• Home
  • Search
  • Diversity-based active learning: creating a representative object detection dataset in 3D point clouds
  • Cite Icon1
  • https://doi.org/10.1117/12.2663179Copy DOI Icon

Diversity-based active learning: creating a representative object detection dataset in 3D point clouds

  • Jun 16, 2023
  • Aimee Moses +4 more
Show More
  • Abstract
  • Literature Map
  • Citations
  • Similar Papers
Abstract

Object detection in 3D point clouds is essential in fields such as geospatial intelligence and autonomous driving. The common machine learning problem of scarce labeled training data is even more acute with 3D point cloud data. Active learning provides a framework to prioritize the additional effort to manually annotate unlabeled training data. Most active learning methods for deep learning fall into one of two categories: uncertainty methods and diversity methods. Uncertainty methods select data by assessing model outputs for their confidence and consistency and are therefore dependent on the expected output of each deep learning task. These methods tend to select batches of informative yet highly similar samples to label. Diversity-based active learning aims to create a labeled dataset that is both varied and representative of the remaining unlabeled data. Diversity methods operate directly on the feature representations of the inputs and are thus more flexible with respect to the specifics of the deep learning task. Our current work explores applying diversity methods and uncertainty-diversity hybrid methods to 3D object detection. We evaluate various approaches to incorporate diversity, including K-Medoids clustering, core set selection, and furthest nearest neighbors. We address the high dimensionality of the features extracted from a VoxelNet-based object detector by varying the distance metric used in the active learning algorithms. Furthermore, we compare our results to those obtained using only uncertainty methods. We assess the performance and efficiency of each active learning method in addition to the representativeness and diversity of the labeled datasets produced. We find that hybrid uncertainty-diversity methods outperform other methods in terms of object detection AP50 throughout active learning, annotation efficiency, and class balance.

Similar Papers
  • Book Chapter

Efficient Training Method for Point Cloud-Based Object Detection Models by Combining Environmental Transitions and Active Learning

  • Jan 01, 2023
  • Takumi Yamamoto +4
  • Conference Article
  • Citations8

Experimental Design with Multiple Kernels

  • Nov 01, 2015
  • Hanmo Wang +5
  • Research Article
  • Citations14

RoofN3D: A Database for 3D Building Reconstruction with Deep Learning

  • Jun 01, 2019
  • Photogrammetric Engineering & Remote Sensing
  • Andreas Wichmann +3
  • Research Article
  • Citations31

MVGCN: Multi-View Graph Convolutional Neural Network for Surface Defect Identification Using Three-Dimensional Point Cloud

  • Dec 02, 2022
  • Journal of Manufacturing Science and Engineering
  • Yinan Wang +4
  • Book Chapter
  • Citations21

DEAL: Deep Evidential Active Learning for Image Classification

  • Nov 13, 2021
  • Patrick Hemmer +2
  • Conference Article
  • Citations16

Challenges and Trends of Implementation of 3D Point Cloud Technologies in Building Information Modeling (BIM): Case Studies

  • Jun 17, 2014
  • T Qu +3
  • Research Article
  • Citations75

Batch Mode Active Sampling Based on Marginal Probability Distribution Matching

  • Sep 01, 2013
  • ACM Transactions on Knowledge Discovery from Data
  • Rita Chattopadhyay +5
  • Research Article
  • Citations4

Active learning for deep object detection by fully exploiting unlabeled data

  • Apr 03, 2023
  • Connection Science
  • Feixiang Tan +1
  • Research Article
  • Citations2

Time Series Remote Sensing Image Classification with a Data-Driven Active Deep Learning Approach.

  • Mar 10, 2025
  • Sensors (Basel, Switzerland)
  • Gaoliang Xie +5
  • Research Article
  • Citations19

Active Learning for Crowdsourced QoE Modeling

  • Dec 01, 2018
  • IEEE Transactions on Multimedia
  • Haw-Shiuan Chang +3
  • Book Chapter
  • Citations18

Warm Start Active Learning with Proxy Labels and Selection via Semi-supervised Fine-Tuning

  • Jan 01, 2022
  • Vishwesh Nath +3
  • Research Article

Virtual structured-light three-dimensional point cloud compression with geometric reshaping

  • Mar 18, 2022
  • Journal of Electronic Imaging
  • Yingchun Wu +4
  • Conference Article
  • Citations18

Active Learning for Black-Box Semantic Role Labeling with Neural Factors

  • Aug 01, 2017
  • Chenguang Wang +2
  • Research Article
  • Citations4

Real-Time Environmental Contour Construction Using 3D LiDAR and Image Recognition with Object Removal

  • Dec 01, 2024
  • Remote Sensing
  • Tzu-Jung Wu +2
  • Research Article
  • Citations58

Adaptive Assessment of Power System Transient Stability Based on Active Transfer Learning With Deep Belief Network

  • Apr 01, 2023
  • IEEE Transactions on Automation Science and Engineering
  • Baoqin Li +1
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.