Dear friends and colleagues,
We are excited to announce the 2010 Spring Release of ImageNet
(
www.image-net.org), now totaling more than 11 million images over
15500 visual categories, ranging from activities, scenes, to objects.
Please check out the details at
http://www.image-net.org/spring2010
Some highlights include:
- We provide bounding box annotations for nearly 1000 popular visual
categories (200 images per category);
- We provide densely sampled SIFT features for 1.2 million images and
1000 categories. We provide both raw descriptors and quantized
codewords, as well as their spatial coordinates.
- We provide a MATLAB Toolbox to help you make use of the WordNet
hierarchical structure behind the visual categories, and enable you to
efficiently download the full resolution images of a subtree.
- Partnering with the PASCAL VOC 2010, we call for participation in the
ImageNet Large-Scale Visual Recognition Challenge 2010 (ILSVRC2010).
Come and show off your algorithms to recognize 1000 visual categories!
For more information, please check out:
http://www.image-net.org/challenges/LSVRC/2010/
- Users can now browse ImageNet categories by their WordNet structure,
as well popularity ranking.
As always, we hope that ImageNet can provide valuable data for your
research. If there is anything we could do to improve and help, please
do not hesitate to contact us at
sup...@image-net.org.
Best regards,
ImageNet Team
--
---------------------------------------------
Li, Fei-Fei Ph.D.
(publish under L. Fei-Fei)
Assistant Professor
Computer Science Dept.
Stanford University
353 Serra Mall, 2A Room 246
Stanford, CA 94305-9025
Tel:
(650)725-3860
Website:
http://vision.stanford.edu
---------------------------------------------