Type: huggingface.object_detection.ObjectDetection
Namespace: huggingface.object_detection
Description
Detects and localizes objects in images. image, object detection, bounding boxes, huggingface
Use cases:
- Identify and count objects in images
- Locate specific items in complex scenes
- Assist in autonomous vehicle vision systems
- Enhance security camera footage analysis
Properties
| Property | Type | Description | Default |
|---|---|---|---|
| model | hf.object_detection |
The model ID to use for object detection | {'type': 'hf.object_detection', 'repo_id': 'facebook/detr-resnet-50', 'path': None, 'variant': None, 'allow_patterns': None, 'ignore_patterns': None} |
| image | image |
The input image for object detection | {'type': 'image', 'uri': '', 'asset_id': None, 'data': None} |
| threshold | float |
Minimum confidence score for detected objects | 0.9 |
| top_k | int |
The number of top predictions to return | 5 |
Outputs
| Output | Type | Description |
|---|---|---|
| output | List[object_detection_result] |
Metadata
Related Nodes
Browse other nodes in the huggingface.object_detection namespace.