CogniLearn HSKT Dataset


Overview
   
We present the CogniLearn HSKT Dataset (Cogni-HSKT) with image data collected from the CogniLearn Interface from participants while they perform the well-known Head-Toes-Knees-Shoulders (HTKS) tasks. This dataset contains RGB data frames collected using Kinect V2 camera from 15 participants (9 Male and 6 Females) of age group 18 to 30. All participants were required to perform four steps of the protocol setup in the CogniLearn recording interface. Details of the four steps are provided in Table-1. Head-Toes-Knees-Shoulders (HTKS) task is known to provides sufficient psychometric properties for assessing cognitive dysfunctions and serves to be a useful measure for behavioral self-regulation.This dataset comprises of over 60,000 frames of RGB data captured for the participants which creates a substantial dataset for cognitive analysis as well a baseline to perform gesture recognition for the physical exercises performed during the HTKS task. We provide annotated gesture class labels for 4,443 of these frames which are the Step-1 recordings for all the participants. TheHTKS task has four sections with up to four paired behavioral rules: touch your head and touch your toes; touch your shoulders and touch your knees. Subjects first respond naturally, and then are instructed to switch rules by responding in the opposite way (e.g., touch their knees when told to touch their shoulders). 


Dataset Description

   
There are 4 steps recorded per participant with audio instructions provided to perform tasks. These instructions are shown in Table-1.  In step-1 there is no task-switching. In step-2 "Knees" and "Shoulders" are swapped. In step-3 "Head" and "Toes" are swapped and in step-4 "Knees" and "Shoulders" are swapped and "Head" and "Toes" are swapped. We also provide timestamps at which each of the RGB frames was recorded and timestamps at which instructions were provided.


Steps
Table-1 HTKS Steps captured



Downloads

   
The images can be downloaded by clicking the link: Cogni-HSKT
    Image frame timestamps can be downloaded from the following link: RGB Timestamps
    Instruction timestamps can be downloaded from the following link: Instruction Timestamps
    Ground Truth gesture class annotations can be downloaded from the following link: Annotations

 


Downloads Description
 


Citation:
If you use our dataset for your research work, please Cite our Paper:
 
CogniLearn: A Deep Learning based Interface for Cognitive Behavior Assessment



Thank You for Visiting our Webpage!