Pose estimation (BOP 2019-2022) – HomebrewedDB

The performance scores are defined in the BOP Challenge 2019 description. The reported time is the average image processing time.

Date (UTC) Submission Test image AR ARVSD ARMSSD ARMSPD Time (s)
2021-12-22 SurfEmb-PBR-RGBD RGB-D 0.866 0.829 0.875 0.893 7.431
2022-07-21 DPODv2 (synthetic train data, RGB + D Kabsch) (RGB YOLO, RGB DPOD, Kabsch, ICP) RGB-D 0.845 0.784 0.872 0.879 0.329
2022-07-21 DPODv2 (synthetic train data, RGB + D Kabsch) (RGB YOLO, RGB DPOD, Kabsch) RGB-D 0.840 0.796 0.856 0.867 0.238
2021-12-22 SurfEmb-PBR-RGB RGB 0.791 0.725 0.760 0.888 7.262
2022-04-24 Coupled Iterative Refinement RGB-D 0.757 0.760 0.753 0.757 -1.000
2020-08-18 CosyPose-ECCV20-SYNT+REAL-8VIEWS RGB 0.746 0.691 0.744 0.804 0.427
2020-08-18 CDPNv2_BOP20 (RGB-only) (Zhigang-CDPNv2 (MODE 1, FCOS)) RGB 0.722 0.614 0.708 0.845 0.273
2020-08-19 CDPNv2_BOP20 (PBR-only & RGB-only) (Zhigang-CDPNv2 (MODE 2, FCOS)) RGB 0.722 0.614 0.708 0.845 0.273
2022-07-21 DPODv2 (synthetic train data, RGB) (RGB YOLO, RGB DPOD, PnP) RGB 0.714 0.627 0.692 0.823 0.163
2022-07-10 PFA-cosypose (7view-Bidir) RGB 0.712 0.693 0.704 0.740 -1.000
2020-08-17 CosyPose-ECCV20-SYNT+REAL-1VIEW-ICP RGB-D 0.712 0.679 0.719 0.737 5.326
2020-08-19 CDPNv2_BOP20 (PBR-only & ICP) (icpv4) RGB-D 0.712 0.629 0.757 0.749 0.713
2020-08-19 CDPNv2_BOP20 (RGB-only & ICP) (icpv4) RGB-D 0.712 0.629 0.757 0.749 0.713
2019-10-22 Vidal-Sensors18 D 0.706 0.707 0.704 0.708 2.608
2022-07-10 PFA-cosypose (1view-Bidir) RGB 0.704 0.682 0.694 0.737 -1.000
2020-08-18 CosyPose-ECCV20-SYNT+REAL-4VIEWS RGB 0.696 0.646 0.685 0.756 0.445
2020-08-17 Pix2Pose-BOP20_w/ICP-ICCV19 RGB-D 0.695 0.640 0.721 0.724 3.248
2020-08-17 CDPN_BOP19 (RGB-only) RGB 0.672 0.557 0.642 0.816 0.280
2019-08-12 Drost-CVPR10-Edges RGB-D 0.671 0.677 0.665 0.670 144.029
2020-08-19 leaping from 2D to 6D (hb) RGB 0.658 0.528 0.583 0.863 0.736
2020-08-18 CosyPose-ECCV20-SYNT+REAL-1VIEW RGB 0.656 0.613 0.634 0.721 0.419
2020-08-18 CosyPose-ECCV20-PBR-1VIEW RGB 0.656 0.613 0.634 0.721 0.417
2020-08-19 Koenig-Hybrid-DL-PointPairs RGB-D 0.651 0.660 0.645 0.649 0.274
2019-08-05 Drost-CVPR10-3D-Edges D 0.623 0.618 0.624 0.626 127.372
2019-07-31 Drost-CVPR10-3D-Only D 0.615 0.593 0.627 0.627 16.136
2019-10-17 Drost-CVPR10-3D-Only-Faster D 0.603 0.585 0.612 0.611 1.659
2022-07-22 OSOP: RGBD (no training on target objects, Kabsch + ICP) RGB-D 0.581 0.571 0.584 0.588 0.438
2020-08-19 EPOS-BOP20-PBR RGB 0.580 0.484 0.527 0.729 0.657
2022-07-22 OSOP: RGBD (no training on target objects, Kabsch) RGB-D 0.560 0.556 0.559 0.566 0.314
2020-08-19 PointVoteNet2 (HB-PVN2-E6) RGB-D 0.556 0.578 0.545 0.546 -1.000
2019-10-14 FĂ©lix&Neves-ICRA2017-IET2019 RGB-D 0.529 0.544 0.520 0.523 58.693
2019-10-22 Sundermeyer-IJCV19+ICP RGB-D 0.506 0.479 0.506 0.533 1.352
2019-10-22 Zhigang-CDPN-ICCV19 (Zhigang-CDPN-ICCV19) RGB 0.470 0.391 0.451 0.569 0.311
2022-07-21 DPODv2 (synthetic train data, RGBD) (RGB YOLO, Depth DPOD, Kabsch, ICP) RGB-D 0.468 0.462 0.468 0.473 0.470
2022-07-22 OSOP: RGB (no training on target objects, PnP) RGB 0.464 0.420 0.436 0.535 0.503
2020-08-18 Pix2Pose-BOP20-ICCV19 RGB 0.446 0.352 0.394 0.594 0.982
2022-07-21 DPODv2 (synthetic train data, RGBD) (RGB YOLO, Depth DPOD, Kabsch) RGB-D 0.382 0.344 0.398 0.403 0.245
2019-10-22 Sundermeyer-IJCV19 RGB 0.346 0.273 0.306 0.461 0.190
2020-08-11 SingleMultiPathEncoder-CVPR20 (Test) RGB 0.293 0.241 0.250 0.388 0.191
2019-10-10 DPOD (synthetic) RGB 0.286 0.218 0.262 0.379 0.180
2022-09-26 ZTE_PPF D 0.000 0.000 0.000 0.000 1.018