Model-based 6D detection of unseen objects – TUD-L

This leaderbord shows the ranking for Model-based 6D detection of unseen objects on TUD-L. The reported time is the average time to generate predictions for all objects in a single image (averaged over the datasets).

Date (UTC) Submission Views Test image AP APMSSD APMSPD Time (s)
2024-11-29 FreeZeV2.1 single RGB-D 0.991 0.989 0.993 13.850
2025-05-31 FreeZeV2.2 Single RGB-D 0.988 0.986 0.991 18.568
2025-05-23 FRTPose-WAPR(Default Detection) Single RGB-D 0.978 0.982 0.973 15.310
2025-05-26 FRTPose-WAPR(Multi 2D Detections) Single RGB-D 0.976 0.980 0.972 51.722
2024-09-19 FreeZeV2 (SAM6D) (SAR threshold = -1) single RGB-D 0.975 0.973 0.977 26.683
2025-05-22 FRTPose-WAPR(MUSE) (conf-0.45) Single RGB-D 0.963 0.966 0.959 0.463
2024-11-02 FRTPose.v1 (Default Detections) single RGB-D 0.946 0.950 0.943 31.950
2024-11-02 FRTPose.v1 (SAM6D-FastSAM) single RGB-D 0.942 0.945 0.939 27.618
2024-11-24 FRTPose.v1 (SAM6D-FastSAM) single RGB-D 0.941 0.944 0.938 24.596
2024-11-24 FRTPose.v1 (Default Detections) single RGB-D 0.941 0.945 0.937 33.027
2024-11-24 FRTPose.v1 (MUSE) single RGB-D 0.940 0.942 0.937 32.447
2024-12-09 FRTPose.v1 (SAM6D-FastSAM, NIDS, CNOS and MUSE) single RGB-D 0.935 0.937 0.933 44.533
2024-09-26 FreeZeV2 (SAM6D, Coarse-to-Fine) (0.45 to 0.25) single RGB-D 0.902 0.900 0.904 4.579
2024-09-26 FreeZeV2 (SAM6D, Coarse-to-Fine) single RGB-D 0.884 0.882 0.886 4.569
2024-11-28 Co-op (F3DT2D, Coarse, RGBD) (sel rescoring, Intel(R) Core(TM) i9-14900K, RTX4090) Single RGB-D 0.841 0.842 0.839 0.892
2024-11-28 Co-op (CNOS, Coarse, RGBD) (BOP 2024) single RGB-D 0.808 0.808 0.807 1.491
2024-11-28 Co-op (CNOS, 1 Hypo, RGBD) (BOP 2024) Single RGB-D 0.769 0.768 0.769 4.615
2024-11-28 Co-op (F3DT2D, 5 Hypo, RGBD) (Intel(R) Core(TM) i9-14900K, RTX4090) Single RGB-D 0.766 0.766 0.765 10.342
2024-09-17 FreeZeV2 (SAM6D) single RGB-D 0.751 0.748 0.754 2.779
2024-11-29 Co-op (CNOS, 5 Hypo, RGBD) (BOP 2024) Single RGB-D 0.738 0.739 0.737 4.443
2024-09-13 GigaPose+GenFlow (RGBD) (threshold = 0.0) Single RGB-D 0.715 0.714 0.717 2.626
2024-09-20 GigaPose+GenFlow (5 hypothesis) single RGB 0.688 0.592 0.783 6.910
2024-11-28 Co-op (F3DT2D, 5 Hypo) (Intel(R) Core(TM) i9-14900K, RTX4090) Single RGB 0.658 0.573 0.743 9.561
2024-11-28 Co-op (CNOS, 1 Hypo) (BOP 2024) Single RGB 0.641 0.555 0.727 4.358
2024-11-28 Co-op (CNOS, Coarse) (BOP 2024) Single RGB 0.640 0.534 0.746 1.510
2024-11-29 Co-op (CNOS, 5 Hypo) (BOP 2024) Single RGB 0.619 0.536 0.702 6.601
2024-09-27 GigaPose + GenFlow (threshold = 0.6) Single RGB 0.608 0.526 0.690 2.703
2024-09-18 GigaPose+GenFlow+kabsch (5 hypothesis) single RGB-D 0.246 0.247 0.244 7.687
2024-09-18 GigaPose-CVPR24 single RGB 0.035 0.017 0.052 0.459

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