Rotation invariance computation

[eokeeffe]

Thanks for the code, the paper was brilliant!
I'm a little lost on just how you implemented the rotation invariance though

In terms of the rotation invariance, the system you have is to build a set of MIMs where the indexes for the first convolution is
treated as a circular list, whereby you start from the 0th index and go to Nth index (N being number of angles), then proceed to increment the starting index i.e.
[0 1 2 3 4 5] (original order)
[1 2 3 4 5 0]
[2 3 4 5 0 1]
[3 4 5 0 1 2]
[4 5 0 1 2 3]
[5 0 1 2 3 4]

I'm a little lost on the dominant orientation method you mentioned in the paper, is this the method used in the likes of HOPC or SIFT ?

[woshiybc]

Thanks for the code, the paper was brilliant! I'm a little lost on just how you implemented the rotation invariance though

In terms of the rotation invariance, the system you have is to build a set of MIMs where the indexes for the first convolution is treated as a circular list, whereby you start from the 0th index and go to Nth index (N being number of angles), then proceed to increment the starting index i.e. [0 1 2 3 4 5] (original order) [1 2 3 4 5 0] [2 3 4 5 0 1] [3 4 5 0 1 2] [4 5 0 1 2 3] [5 0 1 2 3 4]

I'm a little lost on the dominant orientation method you mentioned in the paper, is this the method used in the likes of HOPC or SIFT ?

Hello, I also encountered the same problem. Have you solved the rotation invariance mentioned by the author?

[SteveEinstain]

Hello, I have encountered the same problem, how did you solve it?

[LJY-RS]

你好，请查看我新的代码项目RIFT2，谢谢