Back to Search
Start Over
I3D-Shufflenet Based Human Action Recognition
- Source :
- Algorithms, Volume 13, Issue 11, Algorithms, Vol 13, Iss 301, p 301 (2020)
- Publication Year :
- 2020
- Publisher :
- MDPI AG, 2020.
-
Abstract
- In view of difficulty in application of optical flow based human action recognition due to large amount of calculation, a human action recognition algorithm I3D-shufflenet model is proposed combining the advantages of I3D neural network and lightweight model shufflenet. The 5 &times<br />5 convolution kernel of I3D is replaced by a double 3 &times<br />3 convolution kernels, which reduces the amount of calculations. The shuffle layer is adopted to achieve feature exchange. The recognition and classification of human action is performed based on trained I3D-shufflenet model. The experimental results show that the shuffle layer improves the composition of features in each channel which can promote the utilization of useful information. The Histogram of Oriented Gradients (HOG) spatial-temporal features of the object are extracted for training, which can significantly improve the ability of human action expression and reduce the calculation of feature extraction. The I3D-shufflenet is testified on the UCF101 dataset, and compared with other models. The final result shows that the I3D-shufflenet has higher accuracy than the original I3D with an accuracy of 96.4%.
- Subjects :
- lcsh:T55.4-60.8
Channel (digital image)
Computer science
Feature extraction
Optical flow
02 engineering and technology
lcsh:QA75.5-76.95
Theoretical Computer Science
Convolution
shufflenet
0202 electrical engineering, electronic engineering, information engineering
Feature (machine learning)
lcsh:Industrial engineering. Management engineering
Numerical Analysis
action recognition
Artificial neural network
business.industry
020207 software engineering
Pattern recognition
I3D neural network
Expression (mathematics)
Computational Mathematics
Histogram of oriented gradients
Computational Theory and Mathematics
3D convolution
020201 artificial intelligence & image processing
lcsh:Electronic computers. Computer science
Artificial intelligence
business
Subjects
Details
- ISSN :
- 19994893
- Volume :
- 13
- Database :
- OpenAIRE
- Journal :
- Algorithms
- Accession number :
- edsair.doi.dedup.....d3645a9dc889e1890d0af3ad2a108f6e