Search results with tag "Based generative adversarial"

VIBE: Video Inference for Human Body Pose and Shape …

paired information by training a sequence-based generative adversarial network (GAN) [18]. Here, given the video of a person, we train a temporal model to predict the parame-ters of the SMPL body model for each frame while a mo-tion discriminator tries to distinguish between real and re-gressedsequences. Bydoingso,theregressorisencouraged

  Based, Adversarial, Generative, Discriminator, Based generative adversarial