ICCV 2013

Bayesian Joint Topic Modelling for Weakly Supervised Object Localisation

iccv13

Abstract
We address the problem of localisation of objects as bounding boxes in images with weak labels. This weakly supervised object localisation problem has been tackled in the past using discriminative models where each object class is localised independently from other classes. We propose a novel framework based on Bayesian joint topic modelling. Our framework has three distinctive advantages over previous works: (1) All object classes and image backgrounds are modelled jointly together in a single generative model so that “explaining away” inference can resolve ambiguity and lead to better learning and localisation.  (2) The Bayesian formulation of the model enables easy integration of prior knowledge about object appearance to compensate for limited supervision. (3) Our model can be learned with a mixture of weakly labelled and unlabelled data, allowing the large volume of unlabelled images on the Internet to be exploited for learning. Extensive experiments on the challenging VOC dataset demonstrate that our approach outperforms the state-of-the-art competitors.

 

Demo

 

Visualization

iccv2013_foreground

 

Citation

  1. Bayesian Joint Topic Modelling for Weakly Supervised Object Localisation
    PDF Poster Supplementary material
@InProceedings{Shi_2013_ICCV,
author = {Zhiyuan Shi and Timothy M. Hospedales and Tao Xiang},
title = {Bayesian Joint Topic Modelling for Weakly Supervised Object Localisation},
booktitle = {The IEEE International Conference on Computer Vision (ICCV)},
month = {December},
year = {2013}
}