Introduction

The CVPR 2021 Workshop on Continual Learning (CLVision, 2nd edition) aims to gather researchers and engineers from academia and industry to discuss the latest advances in continual learning. In this one-day workshop, we will have regular paper presentations, invited speakers, and technical benchmark challenges to present the current state of the art, as well as the limitations and future directions for computer vision in continual learning, arguably one of the most crucial milestones of computer vision and AI in general.

You can checkout the accepted works here!

For the previous edition see: https://sites.google.com/view/clvision2020

Livestream

You can access the SLIDES from the Livestream here:

• PART 1: Talk #1 - Razvan Pascanu, Talk #2 - Hatice Gunes, Talk #3 - Joost van de Weijer, Talk #4 - Jim Rehg, Talk #5 - Rahaf Aljundi

• PART 2: Talk #6 - Sayna Ebrahimi, Talk #7 - Andreas Tolias, Talk #8 - Eric Eaton, Talk #9 - Jeff Clune

How to Participate?

The event will be live streamed on YouTube through the ContinualAI Youtube account and you'll be able to interact with the speakers and the other participants via the Youtube chat, a Discuss chat (below) or the ContinualAI slack workspace (Join here).

In order to join the LIVE Poster Sessions you need to register to the event on Eventbrite HERE. Otherwise you can watch the pre-recorded videos in the Accepted Papers page.

In your virtual ticket you will find the links to the different MS Teams rooms (on for paper) and a few additional ones to socialize and interact with the other participants. We ask each author to be present in virtual room with the name corresponding to their paper title during at least one of the two poster sessions.

For more information about the planned schedule visit the program page.

News and Updates

• Check out our list of accepted papers and findings

• The supervised and reinforcement learning tracks of our competition are available, check it out!

• The workshop will take place on June 25th (UTC-5 timezone)

• Workshop officially confirmed for CVPR 2021

Prizes

Thanks to our amazing sponsors, those brave enough to submit a paper or participate in our Continual Learning challenge can be awarded exquisite prizes!

Besides the best paper award, all of the challenge tracks have thrilling prizes provided by our sponsors, and on top of that, the overall challenge winner will be generously awarded as well! The prizes will be announced in the closing remarks of the workshop. So... are you ready for the wall of fame?

Summary of Prizes*:

• Best Paper Award: A Tale of Two CILs: The Connections between Class Incremental Learning and Class Imbalanced Learning, and Beyond

  • 500 USD + 500 USD worth of Huawei cloud credits (HUAWEI)

• Best Library award: Avalanche

  • 1,500 USD (Weights & Biases)

• Supervised-Learning track

  • #1 700 USD + 300 USD worth of Huawei cloud credits (HUAWEI): SAIT-AVA-CL (Seungjun Shin, Suji Kim, Dokwan Oh)

  • #2 500 USD + 100 USD worth of Huawei cloud credits (HUAWEI): HIK_ILG

  • #3 300 USD + 100 USD worth of Huawei cloud credits (HUAWEI): Jodelet

• Reinforcement-Learning track winner:

  • #1 500 USD (ServiceNow): CLers

Participation (Closed)

Paper Submission

We solicit paper submissions on novel methods and application scenarios of continual learning. We accept papers on a variety of topics, including lifelong learning, few-shot learning, meta learning, incremental learning, online learning, multitask learning, etc.. Papers will be peer reviewed under double-blind policy and the submission deadline is 20th of March. Accepted papers will be presented at the poster session, some as orals and one paper will be awarded as the best paper.

Challenge Track

There is currently a challenge track programmed. Check it out!

Invited Speakers

Contact

contvisionworkshop@gmail.com

Sponsors