iCareNet Eating behavior dataset

To develop a context-aware system to help people with eating related disorders like obesity, a method must be available to recognize eating behavior first. To make progress in the field of automatic eating behavior recognition, data must be available to train state-of-the-art machine learning algorithms and test their performance in terms of accuracy.

The iEatSet provides a dataset with 12 participants, recorded 5 times having different meals for 5 days. Ground truth labelling is provided to use for both testing and validation of future algorithm development for eating behavior as well as other activity recognition algorithms.

It contains:

  1. Compressed and synchronized RGB videos of the recordings from the IP cameras and the Kinect.
  2. The calibration data of each camera including the intrinsic and extrinsic parameters.
  3. 13-bit depth data from the Kinect.
  4. The raw, calibrated and synchronized 48-bit data from all the 4 IMUs.
  5. The labelled annotations.
  6. The timestamps generated by all the sensors.
  7. Accompanying software to read the data.

All the data is time synchronized.

Read more about the iEatSet dataset in these articles:

Request the iEatSet dataset

Conditions for download and usage:

  • Datasets are for academic and non-commercial use only.
  • When referring to this set in publications, Noldus Information Technology bv should be mentioned as owner of these data.
  • When applicable, refer to publications by Noldus about this dataset in publications that refer to the use of this set.
  • The datasets are available free of charge, but costs for shipping, handling and hard disks will be charged for.
  • Noldus cannot offer support on usage of these data.