Neural Sign Actors: A diffusion model for 3D sign language production from text

TL;DR

This paper introduces a diffusion-based model for generating realistic 3D sign language avatars from text, leveraging a novel graph neural network on the SMPL-X skeleton to improve sign language production.

Contribution

The paper presents a new diffusion model for 3D sign language production using a graph neural network on the SMPL-X skeleton, trained on a large-scale dataset, advancing realism and semantic accuracy.

Findings

Outperforms previous SLP methods quantitatively

Produces realistic dynamic 3D sign language sequences

Bridges communication gap for Deaf and hearing communities

Abstract

Sign Languages (SL) serve as the primary mode of communication for the Deaf and Hard of Hearing communities. Deep learning methods for SL recognition and translation have achieved promising results. However, Sign Language Production (SLP) poses a challenge as the generated motions must be realistic and have precise semantic meaning. Most SLP methods rely on 2D data, which hinders their realism. In this work, a diffusion-based SLP model is trained on a curated large-scale dataset of 4D signing avatars and their corresponding text transcripts. The proposed method can generate dynamic sequences of 3D avatars from an unconstrained domain of discourse using a diffusion process formed on a novel and anatomically informed graph neural network defined on the SMPL-X body skeleton. Through quantitative and qualitative experiments, we show that the proposed method considerably outperforms previous methods of SLP. This work makes an important step towards realistic neural sign avatars, bridging the communication gap between Deaf and hearing communities.