DetReIDX: A Stress-Test Dataset for Real-World UAV-Based Person Recognition

TL;DR

DetReIDX introduces a large-scale aerial-ground dataset to stress-test UAV-based person ReID under real-world conditions, combining 13+ million bounding boxes, 509 identities, two sessions with clothing variation, multi-view UAV captures from 5.8 to 120 m altitude, and 16 soft biometrics with multi-task labels for detection, tracking, ReID, and action recognition. Systematic benchmarks show SOTA detectors fail dramatically under DetReIDX conditions, with up to 80% drops in detection AP50 and over 70% declines in Rank-1 ReID, highlighting reliance on appearance cues. By coupling aerial-ground cross-domain matching and long-range, low-resolution scenarios, DetReIDX reveals core gaps and provides a rigorous, publicly available benchmark to drive robust cross-view and clothing-invariant methods for UAV-based surveillance.

Abstract

Person reidentification (ReID) technology has been considered to perform relatively well under controlled, ground-level conditions, but it breaks down when deployed in challenging real-world settings. Evidently, this is due to extreme data variability factors such as resolution, viewpoint changes, scale variations, occlusions, and appearance shifts from clothing or session drifts. Moreover, the publicly available data sets do not realistically incorporate such kinds and magnitudes of variability, which limits the progress of this technology. This paper introduces DetReIDX, a large-scale aerial-ground person dataset, that was explicitly designed as a stress test to ReID under real-world conditions. DetReIDX is a multi-session set that includes over 13 million bounding boxes from 509 identities, collected in seven university campuses from three continents, with drone altitudes between 5.8 and 120 meters. More important, as a key novelty, DetReIDX subjects were recorded in (at least) two sessions on different days, with changes in clothing, daylight and location, making it suitable to actually evaluate long-term person ReID. Plus, data were annotated from 16 soft biometric attributes and multitask labels for detection, tracking, ReID, and action recognition. In order to provide empirical evidence of DetReIDX usefulness, we considered the specific tasks of human detection and ReID, where SOTA methods catastrophically degrade performance (up to 80% in detection accuracy and over 70% in Rank-1 ReID) when exposed to DetReIDXs conditions. The dataset, annotations, and official evaluation protocols are publicly available at https://www.it.ubi.pt/DetReIDX/

Figures (14)

• Figure 1: Comparison between the most important features of the publicly available datasets (ground-ground, aerial-aerial, and aerial-ground) and the DetReIDX dataset. Unlike its counterparts, DetReIDX includes clothing variations within subjects, with detection and tracking annotations, action labels, at wide altitude ranges (5.8m–120m).
• Figure 2: Examples of soft biometric annotations for two individuals in the DetReIDX dataset. Each subject is labeled with 16 visual and demographic attributes, facilitating fine-grained person analysis across multiple scenes.
• Figure 3: Satellite view of the data collection sites across the university campuses in Turkey, Angola, and India. The star markers indicate indoor dataset collection, and the green cones represent drone flight zones.
• Figure 4: Overview of the indoor data collection setup: (left) mugshots taken from three angles (left, front, right); (right) gait video.
• Figure 5: UAV-based outdoor capture protocol. Each subject is recorded from 18 drone viewpoints (P1–P18), spanning a wide range of altitudes, distances, and pitch angles. Recordings are repeated across two sessions (S1, S2) with varied clothing for appearance diversity.