Submission details of TEAM_Hojun

Name TEAM_Hojun
Paper Link https://arxiv.org/abs/1506.01497
Code Link https://github.com/open-mmlab/mmdetection
AP 0.659853
Input N/A
Runtime 0.04 s
Environment 4 GPU (Gtx Titan X)
Abstract State-of-the-art object detection networks depend on region proposal algorithms to hypothesize object locations. Advances like SPPnet [1] and Fast R-CNN [2] have reduced the running time of these detection networks, exposing region proposal computation as a bottleneck. In this work, we introduce a Region Proposal Network (RPN) that shares full-image convolutional features with the detection network, thus enabling nearly cost-free region proposals. An RPN is a fully convolutional network that simultaneously predicts object bounds and objectness scores at each position. The RPN is trained end-to-end to generate high-quality region proposals, which are used by Fast R-CNN for detection. We further merge RPN and Fast R-CNN into a single network by sharing their convolutional features—using the recently popular terminology of neural networks with “attention” mechanisms, the RPN component tells the unified network where to look. For the very deep VGG-16 model [3], our detection system has a frame rate of 5fps (inclu

Detailed results

Overall precision/recall curve

Per-sequence results

sequence name AP
cubberly-auditorium-2019-04-22_1 0.70433
discovery-walk-2019-02-28_0 0.762332
discovery-walk-2019-02-28_1 0.7228
food-trucks-2019-02-12_0 0.753931
gates-ai-lab-2019-04-17_0 0.73457
gates-basement-elevators-2019-01-17_0 0.882132
gates-foyer-2019-01-17_0 0.876479
gates-to-clark-2019-02-28_0 0.753789
hewlett-class-2019-01-23_0 0.891039
hewlett-class-2019-01-23_1 0.944583
huang-2-2019-01-25_1 0.575494
huang-intersection-2019-01-22_0 0.6802
indoor-coupa-cafe-2019-02-06_0 0.629018
lomita-serra-intersection-2019-01-30_0 0.628361
meyer-green-2019-03-16_1 0.483064
nvidia-aud-2019-01-25_0 0.603566
nvidia-aud-2019-04-18_1 0.754671
nvidia-aud-2019-04-18_2 0.799069
outdoor-coupa-cafe-2019-02-06_0 0.543669
quarry-road-2019-02-28_0 0.599787
serra-street-2019-01-30_0 0.579609
stlc-111-2019-04-19_1 0.825056
stlc-111-2019-04-19_2 0.806347
tressider-2019-03-16_2 0.590007
tressider-2019-04-26_0 0.577162
tressider-2019-04-26_1 0.662719
tressider-2019-04-26_3 0.636808
total 0.659853