search

OnderT / XoFTR

XoFTR is a cross-modal cross-view method for local feature matching between thermal infrared (TIR) and visible images.
Apache License 2.0
58 stars 0 forks source link

Model training #3

Closed 13559323996 closed 2 months ago

13559323996 commented 2 months ago

Can this dataset be trained on LoFTR models?

OnderT commented 2 months ago

Unfortunately, no. LoFTR uses depth maps to generate ground truth matches during training. Since our dataset (METU-VisTIR) lacks depth information, it is not possible to train LoFTR directly on it.

13559323996 commented 2 months ago

Thank you for your answers.