# A Data-efficient Framework for Robotics Large-scale LiDAR Scene Parsing

arXiv: https://arxiv.org/abs/2312.02208 · HTML: https://mlsanity.com/paper/2312.02208/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 16 of its 31 references that are in the map.
- First posted on arXiv 3 Dec 2023.

## Builds on (most cited first, 16 of 16)

- [Scalability in Perception for Autonomous Driving: Waymo Open Dataset](https://mlsanity.com/paper/1912.04838/) (2019)
- [PointContrast: Unsupervised Pre-training for 3D Point Cloud Understanding](https://mlsanity.com/paper/2007.10985/) (2020)
- [PointGroup: Dual-Set Point Grouping for 3D Instance Segmentation](https://mlsanity.com/paper/2004.01658/) (2020)
- [Exploring Data-Efficient 3D Scene Understanding with Contrastive Scene Contexts](https://mlsanity.com/paper/2012.09165/) (2020)
- [One Thing One Click: A Self-Training Approach for Weakly Supervised 3D Semantic Segmentation](https://mlsanity.com/paper/2104.02246/) (2021)
- [Towards Semantic Segmentation of Urban-Scale 3D Point Clouds: A Dataset, Benchmarks and Challenges](https://mlsanity.com/paper/2009.03137/) (2020)
- [DyCo3D: Robust Instance Segmentation of 3D Point Clouds through Dynamic Convolution](https://mlsanity.com/paper/2011.13328/) (2020)
- [FG-Net: Fast Large-Scale LiDAR Point Clouds Understanding Network Leveraging Correlated Feature Mining and Geometric-Aware Modelling](https://mlsanity.com/paper/2012.09439/) (2020)
- [3D Guided Weakly Supervised Semantic Segmentation](https://mlsanity.com/paper/2012.00242/) (2020)
- [An Integrated LiDAR-SLAM System for Complex Environment with Noisy Point Clouds](https://mlsanity.com/paper/2212.05705/) (2022)
- [Semi-Supervised Confidence-Level-based Contrastive Discrimination for Class-Imbalanced Semantic Segmentation](https://mlsanity.com/paper/2211.15066/) (2022)
- [A Light-Weight LiDAR-Inertial SLAM System with Loop Closing](https://mlsanity.com/paper/2212.05743/) (2022)
- [An Enhanced LiDAR-Inertial SLAM System for Robotics Localization and Mapping](https://mlsanity.com/paper/2212.14209/) (2022)
- [A LiDAR-Inertial-Visual SLAM System with Loop Detection](https://mlsanity.com/paper/2301.05604/) (2023)
- [An Integrated Visual System for Unmanned Aerial Vehicles Tracking and Landing on the Ground Vehicles](https://mlsanity.com/paper/2301.00198/) (2022)
- [Learning-Based Defect Recognitions for Autonomous UAV Inspections](https://mlsanity.com/paper/2302.06093/) (2023)

## Abstract (arXiv)

Existing state-of-the-art 3D point clouds understanding methods only perform well in a fully supervised manner. To the best of our knowledge, there exists no unified framework which simultaneously solves the downstream high-level understanding tasks, especially when labels are extremely limited. This work presents a general and simple framework to tackle point clouds understanding when labels are limited. We propose a novel unsupervised region expansion based clustering method for generating clusters. More importantly, we innovatively propose to learn to merge the over-divided clusters based on the local low-level geometric property similarities and the learned high-level feature similarities supervised by weak labels. Hence, the true weak labels guide pseudo labels merging taking both geometric and semantic feature correlations into consideration. Finally, the self-supervised reconstruction and data augmentation optimization modules are proposed to guide the propagation of labels among semantically similar points within a scene. Experimental Results demonstrate that our framework has the best performance among the three most important weakly supervised point clouds understanding tasks including semantic segmentation, instance segmentation, and object detection even when limited points are labeled, under the data-efficient settings for the large-scale 3D semantic scene parsing. The developed techniques have postentials to be applied to downstream tasks for better representations in robotic manipulation and robotic autonomous navigation. Codes and models are publicly available at: https://github.com/KangchengLiu.