Using the Forest to See the Trees: A Graphical Model Relating Features, Objects, and Scenes

Using the Forest to See the Trees: A Graphical Model Relating Features, Objects, and Scenes (2003)

Cached

Download Links

by Kevin Murphy , Antonio Torralba , William T. Freeman

Citations:

BibTeX

@MISC{Murphy03usingthe,
    author = {Kevin Murphy and Antonio Torralba and William T. Freeman},
    title = {Using the Forest to See the Trees: A Graphical Model Relating Features, Objects, and Scenes},
    year = {2003}
}

OpenURL

Abstract

Standard approaches to object detection focus on local patches of the image, and try to classify them as background or not. We propose to use the scene context (image as a whole) as an extra source of (global) information, to help resolve local ambiguities. We present a conditional random field for jointly solving the tasks of object detection and scene classification.

Keyphrases

graphical model relating feature standard approach local ambiguity local patch detection focus object detection extra source scene context scene classification conditional random field