Upload your images to have them classified by a machine! Upload multiple images using the button below or dropping them on this page. The predicted objects out of 1,000 categories will be refreshed automatically. Images are resized such that the smallest dimension becomes 256, then the center 256x256 crop is used. More about this NYU demo can be found here. The model in the demo is early work towards the following paper: "Visualizing and Understanding Convolutional Networks", Matthew Zeiler and Rob Fergus.
To see a live real-time demo of the current best object recognition system trained on 10,000 categories, checkout www.clarifai.com.