Infrared Human Activity Recognition Dataset - Coventry-2018

Coventry-2018 is a human activity recognition dataset captured by three Panasonic® Grid-EYE (AMG8833) infrared sensors in March 2018. The Grid-EYE sensors represent a 60 field of view scene by an 8 × 8 array named frame. The data streams are synchronized to 10 frames per second and saved as *.csv recordings using the LabVIEW® software. Two layouts are considered in this dataset with different geometry sizes: 1) small layout; and 2) large layout. The three sensors are located 1.5m from the area of interest in the small geometry size, while the distance grows to 2.5m for the large geometry size as illustrated in the dataset image. The dataset consists of 15 gestures distinguished in two categories: 1) single-subject gestures; and 2) double-subject gestures. There exist eight single-subject activities, and seven double-subject activities. Each single-subject gesture is repeated 10 times by three participants, while for double-subject gestures, the three possible combinations of the three subjects are used and similarly, the gestures are repeated 10 times. It needs to be emphasized that the single-subject activities are conducted in both layouts (small layout and large layout), while the double-subject activities are conducted only in the large layout.