EvFlow-GS: Event Enhanced Motion Deblurring with Optical Flow for 3D Gaussian Splatting

2026-04-23

AI Summary

Existing event camera–based methods for motion-blurred 3D reconstruction suffer from inaccurate double-integration priors and event noise, leading to artifacts and blurred textures. This work proposes EvFlow-GS, a framework that jointly optimizes a learnable double-integration model, camera poses, and 3D Gaussian Splatting (3DGS). By incorporating optical flow–guided event edge extraction and an event residual prior, the method enhances supervision of intensity variations across rendered images. A modular event loss function enables end-to-end co-optimization of event data and 3DGS, significantly improving reconstruction quality under motion blur. The approach effectively suppresses artifacts, recovers sharp textures, and achieves state-of-the-art performance on key evaluation metrics.

Technology Category

• Computer Vision: Motion & Tracking
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Multimodal Learning

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Achieving sharp 3D reconstruction from motion-blurred images alone becomes challenging, motivating recent methods to incorporate event cameras, benefiting from microsecond temporal resolution. However, they suffer from residual artifacts and blurry texture details due to misleading supervision from inaccurate event double integral priors and noisy, blurry events. In this study, we propose EvFlow-GS, a unified framework that leverages event streams and optical flow to optimize an end-to-end learnable double integral (LDI), camera poses, and 3D Gaussian Splatting (3DGS) jointly on-the-fly. Specifically, we first extract edge information from the events using optical flow and then formulate a novel event-based loss applied separately to different modules. Additionally, we exploit a novel event-residual prior to strengthen the supervision of intensity changes between images rendered from 3DGS. Finally, we integrate the outputs of both 3DGS and LDI into a joint loss, enabling their optimization to mutually facilitate each other. Experiments demonstrate the leading performance of our EvFlow-GS.

Problem

• motion deblurring
• 3D reconstruction
• event cameras
• optical flow
• 3D Gaussian Splatting

Innovation

• Event Camera
• Optical Flow
• 3D Gaussian Splatting
• Motion Deblurring
• Learnable Double Integral