Cue Integration for Figure/Ground Labeling
Xiaofeng Ren, Charless Fowlkes, Jitendra Malik
icon We present a model of edge and region grouping using a conditional random field built over a scale-invariant representation of images to inte- grate multiple cues. Our model includes potentials that capture low-level similarity, mid-level curvilinear continuity and high-level object shape. Maximum likelihood parameters for the model are learned from human labeled groundtruth on a large collection of horse images using belief propagation. Using held out test data, we quantify the information gained by incorporating generic mid-level cues and high-level shape.

Download: pdf

Text Reference

Xiaofeng Ren, Charless Fowlkes, and Jitendra Malik. Cue integration for figure/ground labeling. In Y. Weiss, B. Schölkopf, and J. Platt, editors, Advances in Neural Information Processing Systems 18, pages 1121–1128. MIT Press, Cambridge, MA, 2006.

BibTeX Reference

@incollection{RenFM_NIPS_2005,
    author = "Ren, Xiaofeng and Fowlkes, Charless and Malik, Jitendra",
    editor = {Weiss, Y. and Sch\"{o}lkopf, B. and Platt, J.},
    title = "Cue Integration for Figure/Ground Labeling",
    booktitle = "Advances in Neural Information Processing Systems 18",
    publisher = "MIT Press",
    address = "Cambridge, MA",
    pages = "1121--1128",
    year = "2006",
    tag = "grouping,object_recognition"
}