Problem
Research questions and friction points this paper is trying to address.
V2X collaborative perception
3D object detection
spatial misalignment
multi-agent collaboration
3D-aware fusion
CoGoal3D: Collaborative 3D Object Detection with 3D-Aware Fusion and Refinement
🤖 AI Summary
Existing 3D object detection methods in vehicle-to-everything (V2X) cooperative perception suffer from performance degradation due to spatial misalignment caused by ignoring height and pose discrepancies among heterogeneous agents. To address this, we propose CoGoal3D, the first framework that explicitly models and corrects 3D spatial misalignment in V2X perception. Our approach operates in two stages: first, it aligns spatial features through multi-scale 3D perception with global feature fusion; second, it refines detection results via an auxiliary 3D point cloud reconstruction task and introduces a multi-agent collaborative data augmentation strategy to enhance training quality. Evaluated on DAIR-V2X, V2V4Real, and V2X-Real datasets, CoGoal3D achieves state-of-the-art performance, improving 3D AP@0.7 by 10.86%, 10.34%, and 10.18%, respectively.
📝 Abstract
V2X collaborative object detection features overcoming the limitations of single-vehicle systems by aggregating environmental features from multiple collaborative agents. However, existing mainstream V2X perception methods mainly focus on 2D BEV object detection. When 3D detection task is concerned, inferior results are obtained because they ignore the 3D spatial misalignment caused by differing height and attitude among the collaborators. In this paper, we propose a novel collaborative 3D object detection framework called CoGoal3D, which extracts and refines the 3D feature gradually in a two-stage pipeline. In the first stage, a multiscale 3D-aware global fusion module is designed to mitigate the 3D spatial misalignment. The resulting proposals are then refined in the second stage with an auxiliary task of 3D point reconstruction. An effective multi-agent collaborative data augmentation strategy is further proposed to enrich the training data while minimizing information loss. Extensive experiments on public real-world datasets demonstrate that our CoGoal3D achieves new state-of-the-art performance, with 3D AP@0.7 improvements of 10.86%, 10.34%, and 10.18% on the DAIR-V2X, V2V4Real, and V2X-Real datasets, respectively. Code is available at https://github.com/Megalo-f/CoGoal3D.
Innovation
Methods, ideas, or system contributions that make the work stand out.
3D-aware fusion
spatial misalignment mitigation
two-stage refinement
3D point reconstruction
multi-agent data augmentation
🔎 Similar Papers
No similar papers found.
