• Home
  • Search
  • ST3D++: Denoised Self-Training for Unsupervised Domain Adaptation on 3D Object Detection.
  • Cite Icon50
  • https://doi.org/10.1109/tpami.2022.3216606Copy DOI Icon

ST3D++: Denoised Self-Training for Unsupervised Domain Adaptation on 3D Object Detection.

  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

In this paper, we present a self-training method, named ST3D++, with a holistic pseudo label denoising pipeline for unsupervised domain adaptation on 3D object detection. ST3D++ aims at reducing noise in pseudo label generation as well as alleviating the negative impacts of noisy pseudo labels on model training. First, ST3D++ pre-trains the 3D object detector on the labeled source domain with random object scaling (ROS) which is designed to reduce target domain pseudo label noise arising from object scale bias of the source domain. Then, the detector is progressively improved through alternating between generating pseudo labels and training the object detector with pseudo-labeled target domain data. Here, we equip the pseudo label generation process with a hybrid quality-aware triplet memory to improve the quality and stability of generated pseudo labels. Meanwhile, in the model training stage, we propose a source data assisted training strategy and a curriculum data augmentation policy to effectively rectify noisy gradient directions and avoid model over-fitting to noisy pseudo labeled data. These specific designs enable the detector to be trained on meticulously refined pseudo labeled target data with denoised training signals, and thus effectively facilitate adapting an object detector to a target domain without requiring annotations. Finally, our method is assessed on four 3D benchmark datasets (i.e., Waymo, KITTI, Lyft, and nuScenes) for three common categories (i.e., car, pedestrian and bicycle). ST3D++ achieves state-of-the-art performance on all evaluated settings, outperforming the corresponding baseline by a large margin (e.g., 9.6% ∼ 38.16% on Waymo → KITTI in terms of AP[Formula: see text]), and even surpasses the fully supervised oracle results on the KITTI 3D object detection benchmark with target prior. Code is available at https://github.com/CVMI-Lab/ST3D.

Similar Papers
  • Conference Article
  • Citations188

ST3D: Self-training for Unsupervised Domain Adaptation on 3D Object Detection

  • Jun 01, 2021
  • Jihan Yang +4
  • Research Article
  • Citations7

Leveraging Self-Paced Semi-Supervised Learning with Prior Knowledge for 3D Object Detection on a LiDAR-Camera System

  • Jan 20, 2023
  • Remote Sensing
  • Pei An +7
  • Research Article

Context-aware reliability exploration for unsupervised domain adaptive person re-identification.

  • Jun 11, 2026
  • Neural networks : the official journal of the International Neural Network Society
  • Jialu Liu +2
  • Conference Article
  • Citations95

Uncertainty-aware Pseudo Label Refinery for Domain Adaptive Semantic Segmentation

  • Oct 01, 2021
  • Yuxi Wang +2
  • Research Article
  • Citations42

Improving pseudo labels with intra-class similarity for unsupervised domain adaptation

  • Feb 02, 2023
  • Pattern Recognition
  • Jie Wang +1
  • Conference Article
  • Citations95

Class-Balanced Pixel-Level Self-Labeling for Domain Adaptive Semantic Segmentation

  • Jun 01, 2022
  • Ruihuang Li +5
  • PDF
  • Research Article
  • Citations5

Improving Object Detection Accuracy with Self-Training Based on Bi-Directional Pseudo Label Recovery

  • Jun 07, 2024
  • Electronics
  • Shoaib Sajid +3
  • Conference Article
  • Citations12

Pairwise Similarity Regularization for Adversarial Domain Adaptation

  • Oct 12, 2020
  • Haotian Wang +5
  • Research Article
  • Citations3

Universal Domain Alignment Framework for Classification and Regression Tasks

  • Jan 01, 2025
  • IEEE Transactions on Artificial Intelligence
  • Xin Wang +2
  • Research Article
  • Citations3

Noise-Resistance Learning via Multi-Granularity Consistency for Unsupervised Domain Adaptive Person Re-Identification

  • Dec 23, 2024
  • ACM Transactions on Multimedia Computing, Communications, and Applications
  • Yangchun Zhu +4
  • Conference Article
  • Citations6

Pseudo-Label Generation-Evaluation Framework For Cross Domain Weakly Supervised Object Detection

  • Sep 19, 2021
  • Shengxiong Ouyang +3
  • Research Article
  • Citations92

SECRET: Self-Consistent Pseudo Label Refinement for Unsupervised Domain Adaptive Person Re-identification

  • Jun 28, 2022
  • Proceedings of the AAAI Conference on Artificial Intelligence
  • Tao He +4
  • Research Article
  • Citations2

MA-ST3D: Motion Associated Self-Training for Unsupervised Domain Adaptation on 3D Object Detection.

  • Jan 01, 2024
  • IEEE transactions on image processing : a publication of the IEEE Signal Processing Society
  • Chi Zhang +3
  • Research Article
  • Citations138

A Free Lunch for Unsupervised Domain Adaptive Object Detection without Source Data

  • May 18, 2021
  • Proceedings of the AAAI Conference on Artificial Intelligence
  • Xianfeng Li +6
  • Book Chapter
  • Citations29

Unsupervised Domain Adaptation for Monocular 3D Object Detection via Self-training

  • Jan 01, 2022
  • Zhenyu Li +6
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.