ABSTRACT
In this chapter, the authors present the explicit-motion-estimation-free Super-Resolution Reconstruction (SRR) from a different perspective. They also present the use of probabilistic motion within the framework of classic SRR, and develop the proposed algorithm. The probabilistic motion field is integrated into the classic SR framework, and ultimately results in a very simple family of algorithms. Classic super-resolution has long relied on very exact motion estimation for the recovery of sub-pixel details. Highly accurate general motion estimation, known as optical flow, is a severely under-determined problem. The warp operators depend on the scene and require highly accurate motion estimation for their construction. The authors describe how the proposed framework can be adapted to other re-sampling tasks, such as de-interlacing, inpainting and more, and start by explaining this extension intuitively. They demonstrate the abilities of the proposed algorithm in super-resolving general content sequences.
