Neural Semantic Surface Maps

TL;DR

This paper introduces an automated method for creating semantic maps between 3D shapes by leveraging pre-trained vision models and multi-view rendering, eliminating the need for manual annotations or 3D training data.

Contribution

It presents a novel approach that distills semantic correspondences from pre-trained visual models to produce accurate, bijective surface maps without requiring annotated 3D data.

Findings

Effective in complex semantic scenarios

Works with non-isometric and nearly isometric shapes

Eliminates manual annotation and 3D training data

Abstract

We present an automated technique for computing a map between two genus-zero shapes, which matches semantically corresponding regions to one another. Lack of annotated data prohibits direct inference of 3D semantic priors; instead, current State-of-the-art methods predominantly optimize geometric properties or require varying amounts of manual annotation. To overcome the lack of annotated training data, we distill semantic matches from pre-trained vision models: our method renders the pair of 3D shapes from multiple viewpoints; the resulting renders are then fed into an off-the-shelf image-matching method which leverages a pretrained visual model to produce feature points. This yields semantic correspondences, which can be projected back to the 3D shapes, producing a raw matching that is inaccurate and inconsistent between different viewpoints. These correspondences are refined and distilled into an inter-surface map by a dedicated optimization scheme, which promotes bijectivity and continuity of the output map. We illustrate that our approach can generate semantic surface-to-surface maps, eliminating manual annotations or any 3D training data requirement. Furthermore, it proves effective in scenarios with high semantic complexity, where objects are non-isometrically related, as well as in situations where they are nearly isometric.