Northampton Electronic Collection of Theses and Research

Two-Stream Retentive Long Short-Term Memory Network for Dense Action Anticipation

Zhao, F., Zhao, J., Li, X., Zhang, Y., Guo, D. and Chen, W. (2022) Two-Stream Retentive Long Short-Term Memory Network for Dense Action Anticipation. Computational Intelligence and Neuroscience. 2022, pp. 1-9. 1687-5265.

[A][B][+][-]
16729:35889
[thumbnail of Zhao_etal_CIN_2022_Two-Stream_Retentive_Long_Short-Term_Memory_Network_for_Dense_Action_Anticipation]
Zhao_etal_CIN_2022_Two-Stream_Retentive_Long_Short-Term_Memory_Network_for_Dense_Action_Anticipation
Zhao_etal_CIN_2022_Two_Stream_Retentive_Long_Short_Term_Memory_Network_for_Dense_Action_Anticipation.pdf - Published Version
Available under License Creative Commons Attribution.
Information
Abstract:
Analyzing and understanding human actions in long-range videos has promising applications, such as video surveillance, automatic driving, and efficient human-computer interaction. Most researches focus on short-range videos that predict a single action in an ongoing video or forecast an action several seconds earlier before it occurs. In this work, a novel method is proposed to forecast a series of actions and their durations after observing a partial video. This method extracts features from both frame sequences and label sequences. A retentive memory module is introduced to richly extract features at salient time steps and pivotal channels. Extensive experiments are conducted on the Breakfast data set and 50 Salads data set. Compared to the state-of-the-art methods, the method achieves comparable performance in most cases.
Additional Information:
Copyright © 2022 Fengda Zhao et al.
Uncontrolled Keywords:
General Mathematics, General Medicine, General Neuroscience, General Computer Science, Neural Networks, Computer, Memory, Long-Term, Humans, Human Activities, Rivers, Memory, Short-Term
Creators:
Zhao, F., Zhao, J., Li, X., Zhang, Y., Guo, D. and Chen, W.
Faculties, Divisions and Institutes:
Faculties > Faculty of Arts, Science & Technology > Computing
Date:
16 May 2022
Date Type:
Publication
Page Range:
pp. 1-9
Journal or Publication Title:
Computational Intelligence and Neuroscience
Volume:
2022
Number of Pages:
9
Language:
English
DOI:
10.1155/2022/4260247
ISSN:
1687-5265
Status:
Published / Disseminated
Refereed:
Yes
Related URLs:
Edit Item