PatchContrast: Self-Supervised Pre-training for 3D Object Detection

TL;DR

PatchContrast is a self-supervised pre-training framework for 3D object detection that leverages proposal and patch levels of abstraction to improve detection accuracy without requiring labeled data.

Contribution

Introduces a novel self-supervised pre-training method using proposal and patch levels for 3D object detection in point clouds.

Findings

Outperforms state-of-the-art models on three datasets.

Enhances downstream 3D detection performance.

Effective across various backbone architectures.

Abstract

Accurately detecting objects in the environment is a key challenge for autonomous vehicles. However, obtaining annotated data for detection is expensive and time-consuming. We introduce PatchContrast, a novel self-supervised point cloud pre-training framework for 3D object detection. We propose to utilize two levels of abstraction to learn discriminative representation from unlabeled data: proposal-level and patch-level. The proposal-level aims at localizing objects in relation to their surroundings, whereas the patch-level adds information about the internal connections between the object's components, hence distinguishing between different objects based on their individual components. We demonstrate how these levels can be integrated into self-supervised pre-training for various backbones to enhance the downstream 3D detection task. We show that our method outperforms existing state-of-the-art models on three commonly-used 3D detection datasets.