SCP: Scene Completion Pre-training for 3D Object Detection

TL;DR

This paper introduces SCP, a pre-training method for 3D object detection that improves model initialization and reduces labeled data requirements by completing scene point clouds, benefiting autonomous driving applications.

Contribution

SCP is a novel scene completion pre-training approach that enhances 3D detectors without needing extra datasets or extensive labeling, enabling effective training with less labeled data.

Findings

Achieves comparable detection performance with only 20% labeled data.

Improves point cloud model initialization through scene completion.

Does not require additional datasets or labeling efforts.

Abstract

3D object detection using LiDAR point clouds is a fundamental task in the fields of computer vision, robotics, and autonomous driving. However, existing 3D detectors heavily rely on annotated datasets, which are both time-consuming and prone to errors during the process of labeling 3D bounding boxes. In this paper, we propose a Scene Completion Pre-training (SCP) method to enhance the performance of 3D object detectors with less labeled data. SCP offers three key advantages: (1) Improved initialization of the point cloud model. By completing the scene point clouds, SCP effectively captures the spatial and semantic relationships among objects within urban environments. (2) Elimination of the need for additional datasets. SCP serves as a valuable auxiliary network that does not impose any additional efforts or data requirements on the 3D detectors. (3) Reduction of the amount of labeled data for detection. With the help of SCP, the existing state-of-the-art 3D detectors can achieve comparable performance while only relying on 20% labeled data.