Limits on Super-Resolution and How to Break Them
IEEE Transactions on Pattern Analysis and Machine Intelligence
Substituting crossed textons for super resolution of video
ACM SIGGRAPH 2006 Research posters
| Hi-index | 0.01 |
A video camera produces an image sequence with a specific frame rate (e.g. 30 frames per second). This imposes limits on the spatial resolution due to limited bandwidth. Image interpolation, such as bi-cubic interpolation, can increase the resolution, but yields a blurring of edges and image details. To create plausible high-frequency details in the blurred image, super-resolution technique has been a long studied area[Baker and Kanade 2002; Capel and Zisserman 2003]. However, it is difficult to apply these methods to a video sequence, since [Capel and Zisserman 2003] requires multiple images, and [Baker and Kanade 2002] requires a high computational cost. In order to resolve these problems, we proposed a method called "texton substitution"[Kamimura et al. 2006]. In this paper, we improve the method of texton substitution by using temporal connection.