We describe a technique of linguistic post-processing of whole-book recognition results. Whole-book recognition is a technique that improves recognition of book images using fully...
The evolution of any software product over its lifetime is unavoidable, caused both by bugs to be fixed and by new requirements appearing in the later stages of the product's...
An increasing number of NLP tasks require semantic labels to be assigned, not only to entities that appear in textual elements, but to the relationships between those entities. In...
In this paper, we propose a new method for rendering image-based objects. An object is decomposed into a set of 1D displacement textures in a preprocessing stage. During rendering...
Abstract. A convolutional network architecture termed sparse convolutional neural network (SCNN) is proposed and tested on a real-world classification task (car classification). In...