[visionlist] iNaturalist 2017 large scale image classification dataset

[Oisin Mac Aodha]

We are pleased to announce a new image classification challenge featuring
over 5,000 natural categories - from Abaeis nicippe to Zosterops lateralis.


The dataset contains 675,000 training images and includes many visually
similar species, captured in a wide variety of situations, by different
individuals, from all over the world. The goal of the competition is to
push the state of the art in automatic image classification for real world
data that features fine-grained categories, big class imbalances, and large
numbers of classes.

Results will be presented at the Fine-Grained Visual Categorization Workshop
<http://www.fgvc.org/> held at CVPR 2017 this summer.


For more details please check out the competition website
<https://github.com/visipedia/inat_comp>.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://visionscience.com/pipermail/visionlist_visionscience.com/attachments/20170413/6e004f81/attachment.html>