Submission details of Team_minjunmin

Name Team_minjunmin
Paper Link N/A
Code Link N/A
AP 0.57262
Input N/A
Runtime 0.04 s
Environment 1 GPU (GTX 1080Ti)
Abstract we focus on exploring the robustness of the3D object detection in point clouds, which has been rarelydiscussed in existing approaches. We observe two crucialphenomena: 1) the detection accuracy of the hard objects,e.g., Pedestrians, is unsatisfactory, 2) when adding additionalnoise points, the performance of existing approaches de-creases rapidly. To alleviate these problems, a novel TANet isintroduced in this paper, which mainly contains a Triple At-tention (TA) module, and a Coarse-to-Fine Regression (CFR)module. By considering the channel-wise, point-wise andvoxel-wise attention jointly, the TA module enhances the cru-cial information of the target while suppresses the unsta-ble cloud points. Besides, the novel stacked TA further ex-ploits the multi-level feature attention. In addition, the CFRmodule boosts the accuracy of localization without excessivecomputation cost. Experimental results on the validation setof KITTI dataset demonstrate that, in the challenging noisycases, i.e., adding additional random noisy points around eachobject, the presented approach goes far beyond state-of-the-art approaches. Furthermore, for the 3D object detection taskof the KITTI benchmark, our approach ranks the first place onPedestrian class, by using the point clouds as the only input.The running speed is around 29 frames per secon

Detailed results

Overall precision/recall curve

Per-sequence results

sequence name AP
cubberly-auditorium-2019-04-22_1 0.737335
discovery-walk-2019-02-28_0 0.69527
discovery-walk-2019-02-28_1 0.596967
food-trucks-2019-02-12_0 0.676792
gates-ai-lab-2019-04-17_0 0.527557
gates-basement-elevators-2019-01-17_0 0.680004
gates-foyer-2019-01-17_0 0.650101
gates-to-clark-2019-02-28_0 0.5088
hewlett-class-2019-01-23_0 0.783555
hewlett-class-2019-01-23_1 0.90934
huang-2-2019-01-25_1 0.436843
huang-intersection-2019-01-22_0 0.512194
indoor-coupa-cafe-2019-02-06_0 0.560983
lomita-serra-intersection-2019-01-30_0 0.530941
meyer-green-2019-03-16_1 0.439266
nvidia-aud-2019-01-25_0 0.525444
nvidia-aud-2019-04-18_1 0.530684
nvidia-aud-2019-04-18_2 0.819395
outdoor-coupa-cafe-2019-02-06_0 0.470216
quarry-road-2019-02-28_0 0.532776
serra-street-2019-01-30_0 0.815933
stlc-111-2019-04-19_1 0.818315
stlc-111-2019-04-19_2 0.732863
tressider-2019-03-16_2 0.614294
tressider-2019-04-26_0 0.589827
tressider-2019-04-26_1 0.603319
tressider-2019-04-26_3 0.531408
total 0.57262