Self-Supervised Pre-Training Boosts Semantic Scene Segmentation on LiDAR Data

Abstract

Airborne LiDAR systems have the capability to capture the Earth's surface by generating extensive point cloud data comprised of points mainly defined by 3D coordinates. However, labeling such points for supervised learning tasks is time-consuming. As a result, there is a need to investigate techniques that can learn from unlabeled data to significantly reduce the number of annotated samples. In this work, we propose to train a self-supervised encoder with Barlow Twins and use it as a pre-trained network in the task of semantic scene segmentation. The experimental results demonstrate that our unsupervised pre-training boosts performance once fine-tuned on the supervised task, especially for under-represented categories.

arXiv commentInternational conference Machine Vision Applications 2023

References in corpus (8)

• A Simple Framework for Contrastive Learning of Visual Representations
• Bootstrap your own latent: A new approach to self-supervised Learning
• Unsupervised Learning of Visual Features by Contrasting Cluster Assignments
• Barlow Twins: Self-Supervised Learning via Redundancy Reduction
• Linking Points With Labels in 3D: A Review of Point Cloud Semantic Segmentation
• Deep Clustering for Unsupervised Learning of Visual Features
• SemDeDup: Data-efficient learning at web-scale through semantic deduplication
• Object Segmentation of Cluttered Airborne LiDAR Point Clouds