July 13th, 2021 
Daniela Lopez de Luise The deadline for submitting results to the Continual Learning challenges co-located with the First International Workshop on Continual Semi-Supervised Learning (CSSL @ IJCAI 2021), has been extended to
***July 29 2021***
We propose both a Continual Activity Recognition (CAR) challenge (https://lnkd.in/dM3rwNX) and a Continual Crowd Counting (CCC) challenge (https://lnkd.in/dNSPstt).
The challenges are based on the first benchmarks for “semi-supervised continual learning”, a setting in which a model can be trained on an initial supervised training set but then needs to be updated using the information provided by an unlabelled data stream.
Computer vision researchers are especially welcome as the challenges are in activity recognition and crowd counting from video frames.
Predictions need to be submitted through eval.ai:
https://eval.ai/web/challenges/challenge-page/986/overview
Important Dates
· Submission of results: July 29 2021
· Announcement of results: July 30 2021
· Challenge event @ workshop: August 21-23 2021
Organisers
Fabio Cuzzolin (Oxford Brookes University)
Kevin Cannons (Huawei)
Vincenzo Lomonaco, PhD (U. Pisa and ContinualAI)
Irina Rish (Mila – Quebec Artificial Intelligence Institute)
Salman Khan (Oxford Brookes)
Mohamad Asiful Hossain (Huawei)
Ajmal Shahbaz (Oxford Brookes)
Posted in