Point-MF: One-step Point Cloud Generation from a Single Image via Mean Flows

TL;DR

Point-MF introduces a fast, one-step point cloud reconstruction method from a single image using mean flows and a specialized transformer, achieving high quality with minimal inference time.

Contribution

The paper presents a novel one-step point cloud reconstruction framework that operates directly in point-cloud space using mean flows, eliminating the need for multiple denoising iterations.

Findings

Achieves high-quality point cloud reconstruction with millisecond latency.

Outperforms multi-step diffusion models in speed while maintaining quality.

Stabilizes large-step generation with a new auxiliary loss.

Abstract

Single-image point cloud reconstruction must infer complete 3D geometry, including occluded parts, from a single RGB image. While diffusion-based reconstructors achieve high accuracy, they typically require many denoising iterations, resulting in slow and expensive inference. We propose Point-MF, a Mean-Flow-based framework for low-NFE single-image point cloud reconstruction that couples a Mean-Flow-compatible architecture with an auxiliary loss. Specifically, Point-MF operates directly in point-cloud space to learn the mean velocity field and enables one-step reconstruction with a single network function evaluation (1-NFE), without relying on VAE-based latent representations. To make Mean Flow effective under large interval jumps, Point-MF employs a Diffusion Transformer tailored to the Mean-Flow setting, conditioned on frozen DINOv3 image features via a lightweight token adapter and equipped with explicit interval/time conditioning. Moreover, we introduce Denoised Space Anchor, a set-distance auxiliary loss on the denoised-space estimate $x_\theta$ induced by the predicted velocity field, to stabilize large-step generation and reduce outliers and density artifacts. On ShapeNet-R2N2 and Pix3D, Point-MF strikes a strong balance between reconstruction quality and inference speed compared to multi-step diffusion baselines and competitive feedforward models, while generating high-quality point clouds with millisecond-level latency.