Label-Guided Auxiliary Training Improves 3D Object Detector

TL;DR

This paper introduces LG3D, a label-guided auxiliary training method that enhances 3D object detection accuracy by improving feature learning without increasing inference costs.

Contribution

The paper presents two novel modules for auxiliary training that boost 3D detection performance while maintaining no extra inference overhead.

Findings

LG3D improves VoteNet by 2.5% mAP on SUN RGB-D.

LG3D improves VoteNet by 3.1% mAP on ScanNetV2.

The auxiliary network is discarded during inference, ensuring no additional computational cost.

Abstract

Detecting 3D objects from point clouds is a practical yet challenging task that has attracted increasing attention recently. In this paper, we propose a Label-Guided auxiliary training method for 3D object detection (LG3D), which serves as an auxiliary network to enhance the feature learning of existing 3D object detectors. Specifically, we propose two novel modules: a Label-Annotation-Inducer that maps annotations and point clouds in bounding boxes to task-specific representations and a Label-Knowledge-Mapper that assists the original features to obtain detection-critical representations. The proposed auxiliary network is discarded in inference and thus has no extra computational cost at test time. We conduct extensive experiments on both indoor and outdoor datasets to verify the effectiveness of our approach. For example, our proposed LG3D improves VoteNet by 2.5% and 3.1% mAP on the SUN RGB-D and ScanNetV2 datasets, respectively.