Uncategorized

80 Million Tiny Images

Take a look at this infographic created by Antonio Torralba, Rob Fergus and William T. Freeman. Torralba teaches in the Computer Science at MIT. His past research centers on creating a lexical understanding of images — linking imagery and language. This work looks at tagged images, and creates an aggregate image, and maps the aggregates in a landscape of meaning.

Of the work he says:
With the advent of the Internet, billions of images are now freely available online and constitute a dense sampling of the visual world. Using a variety of non-parametric methods, we explore this world with the aid of a large dataset of 79,302,017 images collected from the Internet. Motivated by psychophysical results showing the remarkable tolerance of the human visual system to degradations in image resolution, the images in the dataset are stored as 32×32 color images. Each image is loosely labeled with one of the 75,062 non-abstract nouns in English, as listed in the Wordnet lexical database. Hence the image database gives a comprehensive coverage of all object categories and scenes. The semantic information from Wordnet can be used in conjunction with nearest-neighbor methods to perform object classification over a range of semantic levels minimizing the effects of labeling noise. When very many images are available, simple image indexing techniques can be used to retrieve images with similar object arrangements to the query image. If we have a big enough database then we can find, with high probability, images visually close to a query image, containing similar scenes with similar objects arranged in similar spatial configurations. If the images in the retrieval set are partially labeled, then we can propagate the labels to the query image, so performing classification.