STONE: A Submodular Optimization Framework for Active 3D Object Detection

TL;DR

This paper introduces STONE, a submodular optimization framework that significantly reduces labeling costs for 3D object detection by actively selecting the most informative LiDAR point cloud data, addressing data imbalance and coverage.

Contribution

It presents a novel submodular optimization formulation tailored for active 3D object detection, improving labeling efficiency and detection performance.

Findings

Achieves state-of-the-art performance on active 3D detection tasks.

Reduces labeling effort while maintaining high detection accuracy.

Demonstrates high computational efficiency compared to existing methods.

Abstract

3D object detection is fundamentally important for various emerging applications, including autonomous driving and robotics. A key requirement for training an accurate 3D object detector is the availability of a large amount of LiDAR-based point cloud data. Unfortunately, labeling point cloud data is extremely challenging, as accurate 3D bounding boxes and semantic labels are required for each potential object. This paper proposes a unified active 3D object detection framework, for greatly reducing the labeling cost of training 3D object detectors. Our framework is based on a novel formulation of submodular optimization, specifically tailored to the problem of active 3D object detection. In particular, we address two fundamental challenges associated with active 3D object detection: data imbalance and the need to cover the distribution of the data, including LiDAR-based point cloud data of varying difficulty levels. Extensive experiments demonstrate that our method achieves state-of-the-art performance with high computational efficiency compared to existing active learning methods. The code is available at https://github.com/RuiyuM/STONE.