FUSE: A Flow-based Mapping Between Shapes

TL;DR

FUSE introduces a flow-based neural representation for shape matching that is efficient, invertible, and modality-agnostic, enabling accurate cross-representation shape correspondence without extensive training.

Contribution

The paper presents a novel flow-matching model for shape mapping that supports diverse data formats and achieves high accuracy without large-scale training.

Findings

High coverage and accuracy across benchmarks

Effective in cross-representation shape matching

Promising results in UV mapping and registration

Abstract

We introduce a novel neural representation for maps between 3D shapes based on flow-matching models, which is computationally efficient and supports cross-representation shape matching without large-scale training or data-driven procedures. 3D shapes are represented as the probability distribution induced by a continuous and invertible flow mapping from a fixed anchor distribution. Given a source and a target shape, the composition of the inverse flow (source to anchor) with the forward flow (anchor to target), we continuously map points between the two surfaces. By encoding the shapes with a pointwise task-tailored embedding, this construction provides an invertible and modality-agnostic representation of maps between shapes across point clouds, meshes, signed distance fields (SDFs), and volumetric data. The resulting representation consistently achieves high coverage and accuracy across diverse benchmarks and challenging settings in shape matching. Beyond shape matching, our framework shows promising results in other tasks, including UV mapping and registration of raw point cloud scans of human bodies.