Capturing Stable HDR Videos Using a Dual-Camera System

2025-07-09

AI Summary

To address flickering and ghosting artifacts in alternating-exposure HDR video reconstruction caused by exposure fluctuations in reference frames, this paper proposes a dual-camera cooperative capture framework and an Exposure-Adaptive Fusion Network (EAFNet). The primary camera captures a stable single-exposure reference sequence, while the auxiliary camera synchronously acquires multi-exposure auxiliary frames. EAFNet incorporates a pre-alignment subnetwork and a reference-dominant asymmetric cross-feature fusion module to achieve precise feature alignment and dynamic weight assignment across exposures. Additionally, a discrete wavelet transform (DWT)-based multi-scale reconstruction scheme enhances fine-detail fidelity. Extensive evaluations on multiple benchmark datasets demonstrate that our method significantly suppresses flickering and motion-related artifacts, achieving state-of-the-art HDR video reconstruction quality. The source code and dataset are publicly available.

Technology Category

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

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

In HDR video reconstruction, exposure fluctuations in reference images from alternating exposure methods often result in flickering. To address this issue, we propose a dual-camera system (DCS) for HDR video acquisition, where one camera is assigned to capture consistent reference sequences, while the other is assigned to capture non-reference sequences for information supplementation. To tackle the challenges posed by video data, we introduce an exposure-adaptive fusion network (EAFNet) to achieve more robust results. EAFNet introduced a pre-alignment subnetwork to explore the influence of exposure, selectively emphasizing the valuable features across different exposure levels. Then, the enhanced features are fused by the asymmetric cross-feature fusion subnetwork, which explores reference-dominated attention maps to improve image fusion by aligning cross-scale features and performing cross-feature fusion. Finally, the reconstruction subnetwork adopts a DWT-based multiscale architecture to reduce ghosting artifacts and refine features at different resolutions. Extensive experimental evaluations demonstrate that the proposed method achieves state-of-the-art performance on different datasets, validating the great potential of the DCS in HDR video reconstruction. The codes and data captured by DCS will be available at https://github.com/zqqqyu/DCS.

Problem

• Reduces flickering in HDR videos from exposure fluctuations
• Uses dual-camera system for stable reference and supplemental frames
• Introduces EAFNet for exposure-adaptive fusion and artifact reduction

Innovation

• Dual-camera system for stable HDR video
• Exposure-adaptive fusion network for robust results
• DWT-based multiscale architecture reduces ghosting