Suppose only 1MB camera that will result into somewhat readable photos of written text. Now HDRI is able to remove shadows and make photos sometimes better than alone. My friend said that in the future it will be possible to make blurry or even low-resolution images precise. In practice meaning that poor APS-capability-devices could achieve very good photos, APS here. Now what are such algorithms that make let say 10 blurry/poor pictures from the same object into one precise picture from slightly different angles? How is it done and where can I find more info about this?
I often have the problem with different devices such as iPad 2, phone cameras and other. I would like to find some general methods to process such photos and how to do it? And what kind of problems this kind of usually 3D cases have? Does there exist ready linear-programming-reduced algorithms for this?
I think you are asking about Superresolution.
http://en.wikipedia.org/wiki/Superresolution
I was playing around with a matlab program once that took four photos and turned them into a much higher resolution photo--this is one example of Superresolution.
I think it might have been this one: http://lcav.epfl.ch/software/superresolution