Method: Koenig-Hybrid-DL-PointPairs

User rko
Publication
Implementation
Training image modalities RGB
Test image modalities RGB-D
Description

A hybrid model that first finds object instances using either RetinaMask [1] or MaskRCNN [2], whichever has the better mAP on the validation set. For each instance the point-pair voting of [3] is used to recover the pose. For datasets with textured models, symmetries are resolved using the descriptors of [4]. For [3], we use the implementation in MVTec HALCON 20.05.

[1] Fu, C. Y., Shvets, M., & Berg, A. C. (2019). RetinaMask: Learning to predict masks improves state-of-the-art single-shot detection for free. arXiv preprint arXiv:1901.03353. [2] He, K., Gkioxari, G., Dollár, P., & Girshick, R.: Mask r-cnn. ICCV 2017. [3] Drost, B., Ulrich, M., Navab, N., Ilic, S.: Model globally, match locally: Efficient and robust 3d object recognition. CVPR 2010. [4] Lepetit, V., Fua, P.: Keypoint recognition using randomized trees. T-PAMI 2006.

Computer specifications Nvidia GeForce GTX 1080 Ti; Max. 12 Threads on Intel(R) Xeon(R) Silver 4114 CPU @ 2.20GHz

Public submissions

Date Submission name Dataset
2020-08-19 08:23 - T-LESS
2020-08-19 08:24 - TUD-L
2020-08-19 08:24 - LM-O
2020-08-19 08:25 - ITODD
2020-08-19 08:26 - HB
2020-08-19 08:28 - YCB-V
2020-08-19 10:19 - IC-BIN