Conditional random fields (CRF) are widely used for predicting output variables that have some internal structure. Most of the CRF research has been done on structured classificati...
This paper presents a novel system that employs an adaptive neural network for the no-reference assessment of perceived quality of JPEG/JPEG2000 coded images. The adaptive neural ...
Regularities in the world are human defined. Patterns in the observed phenomena are there because we define and recognize them as such. Automatic pattern recognition tries to bridg...
Optical lithography is a critical step in the semiconductor manufacturing process, and one key problem is the design of the photomask for a particular circuit pattern, given the o...
We present an approach for efficiently recognizing all objects in a scene and estimating their full pose from multiple views. Our approach builds upon a state of the art single-vie...