Closed Lucifer1002 closed 2 years ago
Hello, I read your paper, there is a combination of SIFT+Superglue in your comparative experiment, how do you solve the problem that the descriptor and matching network dimensions are inconsistent
Hi, network channels is configurable and we adjust it with descriptor channels number.
Hello, I read your paper, there is a combination of SIFT+Superglue in your comparative experiment, how do you solve the problem that the descriptor and matching network dimensions are inconsistent