SurfFill: Completion of LiDAR Point Clouds via Gaussian Surfel Splatting

2025-12-02

AI Summary

LiDAR point clouds achieve high geometric accuracy on planar surfaces but suffer from geometric detail loss in fine structures, sharp edges, and dark, light-absorbing materials; photogrammetry complements texture and detail but exhibits poor geometric fidelity in textureless regions. To address these limitations, we propose SurfFill—a Gaussian surfel-based multimodal point cloud completion framework. Our method introduces two key innovations: (1) localized Gaussian surfel reconstruction constrained exclusively to blurred, defective regions, and (2) a LiDAR beam divergence–informed density gradient heuristic for precise defect region identification. By integrating a divide-and-conquer strategy with multi-source data fusion, SurfFill robustly restores fine-scale geometry and enhances overall geometric accuracy. Extensive evaluations on both synthetic and real-world datasets demonstrate that SurfFill consistently outperforms state-of-the-art methods in structural completeness and geometric fidelity.

Technology Category

• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs

Abstract

LiDAR-captured point clouds are often considered the gold standard in active 3D reconstruction. While their accuracy is exceptional in flat regions, the capturing is susceptible to miss small geometric structures and may fail with dark, absorbent materials. Alternatively, capturing multiple photos of the scene and applying 3D photogrammetry can infer these details as they often represent feature-rich regions. However, the accuracy of LiDAR for featureless regions is rarely reached. Therefore, we suggest combining the strengths of LiDAR and camera-based capture by introducing SurfFill: a Gaussian surfel-based LiDAR completion scheme. We analyze LiDAR capturings and attribute LiDAR beam divergence as a main factor for artifacts, manifesting mostly at thin structures and edges. We use this insight to introduce an ambiguity heuristic for completed scans by evaluating the change in density in the point cloud. This allows us to identify points close to missed areas, which we can then use to grow additional points from to complete the scan. For this point growing, we constrain Gaussian surfel reconstruction [Huang et al. 2024] to focus optimization and densification on these ambiguous areas. Finally, Gaussian primitives of the reconstruction in ambiguous areas are extracted and sampled for points to complete the point cloud. To address the challenges of large-scale reconstruction, we extend this pipeline with a divide-and-conquer scheme for building-sized point cloud completion. We evaluate on the task of LiDAR point cloud completion of synthetic and real-world scenes and find that our method outperforms previous reconstruction methods.

Problem

• Completes LiDAR point clouds using Gaussian surfel splatting
• Addresses missing details in thin structures and edges
• Combines LiDAR and camera strengths for accurate reconstruction

Innovation

• Combines LiDAR and camera strengths via Gaussian surfel completion
• Uses density changes to identify ambiguous areas for point growth
• Extends pipeline with divide-and-conquer for large-scale reconstruction