This paper presents a new method named text to visual synthesis with appearance models (TEVISAM) for generating videorealistic talking heads. In a first step, the system learns a ...
A method of using a moving texture alpha-mask to represent scientific data is described, for the purpose of visualizing continuous fluid dynamics fields. The method combines strea...
Abstract. This paper evaluates strategies to combine multiple segmentations of the same image, generated for example by different segmentation methods or by different human experts...
Torsten Rohlfing, Daniel B. Russakoff, Calvin R. M...
—When an image of a scene is captured by a camera through a fence, a blurred fence image interrupts objects in the scene. In this paper, we propose a method for a fence removal f...
Recently researchers have introduced methods to develop reusable knowledge in reinforcement learning (RL). In this paper, we define simple principles to combine skills in reinforce...