Training Dataset

The training dataset for this competition is comprised of three files:

• Complete sensor dataset (CSV file) containing all the raw sensor information (IMUs, infrared sensors, brain signals) from the subjects, and the activities (labels) performed.
• Images from camera 1 (folder with ZIP files) containing the frames retrieved from camera-1 recordings, ordered in folders for each subject, activity and trial.
• Images from camera 2 (folder with ZIP files) containing the frames retrieved from camera-2 recordings, ordered in folders for each subject, activity and trial.

Download files [NEW 6 May, 2019 - these files are no longer public here]:

• Complete sensor dataset.
• Images from Camera 1.
• Images from Camera 2.

Testing Dataset

The testing dataset for this competition is comprised of three files. The goal of this competition is the multi-classification of the tags (no activities) reported in the training set:

• Complete sensor dataset (CSV file) containing all the raw sensor information (IMUs, infrared sensors, brain signals) from the subjects. This files are not labeled.
• Images from camera 1 (folder with ZIP files) containing the frames retrieved from camera-1 recordings, ordered in folders for each subject.
• Images from camera 2 (folder with ZIP files) containing the frames retrieved from camera-2 recordings, ordered in folders for each subject.

Data in camera 1 and camera 2 are organized in three folders (one per subject). Inside these folders, you will find images organized in trials and activities. You can easily associate these images with the sensor data using the timestamp values.

Note: You will find the data in the testing set organized in activities, but the activities were renamed as A, B, C, D, E, etc. The order of these labels DO NOT correspond to the order of labels in the training set.

Download files [NEW 6 May, 2019 - these files are no longer public here]:

• Complete sensor dataset.
• Images from Camera 1.
• Images from Camera 2.