AG-VPReID.VIR: Bridging Aerial and Ground Platforms for Video-based Visible-Infrared Person Re-ID

TL;DR

This paper introduces a new aerial-ground cross-modality video dataset for person Re-ID, along with a novel three-stream architecture that effectively handles cross-viewpoint, modality, and temporal challenges, advancing surveillance capabilities.

Contribution

The paper presents the first aerial-ground cross-modality video dataset for person Re-ID and proposes TCC-VPReID, a novel architecture addressing cross-platform and cross-modality challenges.

Findings

TCC-VPReID achieves significant performance improvements on the new dataset.

The dataset presents unique challenges not found in existing Re-ID datasets.

The proposed method effectively bridges domain gaps between aerial and ground perspectives.

Abstract

Person re-identification (Re-ID) across visible and infrared modalities is crucial for 24-hour surveillance systems, but existing datasets primarily focus on ground-level perspectives. While ground-based IR systems offer nighttime capabilities, they suffer from occlusions, limited coverage, and vulnerability to obstructions--problems that aerial perspectives uniquely solve. To address these limitations, we introduce AG-VPReID.VIR, the first aerial-ground cross-modality video-based person Re-ID dataset. This dataset captures 1,837 identities across 4,861 tracklets (124,855 frames) using both UAV-mounted and fixed CCTV cameras in RGB and infrared modalities. AG-VPReID.VIR presents unique challenges including cross-viewpoint variations, modality discrepancies, and temporal dynamics. Additionally, we propose TCC-VPReID, a novel three-stream architecture designed to address the joint challenges of cross-platform and cross-modality person Re-ID. Our approach bridges the domain gaps between aerial-ground perspectives and RGB-IR modalities, through style-robust feature learning, memory-based cross-view adaptation, and intermediary-guided temporal modeling. Experiments show that AG-VPReID.VIR presents distinctive challenges compared to existing datasets, with our TCC-VPReID framework achieving significant performance gains across multiple evaluation protocols. Dataset and code are available at https://github.com/agvpreid25/AG-VPReID.VIR.