Employing machine learning techniques for activity recognition in training with human body landmark data

Rubens de Oliveira Santos, Eric de Paula Ferreira, Fernando Silva Parreiras, Fábio Corrêa, Vinícius Figueiredo de Faria, Renata de Souza França, Dárlinton Barbosa Feres Carvalho

Resumo


Machine learning methods for recognizing human activities have achieved remarkable results recently. Following this trend, this research aims to investigate automated activity recognition in the context of students undergoing training at UniverCemig. This quantitative-exploratory research applied the Support Vector Machine (SVM), K-Nearest Neighbors, Gated Recurrent Unit (GRU), and Long Short-Term Memory (LSTM) techniques. As a result, K-Nearest Neighbors and SVM achieved 91% and 96% performance, respectively, while GRU and LSTM achieved 96% performance for both. In conclusion, automated recognition of human activities is promising and suitable for training students at UniverCemig, helping instructors during training and, consequently, bringing agility to the student's learning certification process.


Texto completo:

PDF

Referências


Aguileta, A., Brena, R., Mayora, O., Molino Minero Re, E., and Trejo, L. 2019. “Multisensor fusion for

activity recognition - a survey,” Sensors (19:17), pp. 1-14.

Baptista, S. and Cunha, M. 2007. “Estudo de usuários: visão global dos métodos de coleta de dados,”

Perspectivas em Ciência da Informação (12), pp. 168-184.

Chung, J., Gulcehre, C., Cho, K., and Bengio, Y. 2014. “Empirical evaluation of gated recurrent neural networks on sequence modeling,” in Proceedings of the NIPS 2014 Deep Learning and Representation Learning Workshop, pp. 1-14.

Gil, A. C. 2008. “Métodos e técnicas de pesquisa social,” 6 ed., São Paulo: Editora Atlas.

Crowley, J. l., Christensen, H. I. 1995. “Vision as process: basic research on computer vision systems,” J. L. Crowley, H. I. Christensen (eds.); with editorial assistance by: Alain Chehikian e. a., pp. 341–371. Berlin: Springer, 1995.

Davila, J., Cretu, A.-M., and Zaremba, M. 2017. “Wearable sensor data classification for human activity recognition based on an iterative learning framework,” Sensors (17:6), pp. 1287.

Dehghani, A., Sarbishei, O., Glatard, T., and Shihab, E. 2019. “A quantitative comparison of overlapping and non-overlapping sliding windows for human activity recognition using inertial sensors,” Sensors (19:22), pp. 5026.

Filntisis, P., Efthymiou, N., Koutras, P., Potamianos, G., and Maragos, P. 2019. “Fusing body posture with facial expressions for joint recognition of affect in child–robot interaction,” IEEE Robotics and Automation Letters (4:4), pp. 4011–4018.

Haykin, S. 2001. “Redes Neurais: Princípios e prática,” São Paulo: Bookman Editora.

Hochreiter, S. and Schmidhuber, J. 1997. “Long short-term memory,” Neural computation (9), pp. 1735– 80.

Janidarmian, M., Roshan Fekr, A., Radecka, K., and Zilic, Z. 1995. “Support-vector networks,” Machine learning (20), pp. 273-297.

Lara, O. D. and Labrador, M. A. 2013. “A survey on human activity recognition using wearable sensors,”

IEEE Communications Surveys Tutorials (15:3), pp. 1192-1209.

Lei, T., Zhang, Y., Wang, S., Dai, H., and Artzi, Y. 2018. “Simple recurrent units for highly parallelizable recurrence,” Preprint, pp. 1-15.

Manzi, A., Dario, P., and Cavallo, F. 2017. “A human activity recognition system based on dynamic clustering of skeleton data,” Sensors (17:5), pp. 1-14.

McCulloch, W. S. and Pitts, W. 1943. “A logical calculus of the ideas immanent in nervous activity,” The bulletin of mathematical biophysics. Cambrigde, Massachussets, USA, 1 edition.

Mitchell, R., Michalski, J., and Carbonell, T. 2013. “An artificial intelligence approach,” Cambrigde, Massachussets, USA, 1 Edition.

Nguyen, P., Akiyama, T., Ohashi, H., Nakahara, G., Yamasaki, K., and Hikaru, S. 2015. “User-friendly activity recognition using svm classifier and informative features,” IEEE, pp. 1-8.

Ordonez, F. and Roggen, D. 2016. “Deep convolutional and lstm recurrent neural networks for multimodal wearable activity recognition,” Sensors (16:1), pp. 115.

Poppe, R. 2010. “A survey on vision-based human action recognition,” Image and Vision Computing 28:6),

pp. 976-990.

Rosati, S., Balestra, G., and Knaflitz, M. 2018. “Comparison of different sets of features for human activity recognition by wearable sensors,” Sensors (18:12), pp. 4189.

Sapiński, T., Kamińska, D., Pelikant, A., and Anbarjafari, G. 2019. “Emotion recognition from skeletal movements,” Entropy (21:7), pp. 646.

Sáez, Y., Gómez, A. B., and Viñuela, P. I. 2016. “A comparison study of classifier algorithms for cross-person physical activity recognition,” Sensors (17:1), pp. 66.

Shotton, J., Fitzgibbon, A., Cook, M., Sharp, T., Finocchio, M., Moore, R., Kipman, A., and Blake, A. 2011. “Real-time human pose recognition in parts from single depth images”, In Proceedings of the CVPR 2011, pp. 1297–1304.

Straka, M., Hauswiesner, S., Rüther, M., and Bischof, H. 2011. “Skeletal graph based human pose estimation in real-time,” In Proceedings of the BMVC 2011, p. 1-12.

Su, X., Tong, H., and Ji, P. (2014). “Activity recognition with smartphone sensors,” Tsinghua Science and Technology (19:2), pp. 35-249.

Sugiyama, M. and Kawanabe, M. 2012. “Machine Learning in Non-Stationary Environments: Introduction to Covariate Shift Adaptation,” volume 4 of 10. The name of the publisher, Cambridge,MA, USA, 1 edition.

Tasnim, N., Islam, M. M., and Baek, J.-H. 2020. “Deep learning-based action recognition using 3d skeleton joints information,” Inventions, (5:3), pp. 49.

Vahora, S. and Chauhan, N. 2019. “Deep neural network model for group activity recognition using contextual relationship, Engineering Science and Technology, an International Journal (22:1), pp. 47- 54.

Wang, Y., Velswamy, K., and Huang, B. (2017). “A long-short term memory recurrent neural network based reinforcement learning controller for office heating ventilation and air conditioning systems,” Processes (5:3), pp. 1-15.


Apontamentos

  • Não há apontamentos.