Submission: Extended FCOS-MixPBR/TUD-L/v2
Download submission
| Submission name |
v2
|
| Submission time (UTC) |
July 23, 2022, 6:29 a.m.
|
| User |
Yang-hai
|
| Task |
Model-based 2D detection of seen objects |
| Dataset |
TUD-L |
| Training model type |
None |
| Training image type |
Synthetic + real |
| Description |
|
| Evaluation scores |
| AP: | 0.836 |
| AP50: | 0.988 |
| AP75: | 0.960 |
| AP_large: | 0.890 |
| AP_medium: | 0.834 |
| AP_small: | 0.776 |
| AR1: | 0.862 |
| AR10: | 0.867 |
| AR100: | 0.867 |
| AR_large: | 0.944 |
| AR_medium: | 0.855 |
| AR_small: | 0.775 |
| average_time_per_image: | 0.033 |
|
| User |
Yang-hai
|
| Publication |
Not yet |
| Implementation |
|
| Training image modalities |
RGB |
| Test image modalities |
RGB |
| Description |
|
| Computer specifications |
NVIDIA 3090 |