Model-based 2D detection of unseen objects – IPD

This leaderbord shows the ranking for Model-based 2D detection of unseen objects on IPD. 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 0.634 0.769 0.743 -1.000 0.507 0.731 0.231 0.844 0.851 -1.000 0.795 0.883 1.604
2025-10-01 3PT-Detection (f.k.a IPT) RGB 0.631 0.771 0.741 -1.000 0.500 0.712 0.230 0.839 0.842 -1.000 0.775 0.874 14.303
2025-09-30 BioDet RGB-D 0.473 0.653 0.523 -1.000 0.416 0.547 0.210 0.630 0.634 -1.000 0.563 0.703 2.075
2025-10-01 FOSOD-D (gdinob-dinov3b-ffa-d-gth01) RGB-D 0.369 0.517 0.428 -1.000 0.326 0.397 0.202 0.597 0.607 -1.000 0.530 0.662 2.271
2025-10-01 FOSOD-D (gdinob-dinov3b-cls-d-gth01) RGB-D 0.354 0.500 0.411 -1.000 0.311 0.396 0.203 0.603 0.614 -1.000 0.529 0.666 2.179
2025-02-28 SAM6D(SAM) RGB-D 0.317 0.473 0.357 -1.000 0.263 0.444 0.155 0.406 0.406 -1.000 0.291 0.629 2.061
2025-03-01 SAM6D(FastSAM) RGB-D 0.300 0.446 0.335 -1.000 0.240 0.417 0.150 0.375 0.375 -1.000 0.271 0.561 0.419
2025-03-01 CNOS(FastSAM) RGB 0.279 0.414 0.311 -1.000 0.234 0.370 0.145 0.373 0.373 -1.000 0.268 0.558 0.184
2025-10-02 FSOD (IPD, 10 sample, conf 0.115) RGB 0.272 0.459 0.272 -1.000 0.287 0.243 0.148 0.525 0.551 -1.000 0.466 0.628 12.172
2025-08-26 MUSE (muse_full) RGB 0.216 0.282 0.250 -1.000 0.206 0.244 0.095 0.241 0.241 -1.000 0.221 0.286 0.987
2025-03-01 CNOS(SAM) RGB 0.208 0.306 0.239 -1.000 0.149 0.301 0.112 0.265 0.265 -1.000 0.172 0.419 1.081

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

Latest submissions to this leaderboard: