Over the recent years, low-level visual descriptors, among which the most popular is the histogram of oriented gradients (HOG), have shown excellent performance in object detection and categorization. We form a hypothesis that the low-level image descriptors can be improved by learning the statistically relevant edge structures from natural images. We validate this hypothesis by introducing a new descriptor called the histogram of compositions (HoC). HoC exploits a learnt vocabulary of parts from a state-of-the-art hierarchical compositional model. Furthermore, we show that HoC is a complementary descriptor to HOG. We experimentally compare our descriptor to the popular HOG descriptor on the task of object categorization. We have observed approximately 4% improved categorization performance of HoC over HOG at lower dimensionality of the descriptor. Furthermore, in comparison to HOG, we show a categorization improvement of approximately 11% when combining HOG with the proposed HoC.
COBISS.SI-ID: 9519956
We have proposed an improved model for visual tracking using an adaptive coupled-layer visual model. The model is capable of tracking articulated objects using simple local visual descriptors, that are weakly connected into a geometric constellation. The model is capable of robustly adding and removing the local descriptors through probabilistic maps generated by the high level features such as motion and color. At the same time, the model allows utilization of additional probabilistic maps of arbitrary high-level features. The model was analyzed on a comprehensive database of video recordings and was compared to eleven current state-of-the-art trackers. The experiments have shown that the proposed tracker outperforms the state-of-the-art under several performance criteria.
COBISS.SI-ID: 9431124
Mobile robots need an effective spatial model for the successful operation in real-world environment. The model should be compact and simultaneously possess large expressive power. Moreover, it should scale well. In this paper we propose a new hierarchical representation of space, whose compositional structure is learned based on statistically significant observations. We have focused on a two dimensional space, since many robots perceive their surroundings in two dimensions with the use of a laser range finder or a sonar. We also propose the use of a low-level image descriptor for addressing the room classification problem, by which we demonstrate the performance of our representation. Using only the lower layers of the hierarchy, we obtain state-of-the-art classification results on demanding datasets.
COBISS.SI-ID: 9671508