Unsupervised deep object discovery for instance recognition
share
Severe background clutter is challenging in many computer vision tasks, including large-scale image retrieval. Global descriptors, that are popular due to their memory and search efficiency, are especially prone to corruption by such clutter. Eliminating the impact of the clutter on the image descriptor increases the chance of retrieving relevant images as well as preventing topic drift by actually retrieving the clutter in the case of query expansion. In this work, we propose a novel salient region detection method. It captures, in an unsupervised manner, patterns that are both discriminative and common in the dataset. The descriptors derived on the salient regions improve particular object retrieval, most noticeably in a large collections containing small objects.
READ FULL TEXT