Label-Guided Auxiliary Training Improves 3D Object Detector
This work addresses the problem of improving 3D object detection accuracy for applications like robotics and autonomous driving, but it is incremental as it builds upon existing detectors with an auxiliary network.
The paper tackles the challenge of 3D object detection from point clouds by proposing a Label-Guided auxiliary training method (LG3D) that enhances feature learning in existing detectors, resulting in improvements such as 2.5% and 3.1% mAP gains on VoteNet for SUN RGB-D and ScanNetV2 datasets.
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.