PUFM++: Point Cloud Upsampling via Enhanced Flow Matching

TL;DR

PUFM++ is an advanced point cloud upsampling method that enhances flow matching with a two-stage strategy, adaptive scheduling, on-manifold constraints, and a recurrent network, achieving state-of-the-art results.

Contribution

The paper introduces PUFM++, a novel framework that significantly improves point cloud upsampling through a multi-faceted flow matching approach and hierarchical feature integration.

Findings

Sets new state-of-the-art in point cloud upsampling

Achieves superior visual fidelity and quantitative accuracy

Demonstrates robustness to noisy and partial inputs

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.