prunetruong / gocor Goto Github PK

Hello, thank you for sharing the work, and congratulations.
I wanted to understand the upper and lower bound of confidence score for Gocor global correspondence, I have implemented your method in my research and I get confidence range in [-2, +7] but sometimes only in the range of [-0.2, +0.3], so I wanted to understand when should I consider it to be a good match? or where should I put a bound to it? looking at the paper I understand that score <= 0 refers to non-match, but for positive confidence, there is no defined bound.
Thank You

Hi, I'm interested in applying this method in my research, do you have an estimate of when to release to code?

Hello,
Thanks for sharing the code. Are the vggPyramid features pretrained and fixed during training the model (I noticed in the test_model they are fixed)? I was wondering if training from scratch would affect the convergence behavior of the gocor module. Thank you in advance for sharing your insight :)

Cheers,
Fatemeh

Hi, thanks for sharing this nice work!
I have a question about Lq loss. Why can query frame objective encapsulate constraints that mentioned in paper 3.5 by minimizing the Lq loss?

Thanks for your execlent work...Im not a math student and I get some confusion about derivation in Appendix A.

1. why in the formula 10 ? It seems that is the taylor expansion of L(w)... So should be Hessian matrix of w?

2. How to expand to Jr,Jq and lambda in the formula 23 specificly? how to use a math formula express relationship between Jn,Jr an Jq?