Weakly Supervised Detection with Posterior Regularization
In Proceedings British Machine Vision Conference 2014
AbstractThis paper focuses on the problem of object localization when the annotation is at training time restricted to presence or absence of object instances at image level. We present a method based on features extracted from a Convolutional Neural Network and latent SVM that can represent and exploit the presence of multiple object instances in an image. Moreover, our approach can better guide the localization of the object instances in the image by incorporating in the learning procedure additional constraints that represent domain-specific knowledge such as symmetry and mutual exclusion. We show that the proposed method outperforms the state-of-the-art in weakly-supervised object localization and object classification on the Pascal VOC 2007 dataset.
FilesExtended Abstract (PDF, 1 page, 402K)
Paper (PDF, 12 pages, 869K)