OpenCOOD-Air: Prompting Heterogeneous Ground-Air Collaborative Perception with Spatial Conversion and Offset Prediction

TL;DR

OpenCOOD-Air introduces a UAV-integrated V2V perception framework that leverages transfer learning, spatial conversion, and offset prediction to enhance sensing beyond ground occlusions, validated by a new benchmark and improved detection metrics.

Contribution

The paper presents a novel UAV integration method with explicit spatial conversion and offset prediction, addressing domain gaps and data sparsity in ground-air collaborative perception.

Findings

Improves 2D AP@0.7 by 4%

Enhances 3D AP@0.7 by 7%

Validates approach with OPV2V-Air benchmark

Abstract

While Vehicle-to-Vehicle (V2V) collaboration extends sensing ranges through multi-agent data sharing, its reliability remains severely constrained by ground-level occlusions and the limited perspective of chassis-mounted sensors, which often result in critical perception blind spots. We propose OpenCOOD-Air, a novel framework that integrates UAVs as extensible platforms into V2V collaborative perception to overcome these constraints. To mitigate gradient interference from ground-air domain gaps and data sparsity, we adopt a transfer learning strategy to fine-tune UAV weights from pre-trained V2V models. To prevent the spatial information loss inherent in this transition, we formulate ground-air collaborative perception as a heterogeneous integration task with explicit altitude supervision and introduce a Cross-Domain Spatial Converter (CDSC) and a Spatial Offset Prediction Transformer (SOPT). Furthermore, we present the OPV2V-Air benchmark to validate the transition from V2V to Vehicle-to-Vehicle-to-UAV. Compared to state-of-the-art methods, our approach improves 2D and 3D AP@0.7 by 4% and 7%, respectively.