| Submission name | submission1 | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Submission time (UTC) | June 29, 2023, 8 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 | Based on FFB6D model, we constrained ROI of RGBD images with YOLOv5 detection model. Also add skip connection from encoder to decoder of FFB6D model. For multi-level fusion with skip connection, we use channel wise attention to filter meaningful feature. | ||||||||||
| Evaluation scores |
|
| User | CW_FLOYD |
|---|---|
| Publication | Not yet |
| Implementation | |
| Training image modalities | RGB-D |
| Test image modalities | RGB-D |
| Description | Based on FFB6D model, we constrained ROI of RGBD images with YOLOv5 detection model. Also add skip connection from encoder to decoder of FFB6D model. For multi-level fusion with skip connection, we use channel wise attention to filter meaningful feature. |
| Computer specifications | NVIDIA RTX A4000 |