WEPDTOF: A Dataset and Benchmark Algorithms for In-the-Wild People Detection and Tracking from Overhead Fisheye Cameras

Mustafa Ozan Tezcan, Zhihao Duan, Mertcan Cokbas, Prakash Ishwar, Janusz Konrad

Abstract: Owing to their large field of view, overhead fisheye cameras are becoming a surveillance modality of choice for large indoor spaces. However, traditional people detection and tracking algorithms developed for side-mounted, rectilinear-lens cameras do not work well on images from overhead fisheye cameras due to their viewpoint and unique optics. While several people-detection algorithms have been recently developed for such cameras, they have all been tested on datasets consisting of "staged" recordings with a limited variety of people, scenes and challenges. Clearly, the performance of these algorithms "in the wild", i.e., on recordings with real-world challenges, remains un-known. In this paper, we introduce a new benchmark dataset of in-the-Wild Events for People Detection and Tracking from Overhead Fisheye cameras (WEPDTOF) ^{http://www.w3.org/1998/Math/MathML" xmlns:xlink="1}" target="_blank" rel="nofollow">http://www.w3.org/1999/xlink">1 . The dataset features 14 YouTube videos captured in a wide range of scenes, 188 distinct person identities consistently labeled across time, and real-world challenges such as extreme occlusions and camouflage. Also, we propose 3 spatiotemporal extensions ^{http://www.w3.org/1998/Math/MathML" xmlns:xlink="2}" target="_blank" rel="nofollow">http://www.w3.org/1999/xlink">2 of a state-of-the-art people-detection algorithm to enhance the coherence of detections across time. Compared to top-performing algorithms, that are purely spatial, the new algorithms offer a significant performance improvement on the new dataset. Finally, we compare the people tracking performance of these algorithms on WEPDTOF.