YouTube Contest On Kaggle
Written by Sue Gee   
Thursday, 07 June 2018

Google AI has announced a new Kaggle video understanding challenge in conjunction with the 2nd Workshop on YouTube-8M Large-Scale Video Understanding, at the 2018 European Conference on Computer Vision in Munich, Germany in September.

This competition is the second Kaggle competition based on YouTube 8M dataset, and is focused on learning video representation under budget constraints.




YouTube-8M is a large-scale labeled video dataset that consists of millions of YouTube video IDs, with high-quality machine-generated annotations from a diverse vocabulary of 3,800+ visual entities. It comes with precomputed audio-visual features from billions of frames and audio segments, designed to fit on a single hard disk. This makes it possible to train a strong baseline model on this dataset in less than a day on a single GPU At the same time, the dataset's scale and diversity can enable deep exploration of complex audio-visual models that can take weeks to train even in a distributed fashion. 

As Joonseok Lee, one of the researchers who created the dataset explains on the the Google AI blog this is the second contest in which the Google AI's YouTube-8M dataset has been used. Last year 742 teams consisting of 946 individuals from 60 countries used the 2017 edition of the dataset to develop classification algorithms which accurately assigned video-level labels. As with this year's competition. the aim was to accelerate improvements in large-scale video understanding, representation learning, noisy data modeling, transfer learning and domain adaptation approaches that can help improve the machine-learning models that classify video. 

In addition to the competition, Google hosted an affiliated workshop at CVPR’17, inviting competition top-performers and researchers and share their ideas on how to advance the state-of-the-art in video understanding. This year the affiliated workshop will be held alongside ECCV'18 and winners of the Kaggle contest will be expected to write papers to present there. As the prize money - five amounts of $5,000-  is described as Travel Grants it looks as though they will be also be expected to attend in person.

The Kaggle contest is already underway with over 180 teams on the leaderboard. The entry deadline is July 30, which is also the team merger deadline. The final submission and model upload deadline is August 6 and August 13 is the paper submission deadline. While winners are obligated to submit papers, all participants are encouraged to do so.


More Information

Announcing an updated YouTube-8M, and the 2nd YouTube-8M Large-Scale Video Understanding Challenge and Workshop

2nd Workshop on YouTube-8M Large-Scale Video Understanding

The 2nd YouTube-8M Video Understanding Challenge


Related Articles

iNaturalist Kaggle Contest

Kaggle Enveloped By Google Cloud

Learn From Top Kagglers

Google AI At I/O

To be informed about new articles on I Programmer, sign up for our weekly newsletter, subscribe to the RSS feed and follow us on Twitter, Facebook or Linkedin.


Microsoft Introduces .NET Smart Components

Microsoft has provided a set of .NET Smart Components, described as a set of genuinely useful AI-powered UI components that you can quickly and easily add to .NET apps. The components are prebuilt end [ ... ]

GitHub Introduces Code Scanning

GitHub has announced a public beta of a code scanner that automatically fixes problems. The new feature was announced back in November, but has now moved to public beta status.  

More News


raspberry pi books



or email your comment to:

Last Updated ( Thursday, 07 June 2018 )