HeAL3D: Heuristical-enhanced Active Learning for 3D Object Detection

2025-05-01

AI Summary

Addressing the high annotation cost of 3D object detection in autonomous driving and the difficulty of selecting informative samples under uncontrolled real-world conditions, this paper proposes a heuristic-enhanced active learning framework. Our method pioneers the integration of physically interpretable, practical features—such as object distance and point cloud cardinality—with localization and classification uncertainties, forming a multi-source weighted uncertainty metric. Coupled with efficient point-cloud detectors (e.g., PointPillars), we design a computationally lightweight sampling strategy. On the KITTI benchmark, our approach achieves full-supervision baseline mAP using only 24% of labeled data, matching state-of-the-art performance. The core contribution lies in bridging theory and practice: systematically incorporating domain-specific priors into uncertainty modeling significantly enhances the practicality and generalizability of active learning in real-world autonomous driving scenarios.

Technology Category

• Machine Learning: Active Learning
• Reasoning under Uncertainty: Uncertainty Representations
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling

Abstract

Active Learning has proved to be a relevant approach to perform sample selection for training models for Autonomous Driving. Particularly, previous works on active learning for 3D object detection have shown that selection of samples in uncontrolled scenarios is challenging. Furthermore, current approaches focus exclusively on the theoretical aspects of the sample selection problem but neglect the practical insights that can be obtained from the extensive literature and application of 3D detection models. In this paper, we introduce HeAL (Heuristical-enhanced Active Learning for 3D Object Detection) which integrates those heuristical features together with Localization and Classification to deliver the most contributing samples to the model's training. In contrast to previous works, our approach integrates heuristical features such as object distance and point-quantity to estimate the uncertainty, which enhance the usefulness of selected samples to train detection models. Our quantitative evaluation on KITTI shows that HeAL presents competitive mAP with respect to the State-of-the-Art, and achieves the same mAP as the full-supervised baseline with only 24% of the samples.

Problem

• Improving sample selection for 3D object detection in autonomous driving
• Addressing challenges in uncontrolled scenarios for active learning
• Integrating heuristical features to enhance training sample usefulness

Innovation

• Integrates heuristical features with Localization and Classification
• Uses object distance and point-quantity for uncertainty estimation
• Achieves competitive mAP with only 24% of samples