ACTOR learns an action-aware latent representation for human motions by training a generative variational autoencoder (VAE). By sampling from this latent space and querying a certain duration through a series of positional encodings, we synthesize variable-length motion sequences conditioned on a categorical action. ACTOR uses a transformer-based architecture to encode and decode a sequence of parametric SMPL human body models estimated from action recognition datasets.

ACTOR learns an action-aware latent representation for human motions by training a generative variational autoencoder (VAE). By sampling from this latent space and querying a certain duration through a series of positional encodings, we synthesize variable-length motion sequences conditioned on a categorical action. ACTOR uses a transformer-based architecture to encode and decode a sequence of parametric SMPL human body models estimated from action recognition datasets.