ECCV 2024 W-CODA: 1st Workshop on Multimodal Perception and Comprehension of Corner Cases in Autonomous Driving

2025-07-02

AI Summary

Autonomous driving faces significant challenges in perceiving and understanding rare yet high-risk corner cases—critical scenarios that are infrequent but pose severe safety risks. Method: This work introduces the first multimodal research platform dedicated to corner-case understanding and mitigation, featuring a dual-track “understanding + generation” challenge paradigm. It integrates vision, LiDAR, and other modalities with scene understanding and controllable generative models, leveraging real-world driving data for modeling and evaluation. Contribution/Results: It establishes the first systematic multimodal benchmark for corner-case perception and reasoning; proposes an evaluable, generation-based data augmentation framework; and organizes a dual-track challenge competition accompanied by a dedicated workshop, attracting broad participation from academia and industry. Outcomes include multiple peer-reviewed publications, five invited talks by domain experts, and fully reproducible technical solutions—collectively advancing robustness research and practical deployment of autonomous driving systems under extreme scenarios.

Technology Category

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

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Security and Privacy: Large-scale security measurements

Abstract

In this paper, we present details of the 1st W-CODA workshop, held in conjunction with the ECCV 2024. W-CODA aims to explore next-generation solutions for autonomous driving corner cases, empowered by state-of-the-art multimodal perception and comprehension techniques. 5 Speakers from both academia and industry are invited to share their latest progress and opinions. We collect research papers and hold a dual-track challenge, including both corner case scene understanding and generation. As the pioneering effort, we will continuously bridge the gap between frontier autonomous driving techniques and fully intelligent, reliable self-driving agents robust towards corner cases.

Problem

• Explore autonomous driving corner case solutions
• Advance multimodal perception and comprehension techniques
• Bridge gap in reliable self-driving for corner cases

Innovation

• Multimodal perception for corner cases
• Dual-track challenge design
• Bridging frontier techniques gap