101 object categories collected by Fei-Fei Li, Marco Andreetto, and Marc 'Aurelio Ranzato. Each object category contains about 40 to 800 images. Size of each image is roughly 300 x 200 pixels. All images are annotated with the following information: a bounding box of the object, and a carefully traced silhouette of the objects by a human subject.
(c) California Institute of Technology, September, 2003.
(for downloading a .tar.gz file of all images in all categories, click here)
(for downloading a .tar file of all annotations of the dataset, click here)
-
L. Fei-Fei, R. Fergus, and P. Perona. Learning generative visual models from few training examples: an incremental Bayesian approach tested on 101 object categories. CVPR 2004, Workshop on Generative-Model Based Vision. 2004
-
L. Fei-Fei, R. Fergus, and P. Perona. One-Shot learning of Object Categories. submitted, 2004.
100 naturally shot photographs collected by Fei-Fei Li. Size of each image is 384 x 256.
(c) California Institute of Technology, July 2002.
-
L. Fei-Fei, A. Iyer, C. Koch and P. Perona. What do we see at a glance of a scene? Proceedings of the Vision Sciences Society annual meeting, Sarasota, 2004.
- L. Fei-Fei, R. Fergus, and P. Perona. A Bayesian approach to unsupervised One-Shot learning of Object categories. Proc. ICCV. 2003
research homepage for Fei-Fei Li