Submission: DPODv2 (synthetic train data, RGB + D Kabsch)/T-LESS/RGB YOLO, RGB DPOD, Kabsch, ICP

Download submission
Submission name RGB YOLO, RGB DPOD, Kabsch, ICP
Submission time (UTC) July 21, 2022, 4:19 p.m.
User ivan_shugurov
Task 6D localization of seen objects
Dataset T-LESS
Training model type CAD
Training image type Synthetic (only PBR images provided for BOP Challenge 2020 were used)
Description
Evaluation scores
AR:0.590
AR_MSPD:0.652
AR_MSSD:0.632
AR_VSD:0.485
average_time_per_image:0.523

Method: DPODv2 (synthetic train data, RGB + D Kabsch)

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

Inference completely in RGB, then the correspondences are projected in 3D

Computer specifications