A novel dataset, called MICC-Flickr101, based on the popular Caltech 101 and collected from Flickr. This new collection was conceived with the idea of cloning a reference datasets as Caltech 101, but using realistic representations and also textual descriptions.
If you use this dataset, please cite our paper:
@inproceedings{ icpr12, author = {Ballan, Lamberto and Bertini, Marco and Del Bimbo, Alberto and Serain, Andrea M. and Serra, Giuseppe and Zaccone, Benito F.}, title = {Combining Generative and Discriminative Models for Classifying Social Images from 101 Object Categories}, booktitle = {Proc. of International Conference on Pattern Recognition (ICPR)}, address = {Tsukuba, Japan}, month = {November}, note = {(Poster)}, year = {2012} }
Pingback: Lamberto Ballan, Ph.D. » Generative & discriminative models for classifying social images on the MICC-Flickr101 dataset
Pingback: Generative & discriminative models for classifying social images on the MICC-Flickr101 dataset | Giuseppe Serra
Pingback: Lamberto Ballan, PhD » Generative & discriminative models for classifying social images on the MICC-Flickr101 dataset
Pingback: Image classification using multimodal generative-discriminative methods | MICC blog - Insights on research