search

Parskatt / DKM

[CVPR 2023] DKM: Dense Kernelized Feature Matching for Geometry Estimation
https://parskatt.github.io/DKM/
Other
378 stars 28 forks source link

if it's possible to use DKMv3 for training when there is no depth information in my own dataset? #49

Open DiKu-nn opened 9 months ago

DiKu-nn commented 9 months ago

Hello, may I ask if it's possible to use DKMv3 for training when there is no depth information in my own dataset? I only need to obtain the horizontal and vertical offsets for perspective transformation.

Parskatt commented 9 months ago

Hi. You would need to rewrite the loss function a bit, but it works just fine for homographies as well. It works with anything that creates a warp.

DiKu-nn commented 9 months ago

Thank you, I'll give it a try.

lqx75 commented 5 months ago

Thank you, I'll give it a try.

Hello, may I ask how you have been trying? If successful, would you be willing to open source? Thank you！