Neural networks are a powerful technology for classification of visual inputs arising from documents. However, there is a confusing plethora of different neural network methods th...
Recent works in object recognition often use visual words, i.e. vector quantized local descriptors extracted from the images. In this paper we present a novel method to build such ...
We present a novel rendering system for defocus blur and lens effects. It supports physically-based rendering and outperforms previous approaches by involving a novel GPU-based tr...
1 In this paper we present a novel multiscale approach to recovery of nonrigid motion from sequences of registered intensity and range images. The main idea of our approach is that...
This paper presents a new reliable detection of core and delta in fingerprints by candidate analysis. The line search based method, which is one of the reliable approaches to dete...