Ryan Pfister
commited on
Commit
·
bf859a1
1
Parent(s):
58343f2
Update README.md with additional information
Browse files
README.md
CHANGED
|
@@ -48,16 +48,16 @@ The model was trained on a filtered version of the COCO dataset containing only
|
|
| 48 |
|
| 49 |
The model achieves the following metrics on the COCO person validation set:
|
| 50 |
|
| 51 |
-
| Metric
|
| 52 |
-
|
|
| 53 |
-
| Box mAP50-95 (COCO)
|
| 54 |
-
| Box mAP50 (COCO)
|
| 55 |
-
| Mask mAP50-95
|
| 56 |
-
| Mask mAP50
|
| 57 |
-
| Box Precision
|
| 58 |
-
| Box Recall
|
| 59 |
-
| Mask Precision
|
| 60 |
-
| Mask Recall
|
| 61 |
|
| 62 |
These metrics were computed on a validation set of 5,000 images with 10,777 instances.
|
| 63 |
|
|
@@ -73,7 +73,7 @@ This model is ideal for applications requiring precise person segmentation:
|
|
| 73 |
|
| 74 |
## Usage
|
| 75 |
|
| 76 |
-
The model can be used directly with Ultralytics
|
| 77 |
|
| 78 |
```python
|
| 79 |
from ultralytics import YOLO
|
|
@@ -136,4 +136,4 @@ if result.masks is not None:
|
|
| 136 |
|
| 137 |
## License
|
| 138 |
|
| 139 |
-
This model is available under the Apache 2.0 license.
|
|
|
|
| 48 |
|
| 49 |
The model achieves the following metrics on the COCO person validation set:
|
| 50 |
|
| 51 |
+
| Metric | Value |
|
| 52 |
+
| ------------------- | ----- |
|
| 53 |
+
| Box mAP50-95 (COCO) | 0.628 |
|
| 54 |
+
| Box mAP50 (COCO) | 0.840 |
|
| 55 |
+
| Mask mAP50-95 | 0.524 |
|
| 56 |
+
| Mask mAP50 | 0.821 |
|
| 57 |
+
| Box Precision | 0.835 |
|
| 58 |
+
| Box Recall | 0.745 |
|
| 59 |
+
| Mask Precision | 0.843 |
|
| 60 |
+
| Mask Recall | 0.723 |
|
| 61 |
|
| 62 |
These metrics were computed on a validation set of 5,000 images with 10,777 instances.
|
| 63 |
|
|
|
|
| 73 |
|
| 74 |
## Usage
|
| 75 |
|
| 76 |
+
The model can be used directly with Ultralytics YOLO:
|
| 77 |
|
| 78 |
```python
|
| 79 |
from ultralytics import YOLO
|
|
|
|
| 136 |
|
| 137 |
## License
|
| 138 |
|
| 139 |
+
This model is available under the Apache 2.0 license.
|