The Microsoft COCO (mscoco) is an image recognition and segmentation dataset which contains more 300k images for more than 70 categories. Other features: More than 70 categories Object segmentation Recognition in Context Multiple objects per image More than 300,000 images More than 2 Million instances See Microsoft COCO: Common Objects in Context Tsung-Yi Lin, Michael Maire, Serge Belongie, James Hays, Pietro Perona, Deva Ramanan, Piotr Dollr, C. Lawrence Zitnick (Submitted on 1 May 2014 (v1), last revised 5 Jul 2014 (this version, v2))

Related datasets