Model-based 2D detection of unseen objects – XYZ-IBD

This leaderbord shows the ranking for Model-based 2D detection of unseen objects on XYZ-IBD. The reported time is the average time to generate predictions for all objects in a single image (averaged over the datasets).

Date (UTC) Submission Test image AP AP50 AP75 APS APM APL AR1 AR10 AR100 ARS ARM ARL Time (s)
2025-11-14 3PT-Detection (Nov 2025) RGB 0.575 0.742 0.666 -1.000 0.382 0.606 0.040 0.360 0.695 -1.000 0.625 0.720 2.040
2025-10-01 3PT-Detection (f.k.a IPT) RGB 0.534 0.699 0.604 -1.000 0.364 0.565 0.041 0.351 0.653 -1.000 0.514 0.685 6.072
2025-10-01 FOSOD-D (gdinob-dinov3b-cls-d-gth005) RGB-D 0.372 0.567 0.376 -1.000 0.185 0.403 0.041 0.276 0.533 -1.000 0.243 0.590 3.205
2025-08-26 MUSE (muse_full) RGB 0.323 0.475 0.332 -1.000 0.136 0.349 0.038 0.270 0.368 -1.000 0.139 0.403 0.671
2025-02-25 SAM6D(SAM) RGB-D 0.296 0.525 0.294 -1.000 0.118 0.335 0.034 0.241 0.357 -1.000 0.135 0.408 1.909
2025-02-25 CNOS(SAM) RGB 0.275 0.489 0.272 -1.000 0.118 0.307 0.034 0.228 0.349 -1.000 0.142 0.396 2.010
2025-09-30 BioDet RGB-D 0.269 0.501 0.251 -1.000 0.119 0.307 0.031 0.228 0.347 -1.000 0.154 0.395 2.082
2025-05-31 SAM6D(FastSAM) RGB-D 0.211 0.367 0.210 -1.000 0.086 0.256 0.037 0.199 0.274 -1.000 0.115 0.321 0.240

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

Latest submissions to this leaderboard: