Probabilities of low and medium level activities and locations based on "OPPORTUNITY" Activity Recognition Dataset
No Thumbnail Available
Date
2018-02-03
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Mendeley Data
Abstract
The OPPORTUNITY Dataset for Human Activity Recognition from Wearable, Object, and Ambient Sensors (hereafter OPPORTUNITY dataset) is a dataset devised to benchmark human activity recognition algorithms (classification, automatic data segmentation, sensor fusion, feature extraction, etc).
Current dataset contains probabilities of subjects' Location and compass, doors states, gestures, gesture, low-level activities for both hands including hand movements, and interactions with specific objects. Thre of most probable candidates are reported as A,B and C. For locations they are calculated for all instances with no label. For other activities, we used activities 1, 2, 3 and 9 (drill) as training and probabilities are calculated for activities 4 and 5. For original signals received from sensors, download the OPPORTUNITY dataset.
Description
Keywords
Activity Recognition, Probabilistic Problem