Ironically this video is too blurry to do the demonstration justice, I would like to see the results in greater detail. From another presentation (deblur portion here) the author mentions that they are collaborating with researchers at MIT so I would suspect that they are using this method. Paper and source code are available if you're interested in learning more or trying it out, though I would suspect that improvements have been made since then. via
0 comments:
Post a Comment