Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Enhancement] Add YOLOv8 OBB Models Directly to Zoo and Document #4238

Merged
merged 2 commits into from
Apr 6, 2024
Merged

[Enhancement] Add YOLOv8 OBB Models Directly to Zoo and Document #4238

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
da43822
Add YOLO OBB models to zoo
jacobmarks Apr 5, 2024
04344ae
document YOLO OBB inference
jacobmarks Apr 5, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
50 changes: 50 additions & 0 deletions docs/source/integrations/ultralytics.rst
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -241,6 +241,56 @@ manually convert Ultralytics predictions to :ref:`FiftyOne format <keypoints>`:
:alt: ultralytics-keypoints
:align: center


.. _ultralytics-oriented-bounding-boxes:

Oriented bounding boxes
-----------------------

You can directly pass Ultralytics YOLO oriented bounding box models to
:meth:`apply_model() <fiftyone.core.collections.SampleCollection.apply_model>`:

.. code-block:: python
:linenos:

model = YOLO("yolov8n-obb.pt")
# model = YOLO("yolov8s-obb.pt")
# model = YOLO("yolov8m-obb.pt")
# model = YOLO("yolov8l-obb.pt")
# model = YOLO("yolov8x-obb.pt")

dataset.apply_model(model, label_field="oriented_boxes")

session = fo.launch_app(dataset)


You can also load YOLOv8 oriented bounding box models from the
:ref:`FiftyOne Model Zoo <model-zoo>`:

.. code-block:: python
:linenos:

model_name = "yolov8n-obb-dotav1-torch"
# model_name = "yolov8s-obb-dotav1-torch"
# model_name = "yolov8m-obb-dotav1-torch"
# model_name = "yolov8l-obb-dotav1-torch"
# model_name = "yolov8x-obb-dotav1-torch"

model = foz.load_zoo_model(model_name)

dataset.apply_model(model, label_field="oriented_boxes")

session = fo.launch_app(dataset)


.. note::

The oriented bounding box models are trained on the `DOTA dataset
<https://captain-whu.github.io/DOTA/index.html>`_, which consists of
drone images with oriented bounding boxes. The models are trained to
predict on bird's eye view images, so applying them to regular images
may not yield good results.

.. _ultralytics-open-vocabulary-object-detection:

Open vocabulary detection
Expand Down
4 changes: 2 additions & 2 deletions fiftyone/utils/ultralytics.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -437,7 +437,7 @@ def predict_all(self, args):
return self._format_predictions(predictions)


class FiftyOneYOLOOBBConfig(FiftyOneYOLOModelConfig):
class FiftyOneYOLOOBBModelConfig(FiftyOneYOLOModelConfig):
pass


Expand Down Expand Up @@ -503,7 +503,7 @@ def _convert_yolo_detection_model(model):


def _convert_yolo_obb_model(model):
config = FiftyOneYOLOOBBConfig({"model": model})
config = FiftyOneYOLOOBBModelConfig({"model": model})
return FiftyOneYOLOOBBModel(config)


Expand Down
160 changes: 160 additions & 0 deletions fiftyone/zoo/models/manifest-torch.json
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2927,6 +2927,166 @@
"tags": ["detection", "torch", "yolo", "zero-shot"],
"date_added": "2024-03-11 19:22:51"
},
{
"base_name": "yolov8n-obb-dotav1-torch",
"base_filename": "yolov8n-obb.pt",
"description": "YOLOv8n Oriented Bounding Box model",
"source": "https://docs.ultralytics.com/tasks/obb/",
"size_bytes": 6548034,
"manager": {
"type": "fiftyone.core.models.ModelManager",
"config": {
"url": "https://github.com/ultralytics/assets/releases/download/v8.1.0/yolov8n-obb.pt"
}
},
"default_deployment_config_dict": {
"type": "fiftyone.utils.ultralytics.FiftyOneYOLOOBBModel",
"config": {}
},
"requirements": {
"packages": [
"torch>=1.7.0",
"torchvision>=0.8.1",
"ultralytics>=8.1.0"
],
"cpu": {
"support": true
},
"gpu": {
"support": true
}
},
"tags": ["detection", "torch", "yolo", "polylines", "obb"],
"date_added": "2024-04-05 19:22:51"
},
{
"base_name": "yolov8s-obb-dotav1-torch",
"base_filename": "yolov8s-obb.pt",
"description": "YOLOv8s Oriented Bounding Box model",
"source": "https://docs.ultralytics.com/tasks/obb/",
"size_bytes": 23245186,
"manager": {
"type": "fiftyone.core.models.ModelManager",
"config": {
"url": "https://github.com/ultralytics/assets/releases/download/v8.1.0/yolov8s-obb.pt"
}
},
"default_deployment_config_dict": {
"type": "fiftyone.utils.ultralytics.FiftyOneYOLOOBBModel",
"config": {}
},
"requirements": {
"packages": [
"torch>=1.7.0",
"torchvision>=0.8.1",
"ultralytics>=8.1.0"
],
"cpu": {
"support": true
},
"gpu": {
"support": true
}
},
"tags": ["detection", "torch", "yolo", "polylines", "obb"],
"date_added": "2024-04-05 19:22:51"
},
{
"base_name": "yolov8m-obb-dotav1-torch",
"base_filename": "yolov8m-obb.pt",
"description": "YOLOv8m Oriented Bounding Box model",
"source": "https://docs.ultralytics.com/tasks/obb/",
"size_bytes": 53304682,
"manager": {
"type": "fiftyone.core.models.ModelManager",
"config": {
"url": "https://github.com/ultralytics/assets/releases/download/v8.1.0/yolov8m-obb.pt"
}
},
"default_deployment_config_dict": {
"type": "fiftyone.utils.ultralytics.FiftyOneYOLOOBBModel",
"config": {}
},
"requirements": {
"packages": [
"torch>=1.7.0",
"torchvision>=0.8.1",
"ultralytics>=8.1.0"
],
"cpu": {
"support": true
},
"gpu": {
"support": true
}
},
"tags": ["detection", "torch", "yolo", "polylines", "obb"],
"date_added": "2024-04-05 19:22:51"
},
{
"base_name": "yolov8l-obb-dotav1-torch",
"base_filename": "yolov8l-obb.pt",
"description": "YOLOv8l Oriented Bounding Box model",
"source": "https://docs.ultralytics.com/tasks/obb/",
"size_bytes": 89504274,
"manager": {
"type": "fiftyone.core.models.ModelManager",
"config": {
"url": "https://github.com/ultralytics/assets/releases/download/v8.1.0/yolov8l-obb.pt"
}
},
"default_deployment_config_dict": {
"type": "fiftyone.utils.ultralytics.FiftyOneYOLOOBBModel",
"config": {}
},
"requirements": {
"packages": [
"torch>=1.7.0",
"torchvision>=0.8.1",
"ultralytics>=8.1.0"
],
"cpu": {
"support": true
},
"gpu": {
"support": true
}
},
"tags": ["detection", "torch", "yolo", "polylines", "obb"],
"date_added": "2024-04-05 19:22:51"
},
{
"base_name": "yolov8x-obb-dotav1-torch",
"base_filename": "yolov8x-obb.pt",
"description": "YOLOv8x Oriented Bounding Box model",
"source": "https://docs.ultralytics.com/tasks/obb/",
"size_bytes": 139533138,
"manager": {
"type": "fiftyone.core.models.ModelManager",
"config": {
"url": "https://github.com/ultralytics/assets/releases/download/v8.1.0/yolov8x-obb.pt"
}
},
"default_deployment_config_dict": {
"type": "fiftyone.utils.ultralytics.FiftyOneYOLOOBBModel",
"config": {}
},
"requirements": {
"packages": [
"torch>=1.7.0",
"torchvision>=0.8.1",
"ultralytics>=8.1.0"
],
"cpu": {
"support": true
},
"gpu": {
"support": true
}
},
"tags": ["detection", "torch", "yolo", "polylines", "obb"],
"date_added": "2024-04-05 19:22:51"
},
{
"base_name": "yolo-nas-torch",
"base_filename": null,
Expand Down
Loading