Model-based 6D detection of unseen objects – ITODD

This leaderbord shows the ranking for Model-based 6D detection of unseen objects on ITODD. The reported time is the average image processing time.

Date (UTC) Submission Test image AP APMSSD APMSPD Time (s)
2024-11-02 FRTPose.v1 (SAM6D-FastSAM) RGB-D 0.767 0.787 0.747 9.643
2024-11-02 FRTPose.v1 (Default Detections) RGB-D 0.764 0.784 0.744 10.238
2024-09-17 Co-op (SAM6D, 5 Hypo, RGBD) (genflow rescoring(no meaning) / Intel(R) Core(TM) i9-14900K, RTX4090) RGB-D 0.620 0.629 0.610 10.822
2024-09-20 FreeZe.v2 (SAM6D) (SAR threshold = -1) RGB-D 0.617 0.629 0.606 53.206
2024-09-17 Co-op (CNOS, 5 Hypo, RGBD) (genflow rescoring(no meaning) / Intel(R) Core(TM) i9-14900K, RTX4090) RGB-D 0.601 0.610 0.593 10.855
2024-09-16 Co-op (SAM6D, Coarse, RGBD) (Coarse only, magsac++, det thres=0.4 / Intel(R) Core(TM) i9-14900K, RTX4090) RGB-D 0.578 0.592 0.565 0.332
2024-09-26 FreeZe.v2 (SAM6D, Coarse-to-Fine) (0.45 to 0.25) RGB-D 0.573 0.584 0.561 19.535
2024-09-26 FreeZe.v2 (SAM6D, Coarse-to-Fine) RGB-D 0.539 0.551 0.528 19.597
2024-09-18 GigaPose+GenFlow+kabsch (5 hypothesis) RGB-D 0.525 0.533 0.517 9.714
2024-09-28 Co-op (SAM6D, 5 Hypo) (genflow coarse score recalc / Intel(R) Core(TM) i9-14900K, RTX4090) RGB 0.497 0.438 0.556 7.075
2024-09-28 Co-op (CNOS, 5 Hypo) (genflow coarse score recalc / Intel(R) Core(TM) i9-14900K, RTX4090) RGB 0.476 0.417 0.535 7.213
2024-09-18 FreeZe.v2 (SAM6D) RGB-D 0.474 0.487 0.462 13.370
2024-09-13 GigaPose+GenFlow (1 hypothesis) (RGBD) (threshold = 0.0) RGB-D 0.447 0.456 0.438 2.748
2024-09-20 GigaPose+GenFlow (5 hypothesis) RGB 0.425 0.356 0.494 8.954
2024-09-27 GigaPose+GenFlow (1 hypothesis) (threshold = 0.2) RGB 0.367 0.315 0.418 2.823
2024-09-15 Co-op (CNOS, Coarse) (Coarse only / Intel(R) Core(TM) i9-14900K, RTX4090) RGB 0.362 0.295 0.430 0.491
2024-09-18 GigaPose-CVPR24 RGB 0.149 0.075 0.223 0.329

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