RED-Net: A Recurrent Encoder-Decoder Network for Video-based Face Alignment.

Xi Peng Rogério Schmidt Feris Xiaoyu Wang Dimitris N. Metaxas

Published in: CoRR (2018)

Keyphrases

face alignment
active appearance models
active shape model
video codec
distributed video coding
face detection
decoding process
computational complexity
motion estimation
rate distortion
facial images
viewpoint
bit rate
detection algorithm
coarse to fine
gender classification