Submission: SAM6D/TUD-L
Download submission
| Submission name |
|
| Submission time (UTC) |
Dec. 5, 2023, 7:46 a.m.
|
| User |
jiehonglin
|
| Task |
Model-based 2D segmentation of unseen objects |
| Dataset |
TUD-L |
| Description |
|
| Evaluation scores |
| AP: | 0.569 |
| AP50: | 0.885 |
| AP75: | 0.674 |
| AP_large: | 0.526 |
| AP_medium: | 0.567 |
| AP_small: | 0.647 |
| AR1: | 0.629 |
| AR10: | 0.649 |
| AR100: | 0.650 |
| AR_large: | 0.612 |
| AR_medium: | 0.640 |
| AR_small: | 0.750 |
| average_time_per_image: | 2.393 |
|
| User |
jiehonglin
|
| Publication |
SAM-6D: Segment Anything Model Meets Zero-Shot 6D Object Pose Estimation |
| Implementation |
https://github.com/JiehongLin/SAM-6D/ |
| Training image modalities |
RGB-D |
| Test image modalities |
RGB-D |
| Description |
|
| Computer specifications |
GeForce RTX 3090 24G； Inter (R) Xeon (R) CPU E5-2678 v3 @ 2.50Ghz |