PV-RCNN++: Semantical point-voxel feature interaction for 3D object detection

Keyphrases

• Voxel-wise 100%
• 3D Object Detection 100%
• Feature Interaction 100%
• Point-Voxel 100%
• Copyright 33%
• Germany 33%
• Time Complexity 33%
• Local Features 33%
• Query-based 33%
• Semantic Information 33%
• Cyclists 33%
• Exclusive License 33%
• Pedestrian 33%
• Receptive Field 33%
• Attention-based 33%
• Detection Result 33%
• Semantic Segmentation 33%
• LiDAR Point Cloud 33%
• PointNet 33%
• Novelty Detection 33%
• Object Detection Network 33%
• KITTI Dataset 33%
• Edge Detector 33%
• Manhattan Distance 33%
• Object Detection 33%

Computer Science

• Object Detection 100%
• 3D Object Detection 100%
• Feature Interaction 100%
• Time Complexity 50%
• Image Segmentation 50%
• local feature 50%
• Detection Result 50%
• Point Cloud 50%
• Receptive Field 50%
• Manhattan Distance 50%
• Neighboring Voxels 50%
• Edge Detector 50%