WACV 2015 Challenge

Pedestrian Detection Challenge. This benchmark contains 8 challenging video sequences (5 train, 3 test) in unconstrained environments. Tracking and evaluation are done in image coordinates. All sequences have been annotated with high accuracy, strictly following a well-defined protocol.

Training Set

SampleName FPS Resolution Length Tracks Boxes DensityDescriptionSourceRef.
1ADL-Rundle-6301920x1080525 (00:18)2450099.5A pedestrian street scene filmed from a low angle.link[1]
2ADL-Rundle-8301920x1080654 (00:22)28678310.4A pedestrian scene filmed at night by a moving cameralink[1]
3Venice-2301920x1080600 (00:20)26714111.9People walking around a large square.link[1]
Total 1779 frm.
(60 s.)
78 18933 10.6

Test Set

SampleName FPS Resolution Length Tracks Boxes DensityDescriptionSourceRef.
1ADL-Rundle-1301920x1080500 (00:17)32930618.6A busy pedestrian street filmed at eye level by a moving cameralink[1]
2ADL-Rundle-3301920x1080625 (00:21)441016616.3A crowded pedestrian street, stationary cameralink[1]
3Venice-1301920x1080450 (00:15)17456310.1People walking around a large square.link[1]
Total 1575 frm.
(53 s.)
93 24035 15.3


References:


[1] Leal-Taixé, L., Milan, A., Reid, I., Roth, S. & Schindler, K. MOTChallenge 2015: Towards a Benchmark for Multi-Target Tracking. arXiv:1504.01942 [cs], 2015., (arXiv: 1504.01942).