Submission: FFB6D-PBR+Real/LM-O/submission1

Download submission
Submission name submission1
Submission time (UTC) March 9, 2023, 5:52 a.m.
User CW_FLOYD
Task Model-based 6D localization of seen objects
Dataset LM-O
Training model type Default
Training image type Synthetic (only PBR images provided for BOP Challenge 2020 were used)
Description We trained FFB6D model with PBR and real dataset.
Evaluation scores
AR:0.677
AR_MSPD:0.776
AR_MSSD:0.723
AR_VSD:0.533
average_time_per_image:0.057

Method: FFB6D-PBR+Real

User CW_FLOYD
Publication ffb6d: a full flow bidirectional fusion network for 6d pose estimation, he et al., CVPR, 2021
Implementation
Training image modalities RGB-D
Test image modalities RGB-D
Description
Computer specifications NVIDIA RTX A4000