TST Intake Monitoring dataset v1

– The dataset is composed by the executions of food and drink intake actions performed by 35 young actors.

Each actor repeated each action 1/2 times, generating a total number of 48 sequences. For each sequence, the following data are available:

• depth frames, resolution of 320x240, captured by Kinect v1 in top-view configuration;
• coordinates of nodes computed by three unsupervised algorithms (SOM, SOM_Ex, GNG) and coordinates of ground-truth nodes (only for hands and head, manually identified): http://www.tlc.dii.univpm.it/dbkinect/FoodIntake/LoadNets_And_GT.zip

This code can be used to read data with MATLAB, and to convert the depth frame from pixel domain to the point cloud domain:

http://www.tlc.dii.univpm.it/dbkinect/FoodIntake/FromDepth2PC.zip

If you use the dataset, please cite the following paper:

S. Gasparrini, E. Cippitelli, E. Gambi, S. Spinsante and F. Florez-Revuelta, “Performance Analysis of Self-Organising Neural Networks Tracking Algorithms for Intake Monitoring Using Kinect,” 1st IET International Conference on Technologies for Active and Assisted Living (TechAAL), 6th November 2015, Kingston upon Thames (UK).