Activity and Context Recognition with Opportunistic Sensor Configurations

Challenge submission

Submissions

Participants should submit their results by email:

activityrecognition.challenge@gmail.com

Submissions should include:

  • Participants name
  • Description of the proposed method including full list of parameters used (only for final submission)
  • Obtained labels in the test dataset (one separate file per subject and task). Data should be formatted as a column vector of the predicted labels.

Files can be in .mat or text format and named as follows:

[Task]_[subject]_[yourName]

Allowed task strings are: A,B1,B2,C. Allowed subject strings are: 2,3,4. e.g., If  the group CNBI submits its labels for the task A, then files should be named as 'A_2_CNBI'  and  'A_3_CNBI' for subjects 2 and 3, respectively.

Moreover, authors are encouraged to also provide accompanying code implementing the proposed method. 


Early feedback

Before the final deadline date participants can submit their results (as described above), along wih a brief description of their methods. We will provide them feedback about the corresponding performance (F-measure).

Performance assessment

Performance will be assessed using the weighted F-measure of all activity classess (i.e. excluding the F-measure of the NULL class).

Prizes

Two prizes will be awarded to the contributions that achieve the highest performance for each task.

  • USD 1000.- for the first place
  • USD 500.- for the 1st runner-up

An additional prize of the jury (USD 500.-) will be awarded by the recomendation of the organizers. It is intended to reward the most novel, innovative contribution (even if it doesn't achieve the highest performance).

Note: People directly involved with the Opportunity project are not eligible for the prizes. They can nevertheless submit their results for comparison purposes.

 

Contact

Do not hesitate to contact the consortium.

About

We develop opportunistic activity recognition systems: goal-oriented sensor assemblies spontaneously arise and self-organize to achieve a common activity and context recognition. We develop algorithms and architectures underlying context recognition in opportunistic systems.

Newsletter

Subscribe to our newsletter for regular project updates.

web design by ybresson.com
© 2017 OPPORTUNITY Consortium
The information on this website does not represent the opinion of the European Community, and is provided "as is". No guarantee or warranty is given that the information is fit for any particular purpose. The user uses the information at its sole risk and liability.