NTIRE 2025 Challenge on Day and Night Raindrop Removal for Dual-Focused Images: Methods and Results

2025-04-17

AI Summary

This work addresses the challenge of realistic raindrop degradation modeling and removal under dual-focus conditions—simultaneously capturing both raindrops and background—in day/night illumination scenarios. To overcome limitations of synthetic data, we introduce Raindrop Clarity, the first real-world multimodal raindrop removal benchmark, systematically covering day/night lighting and raindrop/background dual-focus degradation combinations. Methodologically, we propose the first day/night adaptive dual-focus raindrop degradation model and an end-to-end network integrating multi-scale feature fusion, a lighting-adaptive module, and a focus-aware loss function. Evaluated on 731 real-world test images, the top 32 submitted methods achieve an average PSNR gain of 2.1 dB over prior approaches, establishing new state-of-the-art performance. This work provides both a rigorous benchmark and a methodological framework for vision restoration under complex illumination and focus conditions.

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
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

This paper reviews the NTIRE 2025 Challenge on Day and Night Raindrop Removal for Dual-Focused Images. This challenge received a wide range of impressive solutions, which are developed and evaluated using our collected real-world Raindrop Clarity dataset. Unlike existing deraining datasets, our Raindrop Clarity dataset is more diverse and challenging in degradation types and contents, which includes day raindrop-focused, day background-focused, night raindrop-focused, and night background-focused degradations. This dataset is divided into three subsets for competition: 14,139 images for training, 240 images for validation, and 731 images for testing. The primary objective of this challenge is to establish a new and powerful benchmark for the task of removing raindrops under varying lighting and focus conditions. There are a total of 361 participants in the competition, and 32 teams submitting valid solutions and fact sheets for the final testing phase. These submissions achieved state-of-the-art (SOTA) performance on the Raindrop Clarity dataset. The project can be found at https://lixinustc.github.io/CVPR-NTIRE2025-RainDrop-Competition.github.io/.

Problem

• Remove raindrops from dual-focused day and night images
• Establish benchmark for diverse raindrop degradation removal
• Evaluate solutions using Raindrop Clarity dataset

Innovation

• Diverse Raindrop Clarity dataset for training
• Dual-focused deraining under varying conditions
• State-of-the-art solutions from 32 teams