SALT: A Flexible Semi-Automatic Labeling Tool for General LiDAR Point Clouds with Cross-Scene Adaptability and 4D Consistency

2025-03-31

AI Summary

To address low annotation efficiency and poor consistency in cross-scene and cross-temporal LiDAR point cloud labeling, this paper proposes the first general-purpose semi-automatic pre-segmentation framework for LiDAR. Methodologically: (1) it introduces a novel zero-shot data alignment paradigm that maps raw 3D point clouds to pseudo-images to enable compatibility with vision foundation models; (2) it designs a 4D consistency prompting strategy and a 4D non-maximum suppression module to enhance the spatiotemporal segmentation robustness of SAM2. The key contributions are domain-agnostic generalization without requiring point cloud annotation priors and explicit 4D temporal consistency modeling. Experiments demonstrate that our method achieves an 18.4% higher panoptic quality (PQ) than state-of-the-art zero-shot methods on SemanticKITTI. On multi-source low-resolution LiDAR data, it attains 40–50% of human annotation performance, significantly improving labeling efficiency and scalability.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Semi-Supervised Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Large language models for search
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling

Abstract

We propose a flexible Semi-Automatic Labeling Tool (SALT) for general LiDAR point clouds with cross-scene adaptability and 4D consistency. Unlike recent approaches that rely on camera distillation, SALT operates directly on raw LiDAR data, automatically generating pre-segmentation results. To achieve this, we propose a novel zero-shot learning paradigm, termed data alignment, which transforms LiDAR data into pseudo-images by aligning with the training distribution of vision foundation models. Additionally, we design a 4D-consistent prompting strategy and 4D non-maximum suppression module to enhance SAM2, ensuring high-quality, temporally consistent presegmentation. SALT surpasses the latest zero-shot methods by 18.4% PQ on SemanticKITTI and achieves nearly 40-50% of human annotator performance on our newly collected low-resolution LiDAR data and on combined data from three LiDAR types, significantly boosting annotation efficiency. We anticipate that SALT's open-sourcing will catalyze substantial expansion of current LiDAR datasets and lay the groundwork for the future development of LiDAR foundation models. Code is available at https://github.com/Cavendish518/SALT.

Problem

• Develops a semi-automatic tool for labeling LiDAR point clouds
• Ensures cross-scene adaptability and 4D consistency in labeling
• Improves annotation efficiency and reduces human effort significantly

Innovation

• Zero-shot learning via data alignment
• 4D-consistent prompting strategy
• 4D non-maximum suppression module