Login / Signup

MixSynthFormer: A Transformer Encoder-like Structure with Mixed Synthetic Self-attention for Efficient Human Pose Estimation.

Yuran SunAlan William DoughertyZhuoying ZhangYi-King ChoiChuan Wu
Published in: ICCV (2023)
Keyphrases
  • human pose estimation
  • support vector
  • pose estimation
  • tree models
  • facial expressions
  • saliency map
  • body parts
  • gaussian processes