PUFM++: Point Cloud Upsampling via Enhanced Flow Matching

2025-12-24

AI Summary

This paper addresses point cloud upsampling on sparse, noisy, and incomplete inputs. Methodologically, it introduces an enhanced flow-matching framework featuring: (1) a two-stage flow learning pathway to jointly model coarse-grained structural priors and fine-grained geometric details; (2) a data-driven adaptive time scheduling mechanism to improve dynamic sampling efficiency; (3) the first explicit manifold constraint embedded within flow matching to ensure intrinsic geometric consistency of generated point clouds; and (4) a Recurrent Interface Network (RIN) that strengthens cross-scale feature interaction and enforces geometric regularization. Extensive experiments on both synthetic and real-world scanned datasets demonstrate state-of-the-art performance, achieving significant improvements in Chamfer Distance and F-Score over prior methods, along with superior visual quality. The code and pre-trained models are publicly released.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Learning with Manifolds
• Planning, Routing, and Scheduling: Replanning and Plan Repair

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Recent advances in generative modeling have demonstrated strong promise for high-quality point cloud upsampling. In this work, we present PUFM++, an enhanced flow-matching framework for reconstructing dense and accurate point clouds from sparse, noisy, and partial observations. PUFM++ improves flow matching along three key axes: (i) geometric fidelity, (ii) robustness to imperfect input, and (iii) consistency with downstream surface-based tasks. We introduce a two-stage flow-matching strategy that first learns a direct, straight-path flow from sparse inputs to dense targets, and then refines it using noise-perturbed samples to approximate the terminal marginal distribution better. To accelerate and stabilize inference, we propose a data-driven adaptive time scheduler that improves sampling efficiency based on interpolation behavior. We further impose on-manifold constraints during sampling to ensure that generated points remain aligned with the underlying surface. Finally, we incorporate a recurrent interface network~(RIN) to strengthen hierarchical feature interactions and boost reconstruction quality. Extensive experiments on synthetic benchmarks and real-world scans show that PUFM++ sets a new state of the art in point cloud upsampling, delivering superior visual fidelity and quantitative accuracy across a wide range of tasks. Code and pretrained models are publicly available at https://github.com/Holmes-Alan/Enhanced_PUFM.

Problem

• Reconstructs dense point clouds from sparse, noisy, partial observations
• Improves geometric fidelity, robustness, and downstream task consistency
• Enhances flow matching for high-quality point cloud upsampling

Innovation

• Two-stage flow-matching strategy for dense point cloud reconstruction
• Data-driven adaptive time scheduler to improve sampling efficiency
• On-manifold constraints and recurrent interface network for surface alignment