OpenBox: Annotate Any Bounding Boxes in 3D
This addresses annotation challenges in autonomous driving, offering a more efficient solution, though it appears incremental as it builds on existing unsupervised and open-vocabulary detection methods.
The paper tackles the problem of high annotation costs and limited object recognition in 3D object detection for autonomous driving by proposing OpenBox, a two-stage automatic annotation pipeline that uses a 2D vision foundation model to generate high-quality 3D bounding boxes without self-training, achieving improved accuracy and efficiency on datasets like Waymo, Lyft, and nuScenes.
Unsupervised and open-vocabulary 3D object detection has recently gained attention, particularly in autonomous driving, where reducing annotation costs and recognizing unseen objects are critical for both safety and scalability. However, most existing approaches uniformly annotate 3D bounding boxes, ignore objects' physical states, and require multiple self-training iterations for annotation refinement, resulting in suboptimal quality and substantial computational overhead. To address these challenges, we propose OpenBox, a two-stage automatic annotation pipeline that leverages a 2D vision foundation model. In the first stage, OpenBox associates instance-level cues from 2D images processed by a vision foundation model with the corresponding 3D point clouds via cross-modal instance alignment. In the second stage, it categorizes instances by rigidity and motion state, then generates adaptive bounding boxes with class-specific size statistics. As a result, OpenBox produces high-quality 3D bounding box annotations without requiring self-training. Experiments on the Waymo Open Dataset, the Lyft Level 5 Perception dataset, and the nuScenes dataset demonstrate improved accuracy and efficiency over baselines.