Generic Detector
Detect objects based on the similarity of text prompts and the images of objects in the frame.
Overview
The Generic Detector node is designed to detect objects within a video frame based on the similarity of text prompts and the images of objects. This functionality is useful for applications requiring identification and categorization of objects in real-time.
Inputs & Outputs
- Inputs: 1, Media Format: Raw Video
- Outputs: 1, Media Format: Raw Video
- Output Metadata: Objects
Properties
Property | Description | Type | Default | Required |
---|---|---|---|---|
model_id | The type of model to use. Options: Yolo-Small (yolov8s-world), Yolo-Medium (yolov8m-world), Yolo-Large (yolov8l-world), OWL-Medium (owlvit-base-patch32). | enum | yolov8s-world | Yes |
class_list | A comma-separated list of objects to detect, optionally with an alternate label. For example, a person=person, a blue car=blue_car. | string | null | Yes |
interval | The inference interval. Infer on every nth frame. 1 means infer every frame. The minimum value is 1 frame. | number | 1 | Yes |
confidence_threshold | This property allows you to override the default minimum inference threshold for all classes. The minimum value is 0, the maximum value is 1.0, the step is 0.1, and the scale is 0.1. | slider-optional | 0.1 | No |
per_class_thresholds | A comma-separated list of per-class thresholds. Leave this field empty to use the default threshold for all classes. For example, 0.1,0.2,0.3. | string | null | No |
iou_threshold | This property allows you to increase the threshold to reduce potential duplicate detections of a single object. The minimum value is 0, the maximum value is 1.0, the step is 0.1, and the scale is 0.1. | slider-optional | 0.5 | No |
min_object_size | The minimum size of the object to detect. The size should be specified as width x height. | string | null | No |
enable_max_optimizations | Enable advanced optimizations to improve performance. This currently increases deployment start time. | bool | false | No |
clear_cache | Set to true to clear model cache. This will increase deployment start time. | bool | false | No |
Metadata
The output metadata includes information about the detected objects within the frame. This includes object types, confidence scores, and bounding box coordinates.
Example JSON
Updated about 1 month ago