AIM 2025 Rip Current Segmentation (RipSeg) Challenge Report

2025-08-18

AI Summary

This study addresses the critical need for beach safety by proposing an automated instance segmentation method for rip currents in static images. To support rigorous evaluation, we organized the first Rip Current Segmentation Challenge, built upon RipVIS—the largest globally curated rip current dataset—and introduced a multidimensional evaluation framework integrating F1, F2, and AP-based metrics to enhance application-oriented reliability. Methodologically, our approach synergistically combines deep learning architectures, vision foundation models, domain adaptation techniques, and generalization strategies to robustly handle cross-regional variations, diverse illumination conditions, and complex marine environments. The challenge attracted 75 participants, yielding five valid submissions. The top-performing solution significantly improved boundary localization accuracy for rip currents—particularly under low-contrast conditions and in the presence of dynamic water surface interference—thereby establishing a deployable technical foundation for real-time beach hazard monitoring.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multi-instance/Multi-view Learning
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Security and Privacy: Large-scale security measurements
• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization

Abstract

This report presents an overview of the AIM 2025 RipSeg Challenge, a competition designed to advance techniques for automatic rip current segmentation in still images. Rip currents are dangerous, fast-moving flows that pose a major risk to beach safety worldwide, making accurate visual detection an important and underexplored research task. The challenge builds on RipVIS, the largest available rip current dataset, and focuses on single-class instance segmentation, where precise delineation is critical to fully capture the extent of rip currents. The dataset spans diverse locations, rip current types, and camera orientations, providing a realistic and challenging benchmark. In total, $75$ participants registered for this first edition, resulting in $5$ valid test submissions. Teams were evaluated on a composite score combining $F_1$, $F_2$, $AP_{50}$, and $AP_{[50:95]}$, ensuring robust and application-relevant rankings. The top-performing methods leveraged deep learning architectures, domain adaptation techniques, pretrained models, and domain generalization strategies to improve performance under diverse conditions. This report outlines the dataset details, competition framework, evaluation metrics, and final results, providing insights into the current state of rip current segmentation. We conclude with a discussion of key challenges, lessons learned from the submissions, and future directions for expanding RipSeg.

Problem

• Automating rip current segmentation in still images
• Addressing dangerous rip currents for beach safety
• Advancing single-class instance segmentation techniques

Innovation

• Deep learning architectures for segmentation
• Domain adaptation techniques for generalization
• Pretrained models for improved performance