Abstract:

Topic models (e.g., pLSA, LDA, SLDA) have been widely used for segmenting imagery. These models are confined to crisp segmentation. Yet, there are many images in which some regions cannot be assigned a crisp label (e.g., transition regions between a foggy sky and the ground or between sand and water at a beach). In these cases, a visual word is best represented with partial memberships across multiple topics. To address this, we present a partial membership latent Dirichlet allocation (PM-LDA) model and associated parameter estimation algorithms. Experimental results on two natural image datasets and one SONAR image dataset show that PM-LDA can produce both crisp and soft semantic image segmentations; a capability existing methods do not have.

Links:

Citation:

C. Chen, A. Zare, and J. T. Cobb, “Partial Membership Latent Dirichlet Allocation for Image Segmentation,” in Int. Conf. Pattern Recognition (ICPR), 2016. 

@InProceedings{chen2016partialicpr,
author = {Chen, Chao and Zare, Alina and Cobb, James T.},
title = {Partial Membership Latent Dirichlet Allocation for Image Segmentation},
booktitle = {Int. Conf. Pattern Recognition (ICPR)},
year = {2016},
}