Connectionists: CORe50: a new Dataset and Benchmark for Continuous/Lifelong Object Recognition

[vincenzo lomonaco]

Dear all,

We are happy to announce that the CORe50 dataset from the paper 
“/CORe50: a new Dataset and Benchmark for Continuous Object 
Recognition/” (CoRL, 2017) is now publicly available at the link: 
https://vlomonaco.github.io/core50/

*CORe50*, specifically designed for /Continuous/Lifelong Learning/ and 
/Object Recognition/, is a collection of more than 500 videos (30fps) of 
50 domestic objects belonging to 10 different categories. Instance level 
granularity, temporal coherence, first-person point-of-view and very 
different environmental conditions and backgrounds (outdoor sessions 
included) is what makes CORe50 particularly useful for assessing 
/Continuous Learning/ techniques.

On top of the CORe50 dataset we are also happy to release a 3-way 
benchmark with multiple baselines and a living leaderboard to keep track 
of the progresses made in this new exciting area! For downloading the 
code, the dataset, the benchmark and more, check out the official 
website https://vlomonaco.github.io/core50/

Best regards,
Vincenzo
Vincenzo Lomonaco, M.Sc.
PhD Students' Representative @ DISI, University of Bologna
http://www.vincenzolomonaco.com/
Linkedin <http://it.linkedin.com/in/vincenzolomonaco/> Twitter 
<https://twitter.com/v_lomonaco> Facebook 
<https://www.facebook.com/vincenzo.lomonaco.91> Google+ 
<https://plus.google.com/u/0/+VincenzoLomonaco>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/connectionists/attachments/20180106/26f9f36b/attachment.html>