers from severe quality degradation, while existing multi-exposure methods fail to jointly exploit complementary information across exposures for diverse degradations. Method: We propose the first end-to-end framework unifying multi-exposure image modeling for joint denoising, deblurring, HDR reconstruction, and super-resolution. We introduce a Temporal-Modulated Recurrent Network (TMRNet) and a self-supervised domain adaptation strategy to overcome the absence of real paired multi-exposure data. Our approach combines synthetic pretraining with a controllable exposure simulation pipeline for multi-task optimization. Contribution/Results: Our method achieves significant improvements over state-of-the-art multi-image methods on both synthetic benchmarks and a large-scale real-world nighttime dataset comprising 200 scenes. To foster reproducibility and further research, we publicly release our code and dataset.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data

Abstract

It is highly desired but challenging to acquire high-quality photos with clear content in low-light environments. Although multi-image processing methods (using burst, dual-exposure, or multi-exposure images) have made significant progress in addressing this issue, they typically focus on specific restoration or enhancement problems, and do not fully explore the potential of utilizing multiple images. Motivated by the fact that multi-exposure images are complementary in denoising, deblurring, high dynamic range imaging, and super-resolution, we propose to utilize exposure bracketing photography to get a high-quality image by combining these tasks in this work. Due to the difficulty in collecting real-world pairs, we suggest a solution that first pre-trains the model with synthetic paired data and then adapts it to real-world unlabeled images. In particular, a temporally modulated recurrent network (TMRNet) and self-supervised adaptation method are proposed. Moreover, we construct a data simulation pipeline to synthesize pairs and collect real-world images from 200 nighttime scenarios. Experiments on both datasets show that our method performs favorably against the state-of-the-art multi-image processing ones. Code and datasets are available at https://github.com/cszhilu1998/BracketIRE.

Problem

• Low-light Imaging
• Multi-exposure Fusion
• Image Quality Enhancement

Innovation

• TMRNet
• Multi-Image Processing
• Low-Light Photography