Learning Domain Agnostic Latent Embeddings of 3D Faces for Zero-shot Animal Expression Transfer

TL;DR

This paper introduces a zero-shot method for transferring human facial expressions to 3D animal faces by learning domain-agnostic embeddings that generalize across species without needing animal training data.

Contribution

It proposes a novel latent embedding space that disentangles identity and expression, enabling cross-species expression transfer without animal-specific training.

Findings

Achieves plausible cross-species expression transfer

Effectively narrows geometric gap between human and animal faces

Uses Jacobian, vertex-position, and Laplacian losses for geometric consistency

Abstract

We present a zero-shot framework for transferring human facial expressions to 3D animal face meshes. Our method combines intrinsic geometric descriptors (HKS/WKS) with a mesh-agnostic latent embedding that disentangles facial identity and expression. The ID latent space captures species-independent facial structure, while the expression latent space encodes deformation patterns that generalize across humans and animals. Trained only with human expression pairs, the model learns the embeddings, decoupling, and recoupling of cross-identity expressions, enabling expression transfer without requiring animal expression data. To enforce geometric consistency, we employ Jacobian loss together with vertex-position and Laplacian losses. Experiments show that our approach achieves plausible cross-species expression transfer, effectively narrowing the geometric gap between human and animal facial shapes.