CoopDiff: A Diffusion-Guided Approach for Cooperation under Corruptions

TL;DR

CoopDiff is a diffusion-based framework that enhances cooperative perception robustness against diverse corruptions by employing a denoising mechanism and a teacher-student paradigm, significantly improving performance on multi-degradation benchmarks.

Contribution

This paper introduces CoopDiff, a novel diffusion-guided cooperative perception method that effectively mitigates corruptions through a teacher-student architecture with semantic guidance and adaptive decoding.

Findings

Outperforms prior methods across all corruption types

Reduces relative corruption error significantly

Provides a tunable trade-off between accuracy and efficiency

Abstract

Cooperative perception lets agents share information to expand coverage and improve scene understanding. However, in real-world scenarios, diverse and unpredictable corruptions undermine its robustness and generalization. To address these challenges, we introduce CoopDiff, a diffusion-based cooperative perception framework that mitigates corruptions via a denoising mechanism. CoopDiff adopts a teacher-student paradigm: the Quality-Aware Teacher performs voxel-level early fusion with Quality of Interest weighting and semantic guidance, then produces clean supervision features via a diffusion denoiser. The Dual-Branch Diffusion Student first separates ego and cooperative streams in encoding to reconstruct the teacher's clean targets. And then, an Ego-Guided Cross-Attention mechanism facilitates balanced decoding under degradation by adaptively integrating ego and cooperative features. We evaluate CoopDiff on two constructed multi-degradation benchmarks, OPV2Vn and DAIR-V2Xn, each incorporating six corruption types, including environmental and sensor-level distortions. Benefiting from the inherent denoising properties of diffusion, CoopDiff consistently outperforms prior methods across all degradation types and lowers the relative corruption error. Furthermore, it offers a tunable balance between precision and inference efficiency.