Method: GDRNPP-PBRReal-RGBD-SModel

User zyMeteroid
Publication Not yet
Implementation Pytorch, code can be found at https://github.com/shanice-l/gdrnpp_bop2022
Training image modalities RGB
Test image modalities RGB-D
Description

Submitted to: BOP Challenge 2023

Training data: real + provided PBR

Used 3D models: Reconstructed for T-LESS, default for other datasets

Notes:

Authors: Xingyu Liu, Ruida Zhang, Chenyangguang Zhang, Bowen Fu, Jiwen Tang, Xiquan Liang, Jingyi Tang, Xiaotian Cheng, Yukang Zhang, Gu Wang, and Xiangyang Ji.

Based on GDRNPP-PBRReal-RGB-SModel, we utilize depth information to further refine the estimated pose. We implement a fast refinement module without learned parameters. We compare the rendered object depth and the observed depth to refine translation.

Computer specifications GPU RTX 3090; CPU AMD EPYC 7H12 64-Core Processor.

Public submissions

Date Submission name Dataset
2022-10-12 09:48 - T-LESS
2022-10-12 09:49 - TUD-L
2022-10-12 09:49 - LM-O
2022-10-12 09:50 - IC-BIN
2022-10-12 09:50 - YCB-V
2022-10-13 02:26 - HB
2022-10-14 08:19 - ITODD