Keypoint Detection Model¶
v3¶
Class: RoboflowKeypointDetectionModelBlockV3 (there are multiple versions of this block)
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@v3to 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_mode |
str |
not available. | ✅ |
custom_confidence |
float |
not available. | ✅ |
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 v3.
- inputs:
Morphological Transformation,Image Preprocessing,Email Notification,VLM As Classifier,Halo Visualization,Morphological Transformation,Object Detection Model,Pixel Color Count,Text Display,Template Matching,Image Threshold,Model Monitoring Inference Aggregator,Pixelate Visualization,Keypoint Detection Model,Qwen-VL,OpenAI,CogVLM,Crop Visualization,Dot Visualization,Google Vision OCR,PLC EthernetIP,Detections List Roll-Up,Florence-2 Model,Dimension Collapse,Roboflow Dataset Upload,Qwen3.5-VL,Roboflow Vision Events,Polygon Zone Visualization,Polygon Visualization,SIFT Comparison,Absolute Static Crop,S3 Sink,Twilio SMS/MMS Notification,QR Code Generator,SIFT Comparison,Contrast Enhancement,Single-Label Classification Model,Stitch OCR Detections,Dynamic Zone,OCR Model,Color Visualization,Roboflow Dataset Upload,Roboflow Custom Metadata,LMM For Classification,OpenAI-Compatible LLM,Line Counter Visualization,Image Blur,Stability AI Inpainting,Object Detection Model,Blur Visualization,Current Time,Perspective Correction,Keypoint Visualization,Anthropic Claude,MQTT Writer,MoonshotAI Kimi,Google Gemini,Image Slicer,Identify Changes,Depth Estimation,Detections Consensus,Ellipse Visualization,Google Gemma API,Object Detection Model,PLC ModbusTCP,Slack Notification,Identify Outliers,Image Stack,Google Gemini,Bounding Box Visualization,Label Visualization,Keypoint Detection Model,Stitch OCR Detections,Camera Focus,CSV Formatter,Size Measurement,Keypoint Detection Model,Multi-Label Classification Model,OpenAI,SIFT,Anthropic Claude,Image Convert Grayscale,Roboflow Asset Library Attributes,Florence-2 Model,EasyOCR,Multi-Label Classification Model,Buffer,Twilio SMS Notification,Local File Sink,Single-Label Classification Model,Icon Visualization,Triangle Visualization,Qwen 3.5 API,VLM As Classifier,JSON Parser,OpenRouter,Instance Segmentation Model,Distance Measurement,Instance Segmentation Model,OpenAI,Background Color Visualization,MoonshotAI Kimi,Google Gemini,Grid Visualization,Clip Comparison,Semantic Segmentation Model,Corner Visualization,Reference Path Visualization,Image Slicer,Single-Label Classification Model,Line Counter,Halo Visualization,Dynamic Crop,Webhook Sink,Instance Segmentation Model,Stability AI Outpainting,Detection Event Log,VLM As Detector,Relative Static Crop,Anthropic Claude,Clip Comparison,Multi-Label Classification Model,OpenAI,Llama 3.2 Vision,Motion Detection,Camera Calibration,Model Comparison Visualization,Trace Visualization,Google Gemma,PTZ Tracking (ONVIF),OPC UA Writer Sink,Line Counter,Circle Visualization,Email Notification,LMM,Event Writer,Instance Segmentation Model,Contrast Equalization,Camera Focus,Heatmap Visualization,Background Subtraction,Image Contours,Qwen 3.6 API,GLM-OCR,VLM As Detector,Classification Label Visualization,Llama 3.2 Vision,Stitch Images,Mask Visualization,Microsoft SQL Server Sink,Stability AI Image Generation,Semantic Segmentation Model,Polygon Visualization - outputs:
Dynamic Crop,Detections Classes Replacement,Webhook Sink,Byte Tracker,Color Visualization,Roboflow Dataset Upload,Instance Segmentation Model,Roboflow Custom Metadata,OC-SORT Tracker,Moondream2,Florence-2 Model,Object Detection Model,Blur Visualization,Detections Transformation,SORT Tracker,Multi-Label Classification Model,SAM 3,Multi-Label Classification Model,Segment Anything 2 Model,Keypoint Visualization,Object Detection Model,BoT-SORT Tracker,Single-Label Classification Model,ByteTrack Tracker,Triangle Visualization,Icon Visualization,Detection Offset,Model Monitoring Inference Aggregator,Velocity,Pixelate Visualization,Keypoint Detection Model,Trace Visualization,Model Comparison Visualization,SAM 3,Detections Consensus,Ellipse Visualization,Crop Visualization,Qwen-VL,SAM 3,Dot Visualization,Object Detection Model,Detections Merge,Single-Label Classification Model,Detections List Roll-Up,Circle Visualization,Florence-2 Model,Instance Segmentation Model,Roboflow Dataset Upload,Qwen3.5-VL,Roboflow Vision Events,Qwen3-VL,Qwen2.5-VL,Event Writer,Instance Segmentation Model,Heatmap Visualization,Instance Segmentation Model,SAM2 Video Tracker,GLM-OCR,Background Color Visualization,Qwen3.5,Bounding Box Visualization,Label Visualization,Keypoint Detection Model,Per-Class Confidence Filter,Keypoint Detection Model,Semantic Segmentation Model,Corner Visualization,Multi-Label Classification Model,SmolVLM2,Single-Label Classification Model,Semantic Segmentation Model,Detections Filter
Input and Output Bindings¶
The available connections depend on its binding kinds. Check what binding kinds
Keypoint Detection Model in version v3 has.
Bindings
-
input
images(image): The image to infer on..model_id(roboflow_model_id): Roboflow model identifier..confidence_mode(string): not available.custom_confidence(float_zero_to_one): not available.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 v3
{
"name": "<your_step_name_here>",
"type": "roboflow_core/roboflow_keypoint_detection_model@v3",
"images": "$inputs.image",
"model_id": "my_project/3",
"confidence_mode": "<block_does_not_provide_example>",
"custom_confidence": "<block_does_not_provide_example>",
"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"
}
v2¶
Class: RoboflowKeypointDetectionModelBlockV2 (there are multiple versions of this block)
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:
Morphological Transformation,Image Preprocessing,Email Notification,VLM As Classifier,Halo Visualization,Morphological Transformation,Pixel Color Count,Text Display,Template Matching,Image Threshold,Model Monitoring Inference Aggregator,Pixelate Visualization,Qwen-VL,Crop Visualization,Dot Visualization,PLC EthernetIP,Detections List Roll-Up,Florence-2 Model,Dimension Collapse,Roboflow Dataset Upload,Roboflow Vision Events,Polygon Zone Visualization,Polygon Visualization,SIFT Comparison,Absolute Static Crop,S3 Sink,Twilio SMS/MMS Notification,QR Code Generator,SIFT Comparison,Contrast Enhancement,Single-Label Classification Model,Dynamic Zone,Color Visualization,Roboflow Dataset Upload,Roboflow Custom Metadata,Line Counter Visualization,Image Blur,Stability AI Inpainting,Object Detection Model,Blur Visualization,Perspective Correction,Keypoint Visualization,Anthropic Claude,MQTT Writer,MoonshotAI Kimi,Google Gemini,Image Slicer,Identify Changes,Depth Estimation,Detections Consensus,Ellipse Visualization,Google Gemma API,Object Detection Model,PLC ModbusTCP,Slack Notification,Identify Outliers,Image Stack,Google Gemini,Bounding Box Visualization,Label Visualization,Keypoint Detection Model,Size Measurement,Camera Focus,Keypoint Detection Model,Multi-Label Classification Model,OpenAI,SIFT,Anthropic Claude,Image Convert Grayscale,Roboflow Asset Library Attributes,Florence-2 Model,Multi-Label Classification Model,Buffer,Twilio SMS Notification,Local File Sink,Single-Label Classification Model,Icon Visualization,Triangle Visualization,Qwen 3.5 API,VLM As Classifier,JSON Parser,OpenRouter,Instance Segmentation Model,Distance Measurement,Instance Segmentation Model,OpenAI,Background Color Visualization,MoonshotAI Kimi,Google Gemini,Grid Visualization,Clip Comparison,Semantic Segmentation Model,Corner Visualization,Reference Path Visualization,Image Slicer,Line Counter,Halo Visualization,Dynamic Crop,Webhook Sink,Instance Segmentation Model,Stability AI Outpainting,Detection Event Log,VLM As Detector,Relative Static Crop,Anthropic Claude,Clip Comparison,OpenAI,Llama 3.2 Vision,Motion Detection,PTZ Tracking (ONVIF),Camera Calibration,Model Comparison Visualization,Trace Visualization,Google Gemma,Line Counter,OPC UA Writer Sink,Circle Visualization,Email Notification,Event Writer,Contrast Equalization,Camera Focus,Heatmap Visualization,Background Subtraction,Image Contours,Qwen 3.6 API,VLM As Detector,Classification Label Visualization,Llama 3.2 Vision,Stitch Images,Mask Visualization,Microsoft SQL Server Sink,Stability AI Image Generation,Semantic Segmentation Model,Polygon Visualization - outputs:
Dynamic Crop,Detections Classes Replacement,Webhook Sink,Byte Tracker,Color Visualization,Roboflow Dataset Upload,Instance Segmentation Model,Roboflow Custom Metadata,OC-SORT Tracker,Moondream2,Florence-2 Model,Object Detection Model,Blur Visualization,Detections Transformation,SORT Tracker,Multi-Label Classification Model,SAM 3,Multi-Label Classification Model,Segment Anything 2 Model,Keypoint Visualization,Object Detection Model,BoT-SORT Tracker,Single-Label Classification Model,ByteTrack Tracker,Triangle Visualization,Icon Visualization,Detection Offset,Model Monitoring Inference Aggregator,Velocity,Pixelate Visualization,Keypoint Detection Model,Trace Visualization,Model Comparison Visualization,SAM 3,Detections Consensus,Ellipse Visualization,Crop Visualization,Qwen-VL,SAM 3,Dot Visualization,Object Detection Model,Detections Merge,Single-Label Classification Model,Detections List Roll-Up,Circle Visualization,Florence-2 Model,Instance Segmentation Model,Roboflow Dataset Upload,Qwen3.5-VL,Roboflow Vision Events,Qwen3-VL,Qwen2.5-VL,Event Writer,Instance Segmentation Model,Heatmap Visualization,Instance Segmentation Model,SAM2 Video Tracker,GLM-OCR,Background Color Visualization,Qwen3.5,Bounding Box Visualization,Label Visualization,Keypoint Detection Model,Per-Class Confidence Filter,Keypoint Detection Model,Semantic Segmentation Model,Corner Visualization,Multi-Label Classification Model,SmolVLM2,Single-Label Classification Model,Semantic Segmentation Model,Detections Filter
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)
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:
Morphological Transformation,Image Preprocessing,Email Notification,VLM As Classifier,Halo Visualization,Morphological Transformation,Pixel Color Count,Text Display,Template Matching,Image Threshold,Model Monitoring Inference Aggregator,Pixelate Visualization,Qwen-VL,Crop Visualization,Dot Visualization,PLC EthernetIP,Detections List Roll-Up,Florence-2 Model,Dimension Collapse,Roboflow Dataset Upload,Roboflow Vision Events,Polygon Zone Visualization,Polygon Visualization,SIFT Comparison,Absolute Static Crop,S3 Sink,Twilio SMS/MMS Notification,QR Code Generator,SIFT Comparison,Contrast Enhancement,Single-Label Classification Model,Dynamic Zone,Color Visualization,Roboflow Dataset Upload,Roboflow Custom Metadata,Line Counter Visualization,Image Blur,Stability AI Inpainting,Object Detection Model,Blur Visualization,Perspective Correction,Keypoint Visualization,Anthropic Claude,MQTT Writer,MoonshotAI Kimi,Google Gemini,Image Slicer,Identify Changes,Depth Estimation,Detections Consensus,Ellipse Visualization,Google Gemma API,Object Detection Model,PLC ModbusTCP,Slack Notification,Identify Outliers,Image Stack,Google Gemini,Bounding Box Visualization,Label Visualization,Keypoint Detection Model,Size Measurement,Camera Focus,Keypoint Detection Model,Multi-Label Classification Model,OpenAI,SIFT,Anthropic Claude,Image Convert Grayscale,Roboflow Asset Library Attributes,Florence-2 Model,Multi-Label Classification Model,Buffer,Twilio SMS Notification,Local File Sink,Single-Label Classification Model,Icon Visualization,Triangle Visualization,Qwen 3.5 API,VLM As Classifier,JSON Parser,OpenRouter,Instance Segmentation Model,Distance Measurement,Instance Segmentation Model,OpenAI,Background Color Visualization,MoonshotAI Kimi,Google Gemini,Grid Visualization,Clip Comparison,Semantic Segmentation Model,Corner Visualization,Reference Path Visualization,Image Slicer,Line Counter,Halo Visualization,Dynamic Crop,Webhook Sink,Instance Segmentation Model,Stability AI Outpainting,Detection Event Log,VLM As Detector,Relative Static Crop,Anthropic Claude,Clip Comparison,OpenAI,Llama 3.2 Vision,Motion Detection,PTZ Tracking (ONVIF),Camera Calibration,Model Comparison Visualization,Trace Visualization,Google Gemma,Line Counter,OPC UA Writer Sink,Circle Visualization,Email Notification,Event Writer,Contrast Equalization,Camera Focus,Heatmap Visualization,Background Subtraction,Image Contours,Qwen 3.6 API,VLM As Detector,Classification Label Visualization,Llama 3.2 Vision,Stitch Images,Mask Visualization,Microsoft SQL Server Sink,Stability AI Image Generation,Semantic Segmentation Model,Polygon Visualization - outputs:
Detections Classes Replacement,Morphological Transformation,Image Preprocessing,Email Notification,Morphological Transformation,Halo Visualization,Detections Transformation,Pixel Color Count,Time in Zone,Text Display,BoT-SORT Tracker,Image Threshold,Model Monitoring Inference Aggregator,Pixelate Visualization,Time in Zone,Qwen-VL,OpenAI,CogVLM,Crop Visualization,SAM 3,Dot Visualization,Google Vision OCR,Detections Merge,Detections List Roll-Up,Florence-2 Model,Roboflow Dataset Upload,Qwen3.5-VL,Roboflow Vision Events,Polygon Zone Visualization,Polygon Visualization,S3 Sink,Twilio SMS/MMS Notification,QR Code Generator,SIFT Comparison,Per-Class Confidence Filter,Cache Get,Stitch OCR Detections,Detections Filter,Byte Tracker,Color Visualization,Roboflow Dataset Upload,Roboflow Custom Metadata,LMM For Classification,OpenAI-Compatible LLM,Stability AI Inpainting,Image Blur,Line Counter Visualization,Object Detection Model,Blur Visualization,Path Deviation,Current Time,SAM 3,Perspective Correction,Keypoint Visualization,Detection Offset,Anthropic Claude,MQTT Writer,MoonshotAI Kimi,Google Gemini,SAM 3,Depth Estimation,Detections Stitch,Ellipse Visualization,Google Gemma API,Detections Consensus,Slack Notification,Time in Zone,Google Gemini,Cache Set,Label Visualization,Bounding Box Visualization,Keypoint Detection Model,Stitch OCR Detections,Size Measurement,Multi-Label Classification Model,OpenAI,Perception Encoder Embedding Model,Anthropic Claude,Roboflow Asset Library Attributes,Moondream2,OC-SORT Tracker,CLIP Embedding Model,Florence-2 Model,Seg Preview,YOLO-World Model,Segment Anything 2 Model,Twilio SMS Notification,Local File Sink,Single-Label Classification Model,Triangle Visualization,Icon Visualization,Qwen 3.5 API,Path Deviation,OpenRouter,Instance Segmentation Model,Distance Measurement,Instance Segmentation Model,OpenAI,Background Color Visualization,MoonshotAI Kimi,Google Gemini,Corner Visualization,Reference Path Visualization,Line Counter,Halo Visualization,Webhook Sink,Dynamic Crop,Instance Segmentation Model,Stability AI Outpainting,Anthropic Claude,Clip Comparison,SORT Tracker,OpenAI,Llama 3.2 Vision,ByteTrack Tracker,Velocity,Trace Visualization,Model Comparison Visualization,Google Gemma,PTZ Tracking (ONVIF),OPC UA Writer Sink,Line Counter,Circle Visualization,Email Notification,LMM,Event Writer,Instance Segmentation Model,Contrast Equalization,Heatmap Visualization,GLM-OCR,Qwen 3.6 API,SAM2 Video Tracker,Llama 3.2 Vision,Classification Label Visualization,Mask Visualization,Microsoft SQL Server Sink,Stability AI Image Generation,Semantic Segmentation Model,Polygon Visualization
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"
}