EE Seminar: Dr. Tomer Michaeli, Weizmann Institute

~~סמינר מחלקתי
You are invited to attend a lecture by

Dr. Tomer Michaeli
(Weizmann Institute of Science)

on the subject:

 Blind deblurring and blind super-resolution
using internal patch recurrence

Small image patches tend to recur at multiple scales within high-quality natural images. This fractal-like behavior has been used in the past for various tasks including image compression, super-resolution and denoising. In this talk, I will show that this phenomenon can also be harnessed for "blind deblurring"  and for "blind super-resolution", that is for removing blur or increasing resolution without a-priori knowledge of the associated blur kernel. Our key observation is that the source of the patch recurrence phenomenon is the repetitions of structures at various scales in the continuous scene. Therefore, the way by which this continuous-domain phenomenon is manifested in the discrete image, actually encodes information on the imaging process.
We adopt a continuous-domain model of natural scenes, and use it to study the expected behavior of discrete natural images. This model suggests that the deviations from ideal patch recurrence in a discrete image can be used for recovering the unknown camera blur kernel. More specifically, we show that the correct blur kernel is the one which maximizes the similarity between patches across scales of the image. Extensive experiments indicate that our approach leads to state of the art results, both in deblurring and in super-resolution.