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

TL;DR

DetReIDX is a large, challenging UAV-based person re-identification dataset designed to evaluate and improve long-term, real-world person recognition under diverse and variable conditions, addressing limitations of existing datasets.

Contribution

The paper introduces DetReIDX, a novel large-scale UAV-based dataset with multi-session, long-term variability, and provides benchmark evaluations highlighting current methods' limitations.

Findings

State-of-the-art methods' detection accuracy drops up to 80%.

ReID performance degrades over 70% in Rank-1 accuracy.

DetReIDX reveals significant challenges for current ReID and detection algorithms.

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/