A new multi-pose facial expression dataset with 54 expressions is introduced, augmented by FaPE-GAN, and used to train Fa-Net for end-to-end recognition across unbalanced poses and zero-shot subjects.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
A Fine-Grained Facial Expression Database for End-to-End Multi-Pose Facial Expression Recognition
A new multi-pose facial expression dataset with 54 expressions is introduced, augmented by FaPE-GAN, and used to train Fa-Net for end-to-end recognition across unbalanced poses and zero-shot subjects.