I found the paper very short and dense. As it was stated in previous responses i found the paper somewhat dry also.

I think that the eight-point algorithm described in the paper was novel idea at its time. I wonder what other approaches emerged since than and how the performance of these new approaches compare to the one described in this paper. Mentioning the scenarios in which the algorithm fails was a useful one, however i did not gain enough insight into the details of the algorithms to be able to explain them. It would be nice to hear a presentation where one of these scenarios will be analyzed.