HeAL3D: Heuristical-enhanced Active Learning for 3D Object Detection
Not on the map. Why?
Abstract (arXiv)
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.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 28 of its 44 references that are in the map. First posted on arXiv 1 May 2025.
Builds on
28 of 44 references.
References · 28
- 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,248
- 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,523
- 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,448
- 2019Scalability in Perception for Autonomous Driving: Waymo Open Datasetcited by 2,451
- 2019PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detectioncited by 891
- 2017Deep Bayesian Active Learning with Image Datacited by 819
- 2019Deep Batch Active Learning by Diverse, Uncertain Gradient Lower Boundscited by 610
- 2019Learning Loss for Active Learningcited by 336
- 2018Orthographic Feature Transform for Monocular 3D Object Detectioncited by 202
- 2021RSN: Range Sparse Net for Efficient, Accurate LiDAR 3D Object Detectioncited by 115
- 2021Active Learning for Deep Object Detection via Probabilistic Modelingcited by 79
- 2020Sequential Graph Convolutional Network for Active Learningcited by 69
16 more
- 2022Active Learning by Feature Mixingcited by 63
- 2019Bayesian Generative Active Deep Learningcited by 57
- 2022Entropy-based Active Learning for Object Detection with Progressive Diversity Constraintcited by 53
- 2018Localization-Aware Active Learning for Object Detectioncited by 48
- 2019Deep Active Learning for Efficient Training of a LiDAR 3D Object Detectorcited by 45
- 2024MAN TruckScenes: A multimodal dataset for autonomous trucking in diverse conditionscited by 41
- 2020Deep Active Learning for Biased Datasets via Fisher Kernel Self-Supervisioncited by 32
- 2021Consistency-based Active Learning for Object Detectioncited by 32
- 2021Not All Labels Are Equal: Rationalizing The Labeling Costs for Training Object Detectioncited by 27
- 2023Exploring Active 3D Object Detection from a Generalization Perspectivecited by 26
- 2023Bi3D: Bi-domain Active Learning for Cross-domain 3D Object Detectioncited by 22
- 2024The Why, When, and How to Use Active Learning in Large-Data-Driven 3D Object Detection for Safe Autonomous Driving: An Empirical Explorationcited by 17
- 2024ActiveAnno3D -- An Active Learning Framework for Multi-Modal 3D Object Detectioncited by 17
- 2023KECOR: Kernel Coding Rate Maximization for Active 3D Object Detectioncited by 13
- 2021Bayesian Active Learning by Disagreements: A Geometric Perspectivecited by 9
- 2023BaSAL: Size-Balanced Warm Start Active Learning for LiDAR Semantic Segmentationcited by 8
Built on by
Not cited yet.