Submission: Extended FCOS-PBR/TUD-L/v2
Download submission
| Submission name |
v2
|
| Submission time (UTC) |
July 21, 2022, 7:15 a.m.
|
| User |
Yang-hai
|
| Task |
Model-based 2D detection of seen objects |
| Dataset |
TUD-L |
| Training model type |
None |
| Training image type |
Synthetic (only PBR images provided for BOP Challenge 2020 were used) |
| Description |
|
| Evaluation scores |
| AP: | 0.663 |
| AP50: | 0.905 |
| AP75: | 0.804 |
| AP_large: | 0.574 |
| AP_medium: | 0.678 |
| AP_small: | 0.518 |
| AR1: | 0.713 |
| AR10: | 0.760 |
| AR100: | 0.762 |
| AR_large: | 0.712 |
| AR_medium: | 0.767 |
| AR_small: | 0.600 |
| average_time_per_image: | -1.000 |
|
| User |
Yang-hai
|
| Publication |
Not yet |
| Implementation |
mmdetection |
| Training image modalities |
RGB |
| Test image modalities |
RGB |
| Description |
|
| Computer specifications |
an NVIDIA 3090 |