search

Parskatt / DeDoDe

[3DV 2024 Oral] DeDoDe 🎶 Detect, Don't Describe --- Describe, Don't Detect, for Local Feature Matching
MIT License
368 stars 19 forks source link

Any reason for dual softmax matcher for inference? #18

Open ducha-aiki opened 11 months ago

ducha-aiki commented 11 months ago

Hi,

I am wondering, can we just use mutual nearest neighbor matching for speed at inference time? Dual softmax normalizes the scores, but doesn't change the actual neighbors (matches).

Parskatt commented 11 months ago

We have a confidence threshold for the matches based on the ds scores. If you remove that youll get more outliers.

Otherwise yes, I see no major issue.