NTIRE 2025 Challenge on Event-Based Image Deblurring: Methods and Results

2025-04-16

AI Summary

To address the challenging problem of single-frame motion deblurring in event-camera imaging, this paper proposes the first cross-modal event-image collaborative deblurring framework: it jointly processes the input blurred image and its synchronized event stream within an end-to-end learnable network, leveraging cross-modal feature alignment and PSNR-oriented optimization for high-fidelity reconstruction. Key contributions include: (1) organizing the first international challenge on event-driven image deblurring; (2) pioneering the event-image fusion deblurring paradigm, thereby opening an unconstrained design space for model architecture; and (3) empirically demonstrating that this paradigm substantially extends the performance frontier and methodological diversity for dynamic-scene restoration—evidenced by 199 registered teams and 15 valid submissions to the challenge.

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
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

This paper presents an overview of NTIRE 2025 the First Challenge on Event-Based Image Deblurring, detailing the proposed methodologies and corresponding results. The primary goal of the challenge is to design an event-based method that achieves high-quality image deblurring, with performance quantitatively assessed using Peak Signal-to-Noise Ratio (PSNR). Notably, there are no restrictions on computational complexity or model size. The task focuses on leveraging both events and images as inputs for single-image deblurring. A total of 199 participants registered, among whom 15 teams successfully submitted valid results, offering valuable insights into the current state of event-based image deblurring. We anticipate that this challenge will drive further advancements in event-based vision research.

Problem

• Develop event-based method for high-quality image deblurring
• Assess performance using Peak Signal-to-Noise Ratio (PSNR)
• Leverage events and images for single-image deblurring

Innovation

• Event-based high-quality image deblurring method
• Leveraging events and images as inputs
• No restrictions on computational complexity