Exportar Publicação

Jardim, D., Nunes, L. & Dias, M. (2016). Predicting human activities in sequences of actions in RGB-D videos. In Verikas, A., Radeva, P., Nikolaev, D. P., Zhang, W. and Zhou, J. (Ed.), Proceedings of SPIE, Ninth International Conference on Machine Vision (ICMV 2016). Nice, France: SPIE.

D. W. Jardim et al.,  "Predicting human activities in sequences of actions in RGB-D videos", in Proc. of SPIE, Ninth Int. Conf. on Machine Vision (ICMV 2016), Verikas, A., Radeva, P., Nikolaev, D. P., Zhang, W. and Zhou, J., Ed., Nice, France, SPIE, 2016, vol. 10341

@inproceedings{jardim2016_1730749880045,
	author = "Jardim, D. and Nunes, L. and Dias, M.",
	title = "Predicting human activities in sequences of actions in RGB-D videos",
	booktitle = "Proceedings of SPIE, Ninth International Conference on Machine Vision (ICMV 2016)",
	year = "2016",
	editor = "Verikas, A., Radeva, P., Nikolaev, D. P., Zhang, W. and Zhou, J.",
	volume = "10341",
	number = "",
	series = "",
	doi = "10.1117/12.2268524",
	publisher = "SPIE",
	address = "Nice, France",
	organization = "SPIE",
	url = "https://www.spiedigitallibrary.org/conference-proceedings-of-spie/10341.toc"
}

TY  - CPAPER
TI  - Predicting human activities in sequences of actions in RGB-D videos
T2  - Proceedings of SPIE, Ninth International Conference on Machine Vision (ICMV 2016)
VL  - 10341
AU  - Jardim, D.
AU  - Nunes, L.
AU  - Dias, M.
PY  - 2016
SN  - 0277-786X
DO  - 10.1117/12.2268524
CY  - Nice, France
UR  - https://www.spiedigitallibrary.org/conference-proceedings-of-spie/10341.toc
AB  - In our daily activities we perform prediction or anticipation when interacting with other humans or with objects. Prediction of human activity made by computers has several potential applications: surveillance systems, human computer interfaces, sports video analysis, human-robot-collaboration, games and health-care. We propose a system capable of recognizing and predicting human actions using supervised classifiers trained with automatically labeled data evaluated in our human activity RGB-D dataset (recorded with a Kinect sensor) and using only the position of the main skeleton joints to extract features. Using conditional random fields (CRFs) to model the sequential nature of actions in a sequence has been used before, but where other approaches try to predict an outcome or anticipate ahead in time (seconds), we try to predict what will be the next action of a subject. Our results show an activity prediction accuracy of 89.9% using an automatically labeled dataset.
ER  -