verlab / accelerated_features

Implementation of XFeat (CVPR 2024). Do you need robust and fast local feature extraction? You are in the right place!
https://www.verlab.dcc.ufmg.br/descriptors/xfeat_cvpr24
Apache License 2.0
736 stars 68 forks source link

Usage on multimodal data #25

Open Faheiermann opened 3 weeks ago

Faheiermann commented 3 weeks ago

Nice work!

Did you already tried some tests with multimodal data? Such as depth images and thermal?

guipotje commented 3 weeks ago

Hi @Faheiermann, thank you!

We haven't tested XFeat with multi-modal data, but we believe it is entirely possible to improve multi-modal matching efficiently. Please refer to issue #1 for more information.