Efficientnet Model Based Gesture Recognition Method Research
Pages 255 - 259
Abstract
Aiming at the problem of high computational complexity and accuracy of dynamic gesture recognition, this paper proposes a multi-feature fusion dynamic gesture recognition based on improved EfficientNet[1] model. Firstly, for each video data set, video frames are cut at a certain FPS and saved as training set and test set, and the classification performance of images is regarded as the classification performance of corresponding videos. Then, feature extraction and model fusion are used to extract dynamic gestures of characters, so as to extract key actions and training classification models. After the training, load the model to check and verify all the video frames in the test set, and get the output of top1 accuracy rate and Top5 accuracy rate in the whole test set. Finally, real-time detection is carried out. The experimental analysis of 170 video files shows that the average accuracy of the improved method is significantly higher than that of other common neural networks. The average detection accuracy in various gestures such as click and zoom was 93%.
References
[1]
M. Tan and Q. V. Le, “Rethinking Model Scaling for Convolutional neural Networks,” Internatinal Conference on Machine Learning, 2019.
[2]
Hall ET al, The silent language, vol.3.Doubleday, New York, 1959.
[3]
W. Cheng, Y. Sun, G.Jiang and H.Liu, “Jointly network:a network based on CNN and RBM for gesture recongnition,” Neural Computing and Applications, 2019, 31(Suppl 1): S309-S323.
[4]
S. E and P. N, “Hand gesture recongnition using a neural network shape fitting technique,” Eng Appl Artif Intell 22(8):1141-1158.
[5]
X. Yan and Wu, “A hand gesture recognition algorihm based on DC-CNN,” Multimedia Tools and Applications, 79:9193-9205, 2020.
[6]
R. S and A. A “Vision based hand gesture recognition for human computer interaction:a survey,” Artif Intell Rev43(1)1-5, 2015.
[7]
C. MK and T. RK and H. S, “Accelerometer based static gesture recognition and moblie monitoring system using neural networks,” Proc Comput Sci 70:683-687, 2015.
[8]
K. Miao, W., Li, G., Jiang, G., Fang, Y., Ju, Z., Liu and H., “Optimal grasp planning of multi-fingered robotic hands: a review,” Appl.Comput.Math.14(3), 238-247, 2015
[9]
Hasan, H., Abdul-Kareem and S., “Retracted article: human-computer interaction using vision-based hand gesture recongnition systems: a survey,” Neural Comput.Appl. 25(2), 251-261, 2014
[10]
M. Rautaray, S. S., Agrawal and A., “Vision based hand gesture reconition for human computer interaction: a survey,” Artif. Intell. Rev. 43(1), 1-54, 2015.
[11]
M. A, Z. A and T. PH, “Hand detection using multiple proposals,” In: BMVC, 2011, pp.1-11.
[12]
C. A, R. J, D. K and R. S, “A survey on hand gesture recognition in context of soft copputing,” In: International conference on computer science and information technology. Springer, 2011, pp.46-55.
[13]
E. D. Cubuk, B. Zoph, J. Shlens and Q. V. Le, “RandAugment:Practical automated data augmentation with a reduced search space,” Computer Vision and Pattern Recognition(cs.CV), 2019.
Index Terms
- Efficientnet Model Based Gesture Recognition Method Research
Index terms have been assigned to the content through auto-classification.
Recommendations
Multi-scenario gesture recognition using Kinect
CGAMES '12: Proceedings of the 2012 17th International Conference on Computer Games: AI, Animation, Mobile, Interactive Multimedia, Educational & Serious Games (CGAMES)Hand gesture recognition (HGR) is an important research topic because some situations require silent communication with sign languages. Computational HGR systems assist silent communication, and help people learn a sign language. In this article, a ...
A Dempster-Shafer Evidence Theory Based Multimodal Human Gesture Recognition Method
ICCDE '20: Proceedings of 2020 6th International Conference on Computing and Data EngineeringHand gestures are widely used in human-computer interaction, and dynamic gesture recognition is still a challenging task. In this paper, a Dempster-Shafer evidence theory based multimodal human gesture recognition method is proposed. Firstly, the audio-...
Kinect Based Dynamic Hand Gesture Recognition Algorithm Research
IHMSC '12: Proceedings of the 2012 4th International Conference on Intelligent Human-Machine Systems and Cybernetics - Volume 01In dynamic gesture track recognition, gesture recognition rate has always been a problem, Because the traditional hand gesture recognition in how to separate the hand type from the surrounding environment, and then determine the palm position has been ...
Comments
Information & Contributors
Information
Published In
September 2021
715 pages
ISBN:9781450384087
DOI:10.1145/3488933
Copyright © 2021 ACM.
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee. Request permissions from [email protected]
Publisher
Association for Computing Machinery
New York, NY, United States
Publication History
Published: 25 February 2022
Check for updates
Author Tags
Qualifiers
- Research-article
- Research
- Refereed limited
Conference
AIPR 2021
AIPR 2021: 2021 4th International Conference on Artificial Intelligence and Pattern Recognition
September 24 - 26, 2021
Xiamen, China
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 81Total Downloads
- Downloads (Last 12 months)14
- Downloads (Last 6 weeks)0
Reflects downloads up to 28 Feb 2025
Other Metrics
Citations
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign inFull Access
View options
View or Download as a PDF file.
PDFeReader
View online with eReader.
eReaderHTML Format
View this article in HTML Format.
HTML Format