mattcyu1
commented
8 years ago Dear Ang,
The points were generated using the following MATLAB script (not developed by us): http://www.mathworks.com/matlabcentral/fileexchange/28842-grid-sphere
The alternative is to generate millions of random points, uniformly distributed on the sphere. In our tests, using the fixed points in the MATLAB script I linked gives very similar results while needing to sample less points than when sampling randomly.
Sincerely, Matt
On Tue, Apr 19, 2016 at 3:48 AM, Rouen007 notifications@github.com wrote:
I am total agree with the idea of S-PSNR and L-PSNR described in "A Framework to Evaluate Omnidirectional Video Coding Schemes". But I am confused with how to determine the uniformly distribution on the sphere , e.g., the 655362 positions, when I see the implementation of the source code. Did any information, theories or references support the correctness of the distributation of those points?
Thank you very much.
Ang Lu luangss@zju.edu.cn
— You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub https://github.com/mattcyu1/omnieval/issues/1
Rouen007
I am total agree with the idea of S-PSNR and L-PSNR described in "A Framework to Evaluate Omnidirectional Video Coding Schemes". But I am confused with how to determine the uniformly distribution on the sphere , e.g., the 655362 positions, when I see the implementation of the source code. Did any information, theories or references support the correctness of the distributation of those points?
Thank you very much.
Ang Lu luangss@zju.edu.cn