Submission: FFB6D-CVPR21-PBR-NoRefinement/YCB-V

Submission name
Submission time (UTC) Dec. 27, 2021, 4:25 a.m.
User YishengHe
Dataset YCB-V
Training model type Default
Training image type Synthetic (provided)
Description We use only provided synthesis images for training. No iterative refinement algorithm is applied, i.e., ICP. Differences from the original implementation: - We regenerate object SIFT-FPS 3D keypoint for each object from the BOP YCBV object mesh models. Because the object coordinate in the coordinate system of BOP benchmark dataset is different from the original dataset.
Evaluation scores
bop19_average_recall:0.758
bop19_average_recall_mspd:0.740
bop19_average_recall_mssd:0.827
bop19_average_recall_vsd:0.706
bop19_average_time_per_image:0.199

Method: FFB6D-CVPR21-PBR-NoRefinement

User YishengHe
Publication Yisheng He et al.: FFB6D: A Full Flow Bidirectional Fusion Network for 6D Pose Estimation, CVPR 2021 (Oral)
Implementation https://github.com/ethnhe/FFB6D
Training image modalities RGB-D
Test image modalities RGB-D
Description

FFB6D is a full flow bidirectional fusion network designed for 6D pose estimation from a single RGBD image. We only use the PBR data for training. The predicted result is not refined by any iterative refinement algorithms, i.e., ICP.

Computer specifications CPU: Intel(R) Xeon(R) Gold 6130 CPU @ 2.10GHz; GPU: GeForce RTX 2080Ti