通过双重机制降低3D目标检测标注成本，提升小样本学习效率。
Distribution Discrepancy and Feature Heterogeneity for Active 3D Object Detection
- 结合几何特征与模型嵌入，从实例和帧级双视角评估信息量。
- 在KITTI和Waymo上减少56.3%标注成本，性能超越当前最优方法。
- 适用于一阶段与二阶段模型，有效避免冗余标注，适合资源受限场景。
基于激光雷达的3D目标检测是自动驾驶与机器人发展的关键技术，但数据标注成本高昂制约其进展。本文提出一种新型有效的主动学习方法——分布差异与特征异质性（DDFH），同时考虑几何特征与模型嵌入，从实例级和帧级两个层面评估信息量。分布差异衡量未标记与已标记数据分布间的差异与新颖性，帮助模型在有限数据下高效学习。特征异质性确保帧内实例特征的多样性，避免重复或相似样本，从而降低标注开销。最终通过分位数变换高效聚合多个指标，获得统一的信息量度量。大量实验表明，DDFH在KITTI和Waymo数据集上均优于当前最先进方法，标注框成本降低56.3%，且在单阶段与双阶段模型上均表现稳健。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。