An edge-adaptive structure tensor kernel regression for image interpolation
Hongyi Liu, Zhihui Wei · 2010
This paper investigates image interpolation problems by means of combing structure tensor with kernel regression methods. We propose an anisotropic structure tensor to adapt the diffusivity along or across the geometric structure of image, for instance image edges. Experimental results show that our new algorithm significantly outperforms the steering kernel regression in persevering edge and improving the visual quality.