Abstract
Human actions recognition (HAR) and understanding become very popular topics in the field of computer vision and signal processing. The purpose of human activities recognition is to automatically examine and characterize actions from a video sequence. The main goal of a HAR system is to identify simple actions of everyday life (like walking, running, jumping ...) from videos. Each of these actions, performed by one person or many persons within a specific period of time, must be represented by a simple movement model. In recent years, a large number of applications of HAR have been proposed in literature such as video surveillance, human-computer interaction and video indexing. In this line, we present our method for HAR using Mask Region Based CNN, MRCNN. This technique will help us to make the accent on the body of individual and this step facilitates the recognition of the current action. With the mask RCNN we used key frame extraction and background extraction. This framework was tested on two datasets KTH and WEIZMANN datasets and experimental results showed the performance of the proposed technique.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Willems, G., Tuytelaars, T., Van, L.: An efficient dense and scale-invariant spatio-temporal interest point detector. In: Computer Vision ECCV 2008 10th European Conference on Computer Vision Marseille France (2008)
Wang, H., Wang, L., Member, S.: Beyond joints: learning representations from primitive geometries for skeleton-based action recognition and detection. J. Latex Class Files (2017)
ter Haar, F., Veltkamp, R.: A 3d face matching framework, vol. 2008 (2008)
Binh, H.T., Chau, M.T., Sugimoto, A., Duyp, B.T.: Selecting active frames for action recognition with vote fusion method. In: International Conference on Computer and Communication engineering ICCCE (2018)
Chou, K.P., Prasad, M., Wu, D., Sharma, N., Li, D.L., Lin, Y.F., Michael Blumenstein, W.C.: Feature-based automated multi-view human action recognition system. IEEE Access 6, 15283–15296 (2018)
Obaidi, S.A., Abhayaratne, C.: Action recognition based on multi layer visual features. In: 10th International Conference on Intelligent Human Machine Systems and Cybernetics (2018)
Feng, Q., Ni, G., Dai, B., Dai, J.: Action recognition based on multi-layer visual features. In: 10th International Conference on Intelligent Human Machine Systems and Cybernetics (2018)
Tu, N.A., HuynhThe, T., Lee, K.U.K.Y.K.: Ml-HDP: a hierarchical bayesian nonparametric model for recognizing human actions in video. IEEE Transactions Circuits Systems Video Technology 27(7), 1478–1490 (2018)
Naidoo, D., Tapamo, J.R., Walingo, T.: Human action recognition using spatial-temporal analysis and bag of visual words In: 14th International Conference on Signal-Image Technology Internet-Based Systems (SITIS) (2018)
Girshick, R., Donahue, J., Darrell, T., Malik, J.: Rich feature hierarchies for accurate object detection and semantic segmentation. In: CVPR 2014 (2013)
Girshick, R.: Fast r-cnn. In: ICCV (2015)
He, K., Gkioxari, G., Dollár, P., Girshick, R.: Mask r-cnn. In: Computer Vision and Pattern Recognition (2017)
Lu, D., Hou, Z.: Mimo model-free adaptive control color background image extraction to video. In: IEEE 8th Data Driven Control and Learning Systems Conference (DDCLS) (2019)
Momin, B., Rupnar, G.: Keyframe extraction in surveillance video using correlation (2016)
Zhang, R., Ni, B.: Learning behavior recognition and analysis by using 3d convolutional neural network. In: nternational Conference on Engineering, Applied Sciences and Technology (ICEAST) (2019)
Lian, M.: Captured multi-label relations via joint deep supervised autoencoder. Appl. Soft Comput. J. 74, 709–728 (2018)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Zaghbani, S., Bouhlel, M.S. (2021). Mask RCNN for Human Motion and Actions Recognition. In: Abraham, A., et al. Proceedings of the 12th International Conference on Soft Computing and Pattern Recognition (SoCPaR 2020). SoCPaR 2020. Advances in Intelligent Systems and Computing, vol 1383. Springer, Cham. https://doi.org/10.1007/978-3-030-73689-7_1
Download citation
DOI: https://doi.org/10.1007/978-3-030-73689-7_1
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-73688-0
Online ISBN: 978-3-030-73689-7
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)