Labeling objects in images is an essential prerequisite for many visual learning and recognition applications that depend on training data, such as image retrieval, object detecti...
This paper presents a novel statistical shape model that can be used to detect and localise feature points of a class of objects in images. The shape model is inspired from the 3D...
Abstract. The archetype of many novel research activities is called cognition. Although separate definitions exist to define a technical cognitive system, it is typically character...
We present a novel algorithm, Compact Kd-Trees (CompactKdt), that achieves state-of-the-art performance in searching large scale object image collections. The algorithm uses an or...
We address the problem of learning object class models and object segmentations from unannotated images. We introduce LOCUS (Learning Object Classes with Unsupervised Segmentation...