Human Action Recognition using CNN and LSTM-RNN with Attention Model
Kuppusamy. P1, Harika. C2

1Kuppusamy. P, Department of Computer Science, Madanapalle Institute of Technology & Science, Madanapalle, India.
2Harika. C, Department of Computer Science, Madanapalle Institute of Technology & Science, Madanapalle, India.

Manuscript received on 02 June 2019 | Revised Manuscript received on 10 June 2019 | Manuscript published on 30 June 2019 | PP: 1639-1643 | Volume-8 Issue-8, June 2019 | Retrieval Number: G6246058719/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (

Abstract: The recent advancements in artificial intelligence make the world into recognizing the objects, learning the environment, and predicting the forthcoming sequences. Emerging of embedded technology leads to decrease the cost of surveillance systems. The surveillance systems are capturing the environment and stored in memory. Machine learning is utilized for processing the data to aware of the scenario. This paper is considered the idea of using the video for recognizing the human action and behavior. Tis paper is proposed the integration of convolutional neural network and long short-term memory recurrent neural network for processing the video. The convolution processes the given input that produces the informative spatial features. The extracted features directed into long short-term module to generate temporal features. The feature maps of long short-term memory component fed into proposed attention element. It captures the highly valuable informative features in the frame of video. The actions are recognized from the informative features using softmax module. This model is used to recognize the human actions from video. The experimental results proved that proposed model performed better with accuracy.
Keyword: Attention model, Behavior, CNN, Human action, LSTM.
Scope of the Article: Probabilistic Models and Methods.