Doctoral Thesis
Here you can find all the datasets collected and experimented during my doctoral thesis work [1].
Please cite related papers when using the datasets. Contact by e-mail if you have any further questions.
Gait-based Person Identification
Pressure-sensitive floor matrix is used to identify persons based on gait, i.e., the way their walk.
More details see [1] and [2]. Dataset includes two different settings with 10 and 11 subjects, respectively.
Please, cite the original paper [2] if using the data.
EMFI sensor dataset |
README
Binary switch floor sensor matrix and accelerometer is used to identify persons. These sensor modalities
can be combined or used separately. See [1] and [3] for more details. Datasets contain 9 subjects of normal walking as
well as variability in walking speed and footwear from 4 different subjects. Please, cite the original papers [1] and [3]
if using the data.
Binary switch sensor dataset |
README
Accelerometer sensor dataset |
README
Person Tracking
Binary switch floor sensor matrix is used to track position of multiple moving targets. Dataset contains 3 different
subjects in different two-person tracking scenarios. Details can be found in [1] and [4]. Please, cite the original papers [1] and [4]
if using the data. Example demonstration video:
demo (avi) .
Contact me by e-mail for original videos recorded from the data collection experiments.
Binary switch sensor dataset |
README
Activity Recognition
Multiple 3D acceleration sensors are used for recognition of daily activities. Dataset includes 17 activities
performed by 13 subjects. More details can be found in [1],[5], and [6]. Please, cite the original paper [6] if
using the data.
Activity dataset |
README
References
[1] Suutala J. (2012) Learning Discriminative Models from Structured Multi-sensor Data for Human Context Recognition, Doctoral Thesis,
University of Oulu, 221 p.
[2] Suutala J. & Röning J. (2008) Methods for person identification on a pressure-sensitive floor: Experiments with multiple classifiers and
reject option. Information Fusion Journal, Special Issue on Applications of Ensemble Methods. 9: 21-40.
[3] Suutala J., Fujinami K. & Röning J. (2008) Gaussian Process person identifier based on simple floor sensors.
Proc. 3rd European Conference on Smart Sensing and Context (EuroSSC08), Zürich, Switzerland, 55-68.
[4] Suutala J., Fujinami K. & Röning J. (2010) Persons tracking with Gaussian Process joint Particle Filtering. Proc. IEEE International Workshop
on Machine Learning for Signal Processing (MLSP 2010), Kittilä, Finland.
[5] Suutala J., Pirttikangas S. & Röning J. (2007) Discriminative temporal smoothing for activity recognition from wearable sensors.
Proc. 4th International Symposium on Ubiquitous Computing Systems (UCS07), Tokyo, Japan, pp. 182-195.
[6] Pirttikangas S., Fujinami K. & Nakajima T. (2006) Feature selection and activity recognition from wearable sensors.
Proc. H.Y. Youn, M. Kim, and H. Morikawa (Eds.):UCS 2006, LNCS 4239, Springer-Verlag Berlin Heidelberg, pp. 516-527.
