Behind the Curtain: Learning Occluded Shapes for 3D Object Detection

12/04/2021
by   Qiangeng Xu, et al.
5

Advances in LiDAR sensors provide rich 3D data that supports 3D scene understanding. However, due to occlusion and signal miss, LiDAR point clouds are in practice 2.5D as they cover only partial underlying shapes, which poses a fundamental challenge to 3D perception. To tackle the challenge, we present a novel LiDAR-based 3D object detection model, dubbed Behind the Curtain Detector (BtcDet), which learns the object shape priors and estimates the complete object shapes that are partially occluded (curtained) in point clouds. BtcDet first identifies the regions that are affected by occlusion and signal miss. In these regions, our model predicts the probability of occupancy that indicates if a region contains object shapes. Integrated with this probability map, BtcDet can generate high-quality 3D proposals. Finally, the probability of occupancy is also integrated into a proposal refinement module to generate the final bounding boxes. Extensive experiments on the KITTI Dataset and the Waymo Open Dataset demonstrate the effectiveness of BtcDet. Particularly, for the 3D detection of both cars and cyclists on the KITTI benchmark, BtcDet surpasses all of the published state-of-the-art methods by remarkable margins. Code is released (https://github.com/Xharlie/BtcDethttps://github.com/Xharlie/BtcDet).

READ FULL TEXT

page 1

page 2

page 3

page 5

page 17

page 18

page 20

research
03/03/2023

BSH-Det3D: Improving 3D Object Detection with BEV Shape Heatmap

The progress of LiDAR-based 3D object detection has significantly enhanc...
research
03/01/2020

ZoomNet: Part-Aware Adaptive Zooming Neural Network for 3D Object Detection

3D object detection is an essential task in autonomous driving and robot...
research
07/20/2022

Multimodal Transformer for Automatic 3D Annotation and Object Detection

Despite a growing number of datasets being collected for training 3D obj...
research
04/24/2023

Once Detected, Never Lost: Surpassing Human Performance in Offline LiDAR based 3D Object Detection

This paper aims for high-performance offline LiDAR-based 3D object detec...
research
12/09/2021

3D-VField: Learning to Adversarially Deform Point Clouds for Robust 3D Object Detection

As 3D object detection on point clouds relies on the geometrical relatio...
research
11/09/2018

RoarNet: A Robust 3D Object Detection based on RegiOn Approximation Refinement

We present RoarNet, a new approach for 3D object detection from a 2D ima...
research
11/25/2020

Unsupervised Object Detection with LiDAR Clues

Despite the importance of unsupervised object detection, to the best of ...

Please sign up or login with your details

Forgot password? Click here to reset