Probabilistic joint image segmentation and labeling

A. Ion, J. Carreira, C. Sminchisescu, in:, Neural Information Processing Systems, 2011, pp. 1827–1835.

Download
No fulltext has been uploaded. References only!
Conference Paper | Published | English
Author
; ;
Department
Abstract
We present a joint image segmentation and labeling model (JSL) which, given a bag of figure-ground segment hypotheses extracted at multiple image locations and scales, constructs a joint probability distribution over both the compatible image interpretations (tilings or image segmentations) composed from those segments, and over their labeling into categories. The process of drawing samples from the joint distribution can be interpreted as first sampling tilings, modeled as maximal cliques, from a graph connecting spatially non-overlapping segments in the bag [1], followed by sampling labels for those segments, conditioned on the choice of a particular tiling. We learn the segmentation and labeling parameters jointly, based on Maximum Likelihood with a novel Incremental Saddle Point estimation procedure. The partition function over tilings and labelings is increasingly more accurately approximated by including incorrect configurations that a not-yet-competent model rates probable during learning. We show that the proposed methodologymatches the current state of the art in the Stanford dataset [2], as well as in VOC2010, where 41.7% accuracy on the test set is achieved.
Publishing Year
Date Published
2011-01-01
Page
1827 - 1835
Conference
NIPS: Neural Information Processing Systems
Conference Location
Granada, Spain
Conference Date
2011-12-12 – 2011-12-14
IST-REx-ID

Cite this

Ion A, Carreira J, Sminchisescu C. Probabilistic joint image segmentation and labeling. In: Neural Information Processing Systems; 2011:1827-1835.
Ion, A., Carreira, J., & Sminchisescu, C. (2011). Probabilistic joint image segmentation and labeling (pp. 1827–1835). Presented at the NIPS: Neural Information Processing Systems, Granada, Spain: Neural Information Processing Systems.
Ion, Adrian, Joao Carreira, and Cristian Sminchisescu. “Probabilistic Joint Image Segmentation and Labeling,” 1827–35. Neural Information Processing Systems, 2011.
A. Ion, J. Carreira, and C. Sminchisescu, “Probabilistic joint image segmentation and labeling,” presented at the NIPS: Neural Information Processing Systems, Granada, Spain, 2011, pp. 1827–1835.
Ion A, Carreira J, Sminchisescu C. 2011. Probabilistic joint image segmentation and labeling. NIPS: Neural Information Processing Systems 1827–1835.
Ion, Adrian, et al. Probabilistic Joint Image Segmentation and Labeling. Neural Information Processing Systems, 2011, pp. 1827–35.

Export

Marked Publications

Open Data IST Research Explorer

Search this title in

Google Scholar