☆ 3.8 Proceedings Paper

A Versatile Multi-View Framework for LiDAR-based 3D Object Detection with Guidance from Panoptic Segmentation

2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2022) (2022)

From Points to Parts: 3D Object Detection From Point Cloud With Part-Aware and Part-Aggregation Network

Shaoshuai Shi et al.

Summary: This study proposes a novel point-cloud-based 3D object detection framework, Part-A(2) net, which includes two stages: part-aware and part-aggregation. By utilizing part supervisions from 3D ground-truth boxes and a new-designed RoI-aware point cloud pooling module, the framework achieves high-quality 3D proposals and accurate intra-object part locations prediction.

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE (2021)

添加到收藏夹

Proceedings Paper Computer Science, Artificial Intelligence