You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have questions regarding the ComputeDistance function. If I understand correct, the function compute the similarity beteen the two hybrid descriptor, which is a key step in the top-down loop closure detection introduced in the paper.
Then,
I can not find the function compute the similarity between place descirptors. Do you implement it different from the paper?
This function seems have never been recalled in the program. I can only find it in its .h and .cpp file. Do you use it in the experiment?
The reason I ask is that I want to evaluate the loop closure detection module separately. So, I can avoid deploying the entire SLAM system and prepare light-weighted data. But I need to know how does the front-end works. Since the key loop closure detection function has not been recalled, I don't what it actually reads from the front-end and how does each descriptor been constructed.
Thanks
Chuhao
The text was updated successfully, but these errors were encountered:
Hi
Thanks for your interesting work.
I have questions regarding the ComputeDistance function. If I understand correct, the function compute the similarity beteen the two hybrid descriptor, which is a key step in the top-down loop closure detection introduced in the paper.
Then,
The reason I ask is that I want to evaluate the loop closure detection module separately. So, I can avoid deploying the entire SLAM system and prepare light-weighted data. But I need to know how does the front-end works. Since the key loop closure detection function has not been recalled, I don't what it actually reads from the front-end and how does each descriptor been constructed.
Thanks
Chuhao
The text was updated successfully, but these errors were encountered: