Autoregressive GAN for Semantic Unconditional Head Motion Generation

Louis Airale (1, 2) , Xavier Alameda-Pineda (2) , Stéphane Lathuilière (3, 4, 5) , Dominique Vaufreydaz (1)

1 M-PSI - Multimodal Perception and Sociable Interaction
2 ROBOTLEARN - Vers des robots à l’intelligence sociale au travers de l’apprentissage, de la perception et de la commande
3 IP Paris - Institut Polytechnique de Paris
4 IDS - Département Images, Données, Signal
5 MM - Multimédia

In this work, we address the task of unconditional head motion generation to animate still human faces in a low-dimensional semantic space from a single reference pose. Different from traditional audio-conditioned talking head generation that seldom puts emphasis on realistic head motions, we devise a GAN-based architecture that learns to synthesize rich head motion sequences over long duration while maintaining low error accumulation levels. In particular, the autoregressive generation of incremental outputs ensures smooth trajectories, while a multi-scale discriminator on input pairs drives generation toward better handling of high- and low-frequency signals and less mode collapse. We experimentally demonstrate the relevance of the proposed method and show its superiority compared to models that attained state-of-the-art performances on similar tasks.

Keywords: GAN; Head motion; Face landmarks

Domains: Computer Vision and Pattern Recognition [cs.CV]; Machine Learning [stat.ML]

Louis Airale, Xavier Alameda-Pineda, Stéphane Lathuilière, Dominique Vaufreydaz. Autoregressive GAN for Semantic Unconditional Head Motion Generation. 2023. ⟨hal-03833759v2⟩