Dual-Exposure Imaging with Events

2026-04-11

AI Summary

This work addresses the artifact issues in conventional dual-exposure imaging under low-light conditions, which arise from motion and exposure discrepancies. To mitigate these problems, the authors propose E-DEI, an event-assisted dual-exposure imaging method that integrates short- and long-exposure frames with event camera data. Leveraging the high temporal resolution of events to provide precise motion cues, E-DEI introduces a Dual-path Feature Alignment and Fusion (DFAF) module that jointly performs motion deblurring and low-light enhancement within a unified framework. The study contributes the first real-world dataset, PIED, comprising aligned low-light/normal-light image pairs together with corresponding event streams, and demonstrates through extensive benchmarking that the proposed approach significantly outperforms existing methods in both image quality improvement and artifact suppression.

Technology Category

• Computer Vision: Low Level & Physics-based Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Multimodal Learning

Application Category

• Responsible Web: Data and user privacy-enhancing technologies for the Web
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

By combining complementary benefits of short- and long-exposure images, Dual-Exposure Imaging (DEI) enhances image quality in low-light scenarios. However, existing DEI approaches inevitably suffer from producing artifacts due to spatial displacement from scene motion and image feature discrepancies from different exposure times. To tackle this problem, we propose a novel Event-based DEI (E-DEI) algorithm, which reconstructs high-quality images from dual-exposure image pairs and events, leveraging high temporal resolution of event cameras to provide accurate inter-/intra-frame dynamic information. Specifically, we decompose this complex task into an integration of two sub-tasks, i.e., event-based motion deblurring and low-light image enhancement tasks, which guides us to design E-DEI network as a dual-path parallel feature propagation architecture. We propose a Dual-path Feature Alignment and Fusion (DFAF) module to effectively align and fuse features extracted from dual-exposure images with assistance of events. Furthermore, we build a real-world Dataset containing Paired low-/normal-light Images and Events (PIED). Experiments on multiple datasets show the superiority of our method. The code and dataset are available at github.

Problem

• Dual-Exposure Imaging
• motion artifacts
• exposure discrepancy
• low-light imaging
• image quality degradation

Innovation

• Event-based imaging
• Dual-exposure imaging
• Motion deblurring
• Low-light enhancement
• Feature alignment and fusion