Frustum PointNets for 3D Object Detection from RGB-D Data

CVPR 2018 Charles R. QiWei LiuChenxia WuHao SuLeonidas J. Guibas

In this work, we study 3D object detection from RGB-D data in both indoor and outdoor scenes. While previous methods focus on images or 3D voxels, often obscuring natural 3D patterns and invariances of 3D data, we directly operate on raw point clouds by popping up RGB-D scans... (read more)

PDF Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT LEADERBOARD
Object Localization KITTI Cars Easy Frustum PointNets AP 88.70% # 2
3D Object Detection KITTI Cars Easy Frustum PointNets AP 81.20% # 12
3D Object Detection KITTI Cars Easy val F-PointNet [Qi:2018fd] AP 83.26 # 2
Object Detection KITTI Cars Hard F-PointNet AP 62.19 # 4
3D Object Detection KITTI Cars Hard Frustum PointNets AP 62.19% # 12
Object Localization KITTI Cars Hard Frustum PointNets AP 75.33% # 2
3D Object Detection KITTI Cars Hard val F-PointNet [Qi:2018fd] AP 62.56 # 2
Object Localization KITTI Cars Moderate Frustum PointNets AP 84.00% # 1
3D Object Detection KITTI Cars Moderate Frustum PointNets AP 70.39% # 14
3D Object Detection KITTI Cars Moderate val F-PointNet [Qi:2018fd] AP 69.28 # 2
3D Object Detection KITTI Cyclist Easy val F-PointNet++ [Qi:2018fd] AP 77.15 # 2
3D Object Detection KITTI Cyclist Hard val F-PointNet++ [Qi:2018fd] AP 53.37 # 2
3D Object Detection KITTI Cyclist Moderate val F-PointNet++ [Qi:2018fd] AP 56.49 # 2
3D Object Detection KITTI Cyclists Easy Frustum PointNets AP 71.96% # 7
Object Localization KITTI Cyclists Easy Frustum PointNets AP 75.38% # 1
Object Localization KITTI Cyclists Hard Frustum PointNets AP 54.68% # 1
3D Object Detection KITTI Cyclists Hard Frustum PointNets AP 50.39% # 7
3D Object Detection KITTI Cyclists Moderate Frustum PointNets AP 56.77% # 7
Object Localization KITTI Cyclists Moderate Frustum PointNets AP 61.96% # 1
Birds Eye View Object Detection KITTI Cyclists Moderate F-PointNet AP 61.96% # 4
3D Object Detection KITTI Pedestrian Easy val F-PointNet++ [Qi:2018fd] AP 70.00 # 2
3D Object Detection KITTI Pedestrian Hard val F-PointNet++ [Qi:2018fd] AP 53.59 # 2
3D Object Detection KITTI Pedestrian Moderate val F-PointNet++ [Qi:2018fd] AP 61.32 # 2
3D Object Detection KITTI Pedestrians Easy Frustum PointNets AP 51.21% # 5
Object Localization KITTI Pedestrians Easy Frustum PointNets AP 58.09% # 1
Object Localization KITTI Pedestrians Hard Frustum PointNets AP 47.20% # 1
3D Object Detection KITTI Pedestrians Hard Frustum PointNets AP 40.23% # 6
Birds Eye View Object Detection KITTI Pedestrians Moderate F-PointNet AP 50.22% # 4
Object Localization KITTI Pedestrians Moderate Frustum PointNets AP 50.22% # 1
3D Object Detection KITTI Pedestrians Moderate Frustum PointNets AP 42.15% # 7
3D Object Detection SUN-RGBD Frustum PointNets MAP 54.0 # 2
3D Object Detection SUN-RGBD val F-PointNet MAP 54.0 # 4

Results from Other Papers


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK SOURCE PAPER COMPARE
3D Object Detection KITTI Cyclist Easy val F-PointNet [Qi:2018fd] AP 74.54 # 3
3D Object Detection KITTI Cyclist Hard val F-PointNet [Qi:2018fd] AP 52.65 # 3
3D Object Detection KITTI Cyclist Moderate val F-PointNet [Qi:2018fd] AP 55.95 # 3
3D Object Detection KITTI Pedestrian Easy val F-PointNet [Qi:2018fd] AP 65.08 # 3
3D Object Detection KITTI Pedestrian Hard val F-PointNet [Qi:2018fd] AP 49.28 # 3
3D Object Detection KITTI Pedestrian Moderate val F-PointNet [Qi:2018fd] AP 55.85 # 3

Methods used in the Paper


METHOD TYPE
🤖 No Methods Found Help the community by adding them if they're not listed; e.g. Deep Residual Learning for Image Recognition uses ResNet