Back to Reality: Weakly-supervised 3D Object Detection with Shape-guided Label Enhancement

TL;DR

This paper introduces a weakly-supervised 3D object detection method called Back to Reality (BR) that uses synthetic shapes and domain adaptation to achieve near fully-supervised performance with minimal labeling effort.

Contribution

The paper presents a novel approach that leverages synthetic 3D shapes and virtual scenes to enhance weak labels, enabling effective 3D detection with less than 5% labeling effort.

Findings

Achieves comparable performance to fully-supervised methods on ScanNet.

Introduces a new challenging benchmark with diverse object sizes.

Uses virtual-to-real domain adaptation to refine labels.

Abstract

In this paper, we propose a weakly-supervised approach for 3D object detection, which makes it possible to train a strong 3D detector with position-level annotations (i.e. annotations of object centers). In order to remedy the information loss from box annotations to centers, our method, namely Back to Reality (BR), makes use of synthetic 3D shapes to convert the weak labels into fully-annotated virtual scenes as stronger supervision, and in turn utilizes the perfect virtual labels to complement and refine the real labels. Specifically, we first assemble 3D shapes into physically reasonable virtual scenes according to the coarse scene layout extracted from position-level annotations. Then we go back to reality by applying a virtual-to-real domain adaptation method, which refine the weak labels and additionally supervise the training of detector with the virtual scenes. Furthermore, we propose a more challenging benckmark for indoor 3D object detection with more diversity in object sizes to better show the potential of BR. With less than 5% of the labeling labor, we achieve comparable detection performance with some popular fully-supervised approaches on the widely used ScanNet dataset. Code is available at: https://github.com/wyf-ACCEPT/BackToReality