# Bayesian Joint Modelling for Object Localisation in Weakly Labelled Images

## TL;DR

This paper introduces a Bayesian joint modelling framework for weakly supervised object localisation that models multiple objects and backgrounds simultaneously, leveraging unlabelled data and prior knowledge for improved accuracy.

## Contribution

The novel Bayesian joint topic model captures multiple objects and shared backgrounds in a single framework, enabling better localisation and learning from weakly labelled and unlabelled data.

## Key findings

- Effective localisation on PASCAL VOC, ImageNet, and YouTube-Object datasets.
- Outperforms existing weakly supervised localisation methods.
- Utilizes unlabelled data and prior knowledge to enhance learning.

## Abstract

We address the problem of localisation of objects as bounding boxes in images and videos 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. In this paper, a novel framework based on Bayesian joint topic modelling is proposed, which differs significantly from the existing ones in that: (1) All foreground object classes are modelled jointly in a single generative model that encodes multiple object co-existence so that "explaining away" inference can resolve ambiguity and lead to better learning and localisation. (2) Image backgrounds are shared across classes to better learn varying surroundings and "push out" objects of interest. (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. Moreover, the Bayesian formulation enables the exploitation of various types of prior knowledge to compensate for the limited supervision offered by weakly labelled data, as well as Bayesian domain adaptation for transfer learning. Extensive experiments on the PASCAL VOC, ImageNet and YouTube-Object videos datasets demonstrate the effectiveness of our Bayesian joint model for weakly supervised object localisation.