Closed Parskatt closed 1 year ago
See #12 , basically we have accidentally used LO-RANSAC to evaluate all descriptor methods instead of regular RANSAC. I'm collecting updated results below, snd will update the paper when done.
Mega-1500:
DeDoDe-B: [49.4, 65.5, 77.7] DeDoDe-G: [52.8, 69.7, 82.0] ALIKED: [41.9, 58.4, 71.7] DISK: [35.0, 51.4, 64.9] SiLK: [39.9, 55.1, 66.9] SIFT: [35.0, 50.4, 63.9]
Note that this does not affect the detector scores and inc22 evals of the paper.
Updated scores in paper, arxiv update should come tomorrow or so.
See #12 , basically we have accidentally used LO-RANSAC to evaluate all descriptor methods instead of regular RANSAC. I'm collecting updated results below, snd will update the paper when done.
Mega-1500:
DeDoDe-B: [49.4, 65.5, 77.7] DeDoDe-G: [52.8, 69.7, 82.0] ALIKED: [41.9, 58.4, 71.7] DISK: [35.0, 51.4, 64.9] SiLK: [39.9, 55.1, 66.9] SIFT: [35.0, 50.4, 63.9]