8 citations · 16 across the 6 of their papers we have counts for
6 papers
2D+3D facial expression recognition via embedded tensor manifold regularization
Yunfang Fu, Qiuqi Ruan, Ziyan Luo +3
In this paper, a novel approach via embedded tensor manifold regularization for 2D+3D facial expression recognition (FERETMR) is proposed. Firstly, 3D tensors are constructed from…
Prototype Guided Network for Anomaly Segmentation
Yiqing Hao, Yi Jin, Gaoyun An
Semantic segmentation methods can not directly identify abnormal objects in images. Anomaly Segmentation algorithm from this realistic setting can distinguish between in-distributi…
STSM: Spatio-Temporal Shift Module for Efficient Action Recognition
Zhaoqilin Yang, Gaoyun An
The modeling, computational cost, and accuracy of traditional Spatio-temporal networks are the three most concentrated research topics in video action recognition. The traditional…
E2-Capsule Neural Networks for Facial Expression Recognition Using AU-Aware Attention
Shan Cao, Yuqian Yao, Gaoyun An
Capsule neural network is a new and popular technique in deep learning. However, the traditional capsule neural network does not extract features sufficiently before the dynamic ro…
Squeeze-and-Excitation on Spatial and Temporal Deep Feature Space for Action Recognition
Gaoyun An, Wen Zhou, Yuxuan Wu +2
Spatial and temporal features are two key and complementary information for human action recognition. In order to make full use of the intra-frame spatial characteristics and inter…
Multi-Level Recurrent Residual Networks for Action Recognition
Zhenxing Zheng, Gaoyun An, Qiuqi Ruan
Most existing Convolutional Neural Networks(CNNs) used for action recognition are either difficult to optimize or underuse crucial temporal information. Inspired by the fact that t…