Not Every Side Is Equal: Localization Uncertainty Estimation for Semi-Supervised 3D Object Detection
This work improves semi-supervised 3D object detection for applications like autonomous driving by introducing a novel side-aware approach to handle localization uncertainty, representing an incremental advancement over existing methods.
The paper tackles the problem of semi-supervised 3D object detection from point clouds by addressing the equal treatment of all sides in pseudo-labels, which harms performance due to poor localization quality, and proposes a side-aware framework that estimates localization uncertainty and assigns importance accordingly, achieving state-of-the-art results on three datasets with different labeled ratios.
Semi-supervised 3D object detection from point cloud aims to train a detector with a small number of labeled data and a large number of unlabeled data. The core of existing methods lies in how to select high-quality pseudo-labels using the designed quality evaluation criterion. However, these methods treat each pseudo bounding box as a whole and assign equal importance to each side during training, which is detrimental to model performance due to many sides having poor localization quality. Besides, existing methods filter out a large number of low-quality pseudo-labels, which also contain some correct regression values that can help with model training. To address the above issues, we propose a side-aware framework for semi-supervised 3D object detection consisting of three key designs: a 3D bounding box parameterization method, an uncertainty estimation module, and a pseudo-label selection strategy. These modules work together to explicitly estimate the localization quality of each side and assign different levels of importance during the training phase. Extensive experiment results demonstrate that the proposed method can consistently outperform baseline models under different scenes and evaluation criteria. Moreover, our method achieves state-of-the-art performance on three datasets with different labeled ratios.