Keypoint Detection Model

v2

Class: RoboflowKeypointDetectionModelBlockV2 (there are multiple versions of this block)

Source: inference.core.workflows.core_steps.models.roboflow.keypoint_detection.v2.RoboflowKeypointDetectionModelBlockV2

Warning: This block has multiple versions. Please refer to the specific version for details. You can learn more about how versions work here: Versioning

Run inference on a keypoint detection model hosted on or uploaded to Roboflow.

You can query any model that is private to your account, or any public model available on Roboflow Universe.

You will need to set your Roboflow API key in your Inference environment to use this block. To learn more about setting your Roboflow API key, refer to the Inference documentation.

Type identifier

Use the following identifier in step "type" field: roboflow_core/roboflow_keypoint_detection_model@v2to add the block as as step in your workflow.

Properties

Name	Type	Description	Refs
name	str	Enter a unique identifier for this step..	❌
model_id	str	Roboflow model identifier..	✅
confidence	float	Confidence threshold for predictions..	✅
keypoint_confidence	float	Confidence threshold to predict a keypoint as visible..	✅
class_filter	List[str]	List of accepted classes. Classes must exist in the model's training set..	✅
iou_threshold	float	Minimum overlap threshold between boxes to combine them into a single detection, used in NMS. Learn more..	✅
max_detections	int	Maximum number of detections to return..	✅
class_agnostic_nms	bool	Boolean flag to specify if NMS is to be used in class-agnostic mode..	✅
max_candidates	int	Maximum number of candidates as NMS input to be taken into account..	✅
disable_active_learning	bool	Boolean flag to disable project-level active learning for this block..	✅
active_learning_target_dataset	str	Target dataset for active learning, if enabled..	✅

The Refs column marks possibility to parametrise the property with dynamic values available in workflow runtime. See Bindings for more info.

Available Connections

Compatible Blocks

Check what blocks you can connect to Keypoint Detection Model in version v2.

• inputs: Image Convert Grayscale, Anthropic Claude, OpenAI, Florence-2 Model, Distance Measurement, Pixelate Visualization, Dimension Collapse, SIFT, VLM as Detector, Stability AI Image Generation, SIFT Comparison, Mask Visualization, Image Slicer, Triangle Visualization, VLM as Detector, Polygon Zone Visualization, Model Comparison Visualization, Crop Visualization, Classification Label Visualization, Keypoint Detection Model, Reference Path Visualization, Line Counter, Twilio SMS Notification, Google Gemini, Bounding Box Visualization, Image Contours, Roboflow Custom Metadata, Size Measurement, Circle Visualization, Perspective Correction, Slack Notification, Pixel Color Count, Polygon Visualization, Instance Segmentation Model, VLM as Classifier, Trace Visualization, Webhook Sink, Color Visualization, Identify Changes, Detections Consensus, Image Threshold, SIFT Comparison, JSON Parser, Absolute Static Crop, Clip Comparison, Line Counter Visualization, Stitch Images, Multi-Label Classification Model, Line Counter, Dot Visualization, Identify Outliers, Dynamic Zone, Background Color Visualization, Florence-2 Model, Model Monitoring Inference Aggregator, Image Slicer, Template Matching, Roboflow Dataset Upload, Roboflow Dataset Upload, Image Blur, Email Notification, Camera Focus, Grid Visualization, Object Detection Model, Blur Visualization, Label Visualization, Stability AI Inpainting, Depth Estimation, Image Preprocessing, Llama 3.2 Vision, Ellipse Visualization, ONVIF Control, VLM as Classifier, Halo Visualization, Corner Visualization, Camera Calibration, Clip Comparison, Buffer, Dynamic Crop, Single-Label Classification Model, Relative Static Crop, OpenAI, Keypoint Visualization, Local File Sink
• outputs: Keypoint Detection Model, Detections Classes Replacement, Florence-2 Model, Detection Offset, SmolVLM2, Pixelate Visualization, Single-Label Classification Model, Qwen2.5-VL, Multi-Label Classification Model, Moondream2, Detections Filter, Dot Visualization, Object Detection Model, Instance Segmentation Model, Background Color Visualization, Florence-2 Model, Model Monitoring Inference Aggregator, Triangle Visualization, Roboflow Dataset Upload, Roboflow Dataset Upload, Object Detection Model, Model Comparison Visualization, Crop Visualization, Blur Visualization, Label Visualization, Segment Anything 2 Model, Byte Tracker, Keypoint Detection Model, Multi-Label Classification Model, Ellipse Visualization, Bounding Box Visualization, Roboflow Custom Metadata, Corner Visualization, Circle Visualization, Dynamic Crop, Detections Transformation, Single-Label Classification Model, Instance Segmentation Model, Detections Merge, Keypoint Visualization, Trace Visualization, Webhook Sink, Color Visualization, Detections Consensus

Input and Output Bindings

The available connections depend on its binding kinds. Check what binding kinds Keypoint Detection Model in version v2 has.

Bindings

• input

  • images (image): The image to infer on..
  • model_id (roboflow_model_id): Roboflow model identifier..
  • confidence (float_zero_to_one): Confidence threshold for predictions..
  • keypoint_confidence (float_zero_to_one): Confidence threshold to predict a keypoint as visible..
  • class_filter (list_of_values): List of accepted classes. Classes must exist in the model's training set..
  • iou_threshold (float_zero_to_one): Minimum overlap threshold between boxes to combine them into a single detection, used in NMS. Learn more..
  • max_detections (integer): Maximum number of detections to return..
  • class_agnostic_nms (boolean): Boolean flag to specify if NMS is to be used in class-agnostic mode..
  • max_candidates (integer): Maximum number of candidates as NMS input to be taken into account..
  • disable_active_learning (boolean): Boolean flag to disable project-level active learning for this block..
  • active_learning_target_dataset (roboflow_project): Target dataset for active learning, if enabled..

• output

  • inference_id (inference_id): Inference identifier.
  • predictions (keypoint_detection_prediction): Prediction with detected bounding boxes and detected keypoints in form of sv.Detections(...) object.
  • model_id (roboflow_model_id): Roboflow model id.

Example JSON definition of step Keypoint Detection Model in version v2

{
    "name": "<your_step_name_here>",
    "type": "roboflow_core/roboflow_keypoint_detection_model@v2",
    "images": "$inputs.image",
    "model_id": "my_project/3",
    "confidence": 0.3,
    "keypoint_confidence": 0.3,
    "class_filter": [
        "a",
        "b",
        "c"
    ],
    "iou_threshold": 0.4,
    "max_detections": 300,
    "class_agnostic_nms": true,
    "max_candidates": 3000,
    "disable_active_learning": true,
    "active_learning_target_dataset": "my_project"
}

v1

Class: RoboflowKeypointDetectionModelBlockV1 (there are multiple versions of this block)

Source: inference.core.workflows.core_steps.models.roboflow.keypoint_detection.v1.RoboflowKeypointDetectionModelBlockV1

Warning: This block has multiple versions. Please refer to the specific version for details. You can learn more about how versions work here: Versioning

Run inference on a keypoint detection model hosted on or uploaded to Roboflow.

You can query any model that is private to your account, or any public model available on Roboflow Universe.

You will need to set your Roboflow API key in your Inference environment to use this block. To learn more about setting your Roboflow API key, refer to the Inference documentation.

Type identifier

Use the following identifier in step "type" field: roboflow_core/roboflow_keypoint_detection_model@v1to add the block as as step in your workflow.

Properties

Name	Type	Description	Refs
name	str	Enter a unique identifier for this step..	❌
model_id	str	Roboflow model identifier..	✅
confidence	float	Confidence threshold for predictions..	✅
keypoint_confidence	float	Confidence threshold to predict a keypoint as visible..	✅
class_filter	List[str]	List of accepted classes. Classes must exist in the model's training set..	✅
iou_threshold	float	Minimum overlap threshold between boxes to combine them into a single detection, used in NMS. Learn more..	✅
max_detections	int	Maximum number of detections to return..	✅
class_agnostic_nms	bool	Boolean flag to specify if NMS is to be used in class-agnostic mode..	✅
max_candidates	int	Maximum number of candidates as NMS input to be taken into account..	✅
disable_active_learning	bool	Boolean flag to disable project-level active learning for this block..	✅
active_learning_target_dataset	str	Target dataset for active learning, if enabled..	✅

The Refs column marks possibility to parametrise the property with dynamic values available in workflow runtime. See Bindings for more info.

Available Connections

Compatible Blocks

Check what blocks you can connect to Keypoint Detection Model in version v1.

• inputs: Image Convert Grayscale, Anthropic Claude, OpenAI, Florence-2 Model, Distance Measurement, Pixelate Visualization, Dimension Collapse, SIFT, VLM as Detector, Stability AI Image Generation, SIFT Comparison, Mask Visualization, Image Slicer, Triangle Visualization, VLM as Detector, Polygon Zone Visualization, Model Comparison Visualization, Crop Visualization, Classification Label Visualization, Keypoint Detection Model, Reference Path Visualization, Line Counter, Twilio SMS Notification, Google Gemini, Bounding Box Visualization, Image Contours, Roboflow Custom Metadata, Size Measurement, Circle Visualization, Perspective Correction, Slack Notification, Pixel Color Count, Polygon Visualization, Instance Segmentation Model, VLM as Classifier, Trace Visualization, Webhook Sink, Color Visualization, Identify Changes, Detections Consensus, Image Threshold, SIFT Comparison, JSON Parser, Absolute Static Crop, Clip Comparison, Line Counter Visualization, Stitch Images, Multi-Label Classification Model, Line Counter, Dot Visualization, Identify Outliers, Dynamic Zone, Background Color Visualization, Florence-2 Model, Model Monitoring Inference Aggregator, Image Slicer, Template Matching, Roboflow Dataset Upload, Roboflow Dataset Upload, Image Blur, Email Notification, Camera Focus, Grid Visualization, Object Detection Model, Blur Visualization, Label Visualization, Stability AI Inpainting, Depth Estimation, Image Preprocessing, Llama 3.2 Vision, Ellipse Visualization, ONVIF Control, VLM as Classifier, Halo Visualization, Corner Visualization, Camera Calibration, Clip Comparison, Buffer, Dynamic Crop, Single-Label Classification Model, Relative Static Crop, OpenAI, Keypoint Visualization, Local File Sink
• outputs: CogVLM, Anthropic Claude, Cache Set, Google Vision OCR, OpenAI, Detections Classes Replacement, Florence-2 Model, Detection Offset, Distance Measurement, Pixelate Visualization, OpenAI, Stability AI Image Generation, YOLO-World Model, Mask Visualization, Triangle Visualization, Path Deviation, Polygon Zone Visualization, Model Comparison Visualization, Crop Visualization, LMM, Classification Label Visualization, Segment Anything 2 Model, Reference Path Visualization, Twilio SMS Notification, Line Counter, Google Gemini, Bounding Box Visualization, Roboflow Custom Metadata, Size Measurement, Circle Visualization, Perspective Correction, Slack Notification, Pixel Color Count, Polygon Visualization, Detections Transformation, Instance Segmentation Model, Trace Visualization, Detections Merge, Webhook Sink, Color Visualization, LMM For Classification, Detections Consensus, Image Threshold, Detections Stitch, SIFT Comparison, Line Counter Visualization, Line Counter, Dot Visualization, Detections Filter, Instance Segmentation Model, Florence-2 Model, Background Color Visualization, Model Monitoring Inference Aggregator, Roboflow Dataset Upload, Roboflow Dataset Upload, Image Blur, Time in Zone, Email Notification, Path Deviation, CLIP Embedding Model, Stability AI Inpainting, Label Visualization, Blur Visualization, Llama 3.2 Vision, Image Preprocessing, Byte Tracker, Ellipse Visualization, ONVIF Control, Halo Visualization, Corner Visualization, Clip Comparison, Cache Get, Time in Zone, Dynamic Crop, OpenAI, Keypoint Visualization, Local File Sink

Input and Output Bindings

The available connections depend on its binding kinds. Check what binding kinds Keypoint Detection Model in version v1 has.

Bindings

• input

  • images (image): The image to infer on..
  • model_id (roboflow_model_id): Roboflow model identifier..
  • confidence (float_zero_to_one): Confidence threshold for predictions..
  • keypoint_confidence (float_zero_to_one): Confidence threshold to predict a keypoint as visible..
  • class_filter (list_of_values): List of accepted classes. Classes must exist in the model's training set..
  • iou_threshold (float_zero_to_one): Minimum overlap threshold between boxes to combine them into a single detection, used in NMS. Learn more..
  • max_detections (integer): Maximum number of detections to return..
  • class_agnostic_nms (boolean): Boolean flag to specify if NMS is to be used in class-agnostic mode..
  • max_candidates (integer): Maximum number of candidates as NMS input to be taken into account..
  • disable_active_learning (boolean): Boolean flag to disable project-level active learning for this block..
  • active_learning_target_dataset (roboflow_project): Target dataset for active learning, if enabled..

• output

  • inference_id (string): String value.
  • predictions (keypoint_detection_prediction): Prediction with detected bounding boxes and detected keypoints in form of sv.Detections(...) object.

Example JSON definition of step Keypoint Detection Model in version v1

{
    "name": "<your_step_name_here>",
    "type": "roboflow_core/roboflow_keypoint_detection_model@v1",
    "images": "$inputs.image",
    "model_id": "my_project/3",
    "confidence": 0.3,
    "keypoint_confidence": 0.3,
    "class_filter": [
        "a",
        "b",
        "c"
    ],
    "iou_threshold": 0.4,
    "max_detections": 300,
    "class_agnostic_nms": true,
    "max_candidates": 3000,
    "disable_active_learning": true,
    "active_learning_target_dataset": "my_project"
}