Distribution Discrepancy and Feature Heterogeneity for Active 3D Object Detection

Huang-Yu Chen, Jia-Fong Yeh, Jiawe Jiawei, Pin-Hsuan Peng, Winston H. Hsu
Proceedings of The 8th Conference on Robot Learning, PMLR 270:4967-4980, 2025.

Abstract

LiDAR-based 3D object detection is a critical technology for the development of autonomous driving and robotics. However, the high cost of data annotation limits its advancement. We propose a novel and effective active learning (AL) method called Distribution Discrepancy and Feature Heterogeneity (DDFH), which simultaneously considers geometric features and model embeddings, assessing information from both the instance-level and frame-level perspectives. Distribution Discrepancy evaluates the difference and novelty of instances within the unlabeled and labeled distributions, enabling the model to learn efficiently with limited data. Feature Heterogeneity ensures the heterogeneity of intra-frame instance features, maintaining feature diversity while avoiding redundant or similar instances, thus minimizing annotation costs. Finally, multiple indicators are efficiently aggregated using Quantile Transform, providing a unified measure of informativeness. Extensive experiments demonstrate that DDFH outperforms the current state-of-the-art (SOTA) methods on the KITTI and Waymo datasets, effectively reducing the bounding box annotation cost by 56.3% and showing robustness when working with both one-stage and two-stage models.

Cite this Paper


BibTeX
@InProceedings{pmlr-v270-chen25h, title = {Distribution Discrepancy and Feature Heterogeneity for Active 3D Object Detection}, author = {Chen, Huang-Yu and Yeh, Jia-Fong and Jiawei, Jiawe and Peng, Pin-Hsuan and Hsu, Winston H.}, booktitle = {Proceedings of The 8th Conference on Robot Learning}, pages = {4967--4980}, year = {2025}, editor = {Agrawal, Pulkit and Kroemer, Oliver and Burgard, Wolfram}, volume = {270}, series = {Proceedings of Machine Learning Research}, month = {06--09 Nov}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v270/main/assets/chen25h/chen25h.pdf}, url = {https://proceedings.mlr.press/v270/chen25h.html}, abstract = {LiDAR-based 3D object detection is a critical technology for the development of autonomous driving and robotics. However, the high cost of data annotation limits its advancement. We propose a novel and effective active learning (AL) method called Distribution Discrepancy and Feature Heterogeneity (DDFH), which simultaneously considers geometric features and model embeddings, assessing information from both the instance-level and frame-level perspectives. Distribution Discrepancy evaluates the difference and novelty of instances within the unlabeled and labeled distributions, enabling the model to learn efficiently with limited data. Feature Heterogeneity ensures the heterogeneity of intra-frame instance features, maintaining feature diversity while avoiding redundant or similar instances, thus minimizing annotation costs. Finally, multiple indicators are efficiently aggregated using Quantile Transform, providing a unified measure of informativeness. Extensive experiments demonstrate that DDFH outperforms the current state-of-the-art (SOTA) methods on the KITTI and Waymo datasets, effectively reducing the bounding box annotation cost by 56.3% and showing robustness when working with both one-stage and two-stage models.} }
Endnote
%0 Conference Paper %T Distribution Discrepancy and Feature Heterogeneity for Active 3D Object Detection %A Huang-Yu Chen %A Jia-Fong Yeh %A Jiawe Jiawei %A Pin-Hsuan Peng %A Winston H. Hsu %B Proceedings of The 8th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2025 %E Pulkit Agrawal %E Oliver Kroemer %E Wolfram Burgard %F pmlr-v270-chen25h %I PMLR %P 4967--4980 %U https://proceedings.mlr.press/v270/chen25h.html %V 270 %X LiDAR-based 3D object detection is a critical technology for the development of autonomous driving and robotics. However, the high cost of data annotation limits its advancement. We propose a novel and effective active learning (AL) method called Distribution Discrepancy and Feature Heterogeneity (DDFH), which simultaneously considers geometric features and model embeddings, assessing information from both the instance-level and frame-level perspectives. Distribution Discrepancy evaluates the difference and novelty of instances within the unlabeled and labeled distributions, enabling the model to learn efficiently with limited data. Feature Heterogeneity ensures the heterogeneity of intra-frame instance features, maintaining feature diversity while avoiding redundant or similar instances, thus minimizing annotation costs. Finally, multiple indicators are efficiently aggregated using Quantile Transform, providing a unified measure of informativeness. Extensive experiments demonstrate that DDFH outperforms the current state-of-the-art (SOTA) methods on the KITTI and Waymo datasets, effectively reducing the bounding box annotation cost by 56.3% and showing robustness when working with both one-stage and two-stage models.
APA
Chen, H., Yeh, J., Jiawei, J., Peng, P. & Hsu, W.H.. (2025). Distribution Discrepancy and Feature Heterogeneity for Active 3D Object Detection. Proceedings of The 8th Conference on Robot Learning, in Proceedings of Machine Learning Research 270:4967-4980 Available from https://proceedings.mlr.press/v270/chen25h.html.

Related Material