Many recently proposed perceptual image quality assessment algorithms are implemented in two stages. In the first stage, image quality is evaluated within local regions. This resu...
In this paper, we propose a novel patch-based face hallucination framework, which employs a dual model to hallucinate different components associated with one facial image. Our mod...
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 ...
In this paper two efficient unsupervised video object segmentation approaches are proposed and then extensively compared in terms of computational cost and quality of segmentation...
Nikolaos D. Doulamis, Anastasios D. Doulamis, Stef...
This paper describes a novel approach to 3D motion estimation of planar objects based on eigennormalization, Expansion Matching (EXM) and a scaled orthographic projection model. O...