Natural Scene Categorization in Humans and Computers

For both humans and machines, the ability to learn and categorize natural scenes as well as the objects within is an essential and important functionality. The bulk of this talk will focus on a computer vision model we developed recently to tackle to the problem of categorizing complex real-life images. To motivate this topic, I will present a series of recent human psychophysics studies on natural scene recognition. All these experiments converge to one prominent phenomena of the human visual system: humans are extremely efficient and rapid in capturing the overall gist of natural images. We can categorize a scene as a beach image or a rock concert image in literally a split of a second. Can we achieve such a feat in computer vision modeling? We propose here a generative Bayesian hierarchical model that learns to categorize natural images in a weakly supervised fashion. We represent an image by a collection of local regions, denoted as codewords obtained by unsupervised clustering. Each region is then represented as part of a `theme’. In previous work, such themes were learnt from hand-annotations of experts, while our method learns the theme distribution as well as the codewords distribution over the themes without such supervision. We report excellent categorization performances on a large set of 13 categories of complex scenes.

Speaker Details

Prof. Fei-Fei Li’s main research interest is vision, particularly high-level visual recognition. In computer vision, Fei-Fei has worked on both object and natural scene recognition. In human vision, she has studied the interaction of attention and natural scene and object recognition. Fei-Fei graduated from Princeton University in 1999 with a physics degree. She received her PhD in electrical engineering from the California Institute of Technology in 2005. In the spring of 2005, she was a visiting scientist at the Microsoft Research Center in Cambridge, UK. Fei-Fei became an Assistant Professor of Electrical and Computer Engineering at UIUC in July 2005. She is now a full-time faculty member at the Beckman Institute.

Date:
Speakers:
Fei-Fei Li
Affiliation:
University of Illinois Urbana-Champaign (UIUC)
    • Portrait of Jeff Running

      Jeff Running