• Home
  • Search
  • Self-Supervised Learning for Point Clouds through Multi-crop Mutual Prediction
  • https://doi.org/10.1109/prml59573.2023.10348385Copy DOI Icon

Self-Supervised Learning for Point Clouds through Multi-crop Mutual Prediction

  • Aug 4, 2023
  • Changyu Zeng +4 more
Show More
  • Abstract
  • Literature Map
  • References
  • Similar Papers
Abstract

As the demand for high-level autonomous driving continues to grow, efficient processing of large amount of point cloud data from LiDAR sensors has become a severe issue. Despite its great success, it is extremely difficult, if not impossible, to apply supervised learning in many real-world applications with natural scenes lacking annotations. In addition, removing label dependency and improving algorithm generalization ability are the current key research challenges in 3D deep learning. We propose a crop-based self-supervised learning framework for point clouds called Multi-Crop Mutual Prediction (MCMP). By cropping the original point clouds into different number of multiscale patches and making patches from the same object as close as possible through dynamic updates of online prototypes, the encoder is well able to maintain consistency between both local and global features. MCMP framework is model-agnostic so it is convenient to integrate it into encoders of any architectures. We pre-train three commonly used point cloud feature extraction networks on a single object-level dataset MondelNet40 and evaluate MCMP on three downstream tasks, namely, object classification, partial segmentation and semantic segmentation. Experimental results show that MCMP outperforms the previous crop-based self-supervised approaches in the object classification and semantic segmentation tasks and produces comparable results with the supervised benchmark methods.

Similar Papers
  • PDF
  • Research Article
  • Citations27

A Point Cloud Registration Algorithm Based on Feature Extraction and Matching

  • Dec 24, 2018
  • Mathematical Problems in Engineering
  • Yongshan Liu +4
  • Research Article
  • Citations1

SeqConv-Net: A Deep Learning Segmentation Framework for Airborne LiDAR Point Clouds Based on Spatially Ordered Sequences

  • Jun 01, 2025
  • Remote Sensing
  • Bin Guo +4
  • Research Article
  • Citations8

Point-to-Pixel Prompting for Point Cloud Analysis With Pre-Trained Image Models.

  • Jun 01, 2024
  • IEEE Transactions on Pattern Analysis and Machine Intelligence
  • Ziyi Wang +4
  • Research Article
  • Citations28

Improving performance of deep learning models for 3D point cloud semantic segmentation via attention mechanisms

  • Jun 26, 2022
  • Computers & Graphics
  • Vazgen Vanian +2
  • Research Article
  • Citations26

Transfer learning and performance enhancement techniques for deep semantic segmentation of built heritage point clouds

  • Jul 14, 2021
  • Virtual Archaeology Review
  • Francesca Matrone +1
  • PDF
  • Research Article
  • Citations11

Iterative automatic global registration algorithm for multi-view point cloud of underground tunnel space

  • Mar 01, 2021
  • Measurement and Control
  • Ming Guo +6
  • Research Article
  • Citations8

Improved building MEP systems semantic segmentation in point clouds using a novel multi-class dataset and local–global vector transformer network

  • Aug 07, 2024
  • Journal of Building Engineering
  • Shuju Jing +4
  • Conference Article
  • Citations104

Image-to-Lidar Self-Supervised Distillation for Autonomous Driving Data

  • Jun 01, 2022
  • Corentin Sautier +5
  • PDF
  • Research Article
  • Citations10

Exploiting Structured CNNs for Semantic Segmentation of Unstructured Point Clouds from LiDAR Sensor

  • Sep 10, 2021
  • Remote Sensing
  • Muhammad Ibrahim +3
  • Conference Article
  • Citations27

Continual Learning for Class- and Domain-Incremental Semantic Segmentation

  • Jul 11, 2021
  • Tobias Kalb +3
  • Conference Article
  • Citations2

Self-Supervised Feature Learning from Partial Point Clouds via Pose Disentanglement

  • Oct 23, 2022
  • Meng-Shiun Tsai +3
  • Book Chapter
  • Citations44

Chapter 12 - Semantic scene segmentation for robotics

  • Jan 01, 2022
  • Deep Learning for Robot Perception and Cognition
  • Juana Valeria Hurtado +1
  • Conference Article

Pipe environment elbow features extracting based on 2D Point Cloud

  • Nov 18, 2022
  • Yu Ding +2
  • Research Article
  • Citations26

Deep learning-based semantic segmentation of three-dimensional point cloud: a comprehensive review

  • Jan 17, 2024
  • International Journal of Remote Sensing
  • Dheerendra Pratap Singh +1
  • PDF
  • Research Article
  • Citations1

An End-to-End Geometric Characterization-aware Semantic Instance Segmentation Network for ALS Point Clouds

  • Jun 11, 2024
  • The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
  • Jinhong Wang +1
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.