Micro Expression Recognition based on Graph Convolutional Networks with LSTM
Lists
Only you see this
Abstract
To improve the recognition rate of micro-expression algorithms, we propose an algorithm based on Graph Convolutional Networks with Long Short-Term Memory(LSTM). The algorithm makes full use of the spatial-temporal features of micro-expressions, first extracting the spatial features of micro-expressions through graph convolutional networks, then using LSTM to extract its temporal parts. Extensive experiments on MMEW and SAMM datasets demonstrate that our proposed method improves the accuracy of micro-expression recognition.
BibTeX
@misc{cai2023micro,
author = {Xiong Cai and Hui Tang and Li Chai},
title = {{Micro Expression Recognition based on Graph Convolutional Networks with LSTM}},
year = {2023},
doi = {10.1109/ccdc58219.2023.10327083},
}Loading related content...