Title
Spatially-Varying Super-Resolution for HDTV


Authors
Chih-Tsung Shen, Graduate Institute of Networking and Multimedia, National Taiwan University, Taipei
Hung-Hsun Liu, Telecommunication Laboratories, Chunghwa Telecom Co., Taoyuan.
Ming-Sui Lee, Graduate Institute of Networking and Multimedia, National Taiwan University, Taipei
Yi-Ping Hung, Graduate Institute of Networking and Multimedia, National Taiwan University, Taipei
Soo-Chang Pei, Department of Electrical Engineering, National Taiwan University, Taipei.


Abstract
We propose a system to up-sample visual signals
for high definition televisions (HDTVs). Although the original
visual signals are degraded and limited, we still try to solve
these problems by using super-resolution. First, we interpolate
the visual signals with a edge taper to the desired size. Then, we
combine L1 data term, L2 data term, Tikhonov-like regularizer
and total variation regularizer with a saliency weighting for our
L12TTV deblurring method. Adopting our L12TTV deblurring,
we can remove the spatially-varying blurry conditions of the
interpolated signals. Experimental results show that our system
outperforms than comparisons in both image and video cases.


Talk Slides
We provide the slides: [Slides]

Experimental Results on Images (Blurry Input, Bicubic, GeniueFractals2012, Ours)









Experimental Resulting Frames on Videos

Bicubic interpolation, Video Enhancer, Shan et al.'s resulting frame, and our resulting frame.
Please notice the regions pointed by the red pointers.


Bicubic interpolation, QE super-resolution, Shan et al.'s resulting frame, and our resulting frame.
Please notice the regions pointed by the red pointers.


Bicubic interpolation, QE super-resolution, Video Enhancer, and our resulting frame.
Please notice the regions pointed by the red pointers.
The original image is from a MPEG-1 clip.


Bicubic interpolation, QE super-resolution, Video Enhancer, and our resulting frame.
Please notice the regions pointed by the red pointers.
The original image is from a MPEG-1 clip.


Bicubic interpolation, QE super-resolution, Video Enhancer, and our resulting frame.
Please notice the regions pointed by the red pointers. Jet Li's eyes become clearer. The background regions become sharper.
The original image is from a MPEG-1 clip.


Bicubic interpolation, QE super-resolution, Video Enhancer, and our resulting frame.
Please notice the regions pointed by the red pointers. The girl's eyes become clearer. The Chinese characters and the background become sharper.
The original image is from a MPEG-1 clip.