Similar to the evaluation method used in MS COCO[1], we use AP, AP_{IOU = 0.50}, AP_{IOU = 0.75}, AR_{max = 10}, AR_{max = 100}, AR_{max = 500} and other indicators to evaluate the results of the detection algorithm.
Similar to the evaluation method used in MOTChallenge[2], we use MOTA, MOTP, IDF1, FAR, MT, and Hz to evaluate the results of the multi-object tracking algorithm.
[1] T. Lin, M. Maire, S. J. Belongie, J. Hays, P. Perona, D. Ramanan, P. Dollar, and C. L. Zitnick, “Microsoft COCO: common objects in context,” in Proceedings of European Conference on Computer Vision, 2014, pp. 740–755
[2] Milan A, Leal-Taixé L, Reid I, et al. MOT16: A benchmark for multi-object tracking[J]. arXiv preprint arXiv:1603.00831, 2016.
GigaVision Challenge 2020: Task 2
¥ 35,000
164 participants
start
Final Submissions
2020-03-20
2020-08-15
Sponsor:Tsinghua University & ECCV