2D segmentation of unseen objects – Core datasets

This leaderbord shows the overall ranking for Task 6 on the core datasets (LM-O, T-LESS, TUD-L, IC-BIN, ITODD, HB, YCB-V). For each method, the date of the latest considered submission is reported. If more submissions of a method are available for a dataset, the submission with the highest APCore score is considered. The metrics are defined in Section 2 of the BOP challenge 2022 paper. The reported time is the average image processing time averaged over the core datasets.

Date (UTC) Method Test image APCore APLM-O APT-LESS APTUD-L APIC-BIN APITODD APHB APYCB-V Time (s)
2023-11-21 ViewInvDet RGB 0.426 0.410 0.385 0.464 0.312 0.252 0.545 0.616 1.700
2023-08-02 CNOS (FastSAM) RGB 0.412 0.397 0.374 0.480 0.270 0.254 0.511 0.599 0.221
2023-08-02 CNOS (SAM) RGB 0.403 0.396 0.397 0.391 0.284 0.282 0.480 0.595 1.847
2023-12-05 SAM6D RGB-D 0.481 0.460 0.451 0.569 0.357 0.332 0.593 0.605 2.795
2023-08-23 ZeroPose RGB 0.372 0.356 0.337 0.421 0.293 0.210 0.453 0.534 3.821
2023-09-28 lcc-fastsam RGB 0.149 0.167 0.097 0.153 0.110 0.041 0.253 0.220 1.182
2023-12-05 SAM6D-FastSAM RGB-D 0.449 0.422 0.420 0.517 0.293 0.319 0.548 0.621 0.445

Tip: Hover over the numbers to see more decimal places.

Latest submissions to this leaderboard: