SPU-IMR: Self-supervised Arbitrary-scale Point Cloud Upsampling via Iterative Mask-recovery Network

2025-02-26

AI Summary

Point cloud upsampling aims to generate dense, uniformly distributed, and geometrically faithful point sets from sparse inputs. This work departs from conventional interpolation-based paradigms by reformulating the task as a global shape completion problem. Methodologically, we propose a block-wise masking strategy coupled with a self-supervised iterative reconstruction framework: multiple rounds of stochastic block masking and recovery enable arbitrary-scale upsampling; multi-sequence masking fusion and unsupervised geometric consistency constraints eliminate reliance on paired ground truth or predefined interpolation structures. Evaluated across multiple benchmarks, our approach consistently outperforms both supervised and self-supervised state-of-the-art methods. Quantitative metrics—including Chamfer Distance, Earth Mover’s Distance, and coverage—show significant improvement. Qualitatively, the results exhibit superior density uniformity and structural completeness, particularly in fine-grained geometric details and occluded regions.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Unsupervised & Self-Supervised Learning
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics

Abstract

Point cloud upsampling aims to generate dense and uniformly distributed point sets from sparse point clouds. Existing point cloud upsampling methods typically approach the task as an interpolation problem. They achieve upsampling by performing local interpolation between point clouds or in the feature space, then regressing the interpolated points to appropriate positions. By contrast, our proposed method treats point cloud upsampling as a global shape completion problem. Specifically, our method first divides the point cloud into multiple patches. Then, a masking operation is applied to remove some patches, leaving visible point cloud patches. Finally, our custom-designed neural network iterative completes the missing sections of the point cloud through the visible parts. During testing, by selecting different mask sequences, we can restore various complete patches. A sufficiently dense upsampled point cloud can be obtained by merging all the completed patches. We demonstrate the superior performance of our method through both quantitative and qualitative experiments, showing overall superiority against both existing self-supervised and supervised methods.

Problem

• Self-supervised point cloud upsampling
• Global shape completion via iterative mask-recovery
• Generating dense, uniform point sets

Innovation

• Self-supervised arbitrary-scale upsampling
• Iterative mask-recovery network
• Global shape completion approach