Instance Segmentation Model¶
v2¶
Class: RoboflowInstanceSegmentationModelBlockV2 (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 an instance segmentation 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_instance_segmentation_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.. | ✅ |
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.. | ✅ |
mask_decode_mode |
str |
Parameter of mask decoding in prediction post-processing.. | ✅ |
tradeoff_factor |
float |
Post-processing parameter to dictate tradeoff between fast and accurate.. | ✅ |
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 Instance Segmentation Model in version v2.
- inputs:
Image Threshold,Email Notification,Corner Visualization,Roboflow Dataset Upload,Object Detection Model,Stitch OCR Detections,Dimension Collapse,Stability AI Image Generation,Grid Visualization,Dynamic Crop,Image Slicer,Image Preprocessing,Instance Segmentation Model,SIFT,Line Counter Visualization,Trace Visualization,Halo Visualization,Roboflow Custom Metadata,Pixelate Visualization,Circle Visualization,Semantic Segmentation Model,S3 Sink,Keypoint Detection Model,Twilio SMS Notification,Halo Visualization,SIFT Comparison,Anthropic Claude,Polygon Visualization,Detections Consensus,Identify Changes,Crop Visualization,Roboflow Dataset Upload,Mask Visualization,Heatmap Visualization,Webhook Sink,Detections List Roll-Up,Google Vision OCR,Florence-2 Model,Florence-2 Model,VLM As Classifier,Anthropic Claude,OpenAI,VLM As Detector,OpenAI,PTZ Tracking (ONVIF),Background Color Visualization,Template Matching,Anthropic Claude,Background Subtraction,SIFT Comparison,Multi-Label Classification Model,Keypoint Visualization,Stitch OCR Detections,LMM,Identify Outliers,Motion Detection,Dynamic Zone,Single-Label Classification Model,Object Detection Model,Roboflow Vision Events,VLM As Classifier,Triangle Visualization,Distance Measurement,Google Gemini,Image Slicer,Image Contours,Model Comparison Visualization,Stability AI Outpainting,Stitch Images,Image Blur,Ellipse Visualization,OpenAI,Depth Estimation,EasyOCR,Absolute Static Crop,Multi-Label Classification Model,JSON Parser,CogVLM,Google Gemini,Relative Static Crop,Morphological Transformation,LMM For Classification,Detection Event Log,Dot Visualization,GLM-OCR,Model Monitoring Inference Aggregator,Keypoint Detection Model,Pixel Color Count,Image Convert Grayscale,Icon Visualization,QR Code Generator,Camera Focus,OCR Model,Text Display,Reference Path Visualization,Instance Segmentation Model,Llama 3.2 Vision,CSV Formatter,Label Visualization,Classification Label Visualization,Polygon Zone Visualization,Stability AI Inpainting,Google Gemini,Perspective Correction,Camera Calibration,Qwen3.5-VL,Size Measurement,Email Notification,Contrast Equalization,Line Counter,Single-Label Classification Model,Line Counter,Color Visualization,OpenAI,Local File Sink,Clip Comparison,Twilio SMS/MMS Notification,Buffer,Clip Comparison,Blur Visualization,Bounding Box Visualization,Camera Focus,Polygon Visualization,VLM As Detector,Slack Notification - outputs:
Corner Visualization,Ellipse Visualization,Roboflow Dataset Upload,Time in Zone,Object Detection Model,Multi-Label Classification Model,Time in Zone,Dynamic Crop,Velocity,Instance Segmentation Model,Detections Combine,Trace Visualization,Detection Event Log,Halo Visualization,Dot Visualization,ByteTrack Tracker,GLM-OCR,Model Monitoring Inference Aggregator,Roboflow Custom Metadata,Keypoint Detection Model,Pixelate Visualization,Circle Visualization,Semantic Segmentation Model,Icon Visualization,Detections Classes Replacement,Keypoint Detection Model,Detections Stabilizer,Halo Visualization,Camera Focus,SAM 3,OC-SORT Tracker,Polygon Visualization,Detections Consensus,Qwen2.5-VL,Instance Segmentation Model,Qwen3-VL,Crop Visualization,Roboflow Dataset Upload,Mask Visualization,Detection Offset,SORT Tracker,Heatmap Visualization,Webhook Sink,Byte Tracker,Byte Tracker,Label Visualization,Detections List Roll-Up,Florence-2 Model,Segment Anything 2 Model,Florence-2 Model,Stability AI Inpainting,Overlap Filter,SAM 3,Perspective Correction,Distance Measurement,PTZ Tracking (ONVIF),Bounding Rectangle,Qwen3.5-VL,Background Color Visualization,Size Measurement,Multi-Label Classification Model,Time in Zone,Line Counter,Path Deviation,Detections Filter,Detections Merge,SmolVLM2,Single-Label Classification Model,Detections Transformation,Byte Tracker,Line Counter,SAM 3,Color Visualization,Dynamic Zone,Single-Label Classification Model,Object Detection Model,Roboflow Vision Events,Mask Area Measurement,Detections Stitch,Triangle Visualization,Blur Visualization,Bounding Box Visualization,Polygon Visualization,Path Deviation,Moondream2,Model Comparison Visualization
Input and Output Bindings¶
The available connections depend on its binding kinds. Check what binding kinds
Instance Segmentation 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..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..mask_decode_mode(string): Parameter of mask decoding in prediction post-processing..tradeoff_factor(float_zero_to_one): Post-processing parameter to dictate tradeoff between fast and accurate..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(instance_segmentation_prediction): Prediction with detected bounding boxes and segmentation masks in form of sv.Detections(...) object.model_id(roboflow_model_id): Roboflow model id.
Example JSON definition of step Instance Segmentation Model in version v2
{
"name": "<your_step_name_here>",
"type": "roboflow_core/roboflow_instance_segmentation_model@v2",
"images": "$inputs.image",
"model_id": "my_project/3",
"confidence": 0.3,
"class_filter": [
"a",
"b",
"c"
],
"iou_threshold": 0.4,
"max_detections": 300,
"class_agnostic_nms": true,
"max_candidates": 3000,
"mask_decode_mode": "accurate",
"tradeoff_factor": 0.3,
"disable_active_learning": true,
"active_learning_target_dataset": "my_project"
}
v1¶
Class: RoboflowInstanceSegmentationModelBlockV1 (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 an instance segmentation 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_instance_segmentation_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.. | ✅ |
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.. | ✅ |
mask_decode_mode |
str |
Parameter of mask decoding in prediction post-processing.. | ✅ |
tradeoff_factor |
float |
Post-processing parameter to dictate tradeoff between fast and accurate.. | ✅ |
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 Instance Segmentation Model in version v1.
- inputs:
Image Threshold,Email Notification,Corner Visualization,Roboflow Dataset Upload,Object Detection Model,Stitch OCR Detections,Dimension Collapse,Stability AI Image Generation,Grid Visualization,Dynamic Crop,Image Slicer,Image Preprocessing,Instance Segmentation Model,SIFT,Line Counter Visualization,Trace Visualization,Halo Visualization,Roboflow Custom Metadata,Pixelate Visualization,Circle Visualization,Semantic Segmentation Model,S3 Sink,Keypoint Detection Model,Twilio SMS Notification,Halo Visualization,SIFT Comparison,Anthropic Claude,Polygon Visualization,Detections Consensus,Identify Changes,Crop Visualization,Roboflow Dataset Upload,Mask Visualization,Heatmap Visualization,Webhook Sink,Detections List Roll-Up,Google Vision OCR,Florence-2 Model,Florence-2 Model,VLM As Classifier,Anthropic Claude,OpenAI,VLM As Detector,OpenAI,PTZ Tracking (ONVIF),Background Color Visualization,Template Matching,Anthropic Claude,Background Subtraction,SIFT Comparison,Multi-Label Classification Model,Keypoint Visualization,Stitch OCR Detections,LMM,Identify Outliers,Motion Detection,Dynamic Zone,Single-Label Classification Model,Object Detection Model,Roboflow Vision Events,VLM As Classifier,Triangle Visualization,Distance Measurement,Google Gemini,Image Slicer,Image Contours,Model Comparison Visualization,Stability AI Outpainting,Stitch Images,Image Blur,Ellipse Visualization,OpenAI,Depth Estimation,EasyOCR,Absolute Static Crop,Multi-Label Classification Model,JSON Parser,CogVLM,Google Gemini,Relative Static Crop,Morphological Transformation,LMM For Classification,Detection Event Log,Dot Visualization,GLM-OCR,Model Monitoring Inference Aggregator,Keypoint Detection Model,Pixel Color Count,Image Convert Grayscale,Icon Visualization,QR Code Generator,Camera Focus,OCR Model,Text Display,Reference Path Visualization,Instance Segmentation Model,Llama 3.2 Vision,CSV Formatter,Label Visualization,Classification Label Visualization,Polygon Zone Visualization,Stability AI Inpainting,Google Gemini,Perspective Correction,Camera Calibration,Qwen3.5-VL,Size Measurement,Email Notification,Contrast Equalization,Line Counter,Single-Label Classification Model,Line Counter,Color Visualization,OpenAI,Local File Sink,Clip Comparison,Twilio SMS/MMS Notification,Buffer,Clip Comparison,Blur Visualization,Bounding Box Visualization,Camera Focus,Polygon Visualization,VLM As Detector,Slack Notification - outputs:
Image Threshold,Email Notification,Corner Visualization,Roboflow Dataset Upload,Stitch OCR Detections,Stability AI Image Generation,Time in Zone,Dynamic Crop,Instance Segmentation Model,Image Preprocessing,Line Counter Visualization,Detections Combine,Trace Visualization,Halo Visualization,ByteTrack Tracker,Cache Get,Roboflow Custom Metadata,Pixelate Visualization,Circle Visualization,S3 Sink,Detections Classes Replacement,Twilio SMS Notification,Halo Visualization,Anthropic Claude,OC-SORT Tracker,Polygon Visualization,Detections Consensus,Roboflow Dataset Upload,Crop Visualization,Mask Visualization,Detection Offset,CLIP Embedding Model,Heatmap Visualization,Webhook Sink,Cache Set,Google Vision OCR,Detections List Roll-Up,Florence-2 Model,Florence-2 Model,Overlap Filter,Anthropic Claude,OpenAI,OpenAI,PTZ Tracking (ONVIF),Bounding Rectangle,Background Color Visualization,Anthropic Claude,SIFT Comparison,Keypoint Visualization,Time in Zone,Detections Filter,Stitch OCR Detections,LMM,Detections Merge,Detections Transformation,Perception Encoder Embedding Model,SAM 3,Seg Preview,Dynamic Zone,Roboflow Vision Events,Detections Stitch,Triangle Visualization,Distance Measurement,Google Gemini,Path Deviation,Model Comparison Visualization,Stability AI Outpainting,Image Blur,Ellipse Visualization,OpenAI,Time in Zone,Depth Estimation,CogVLM,Google Gemini,Velocity,Morphological Transformation,LMM For Classification,Detection Event Log,Dot Visualization,GLM-OCR,Model Monitoring Inference Aggregator,Pixel Color Count,Icon Visualization,QR Code Generator,Detections Stabilizer,Camera Focus,SAM 3,Text Display,Reference Path Visualization,Instance Segmentation Model,Llama 3.2 Vision,SORT Tracker,Byte Tracker,Label Visualization,Classification Label Visualization,Byte Tracker,Segment Anything 2 Model,Polygon Zone Visualization,Stability AI Inpainting,Google Gemini,SAM 3,Perspective Correction,Size Measurement,Email Notification,Contrast Equalization,Line Counter,Path Deviation,Byte Tracker,Line Counter,Color Visualization,OpenAI,Local File Sink,Mask Area Measurement,Twilio SMS/MMS Notification,YOLO-World Model,Clip Comparison,Bounding Box Visualization,Blur Visualization,Polygon Visualization,Moondream2,Slack Notification
Input and Output Bindings¶
The available connections depend on its binding kinds. Check what binding kinds
Instance Segmentation 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..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..mask_decode_mode(string): Parameter of mask decoding in prediction post-processing..tradeoff_factor(float_zero_to_one): Post-processing parameter to dictate tradeoff between fast and accurate..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(instance_segmentation_prediction): Prediction with detected bounding boxes and segmentation masks in form of sv.Detections(...) object.
Example JSON definition of step Instance Segmentation Model in version v1
{
"name": "<your_step_name_here>",
"type": "roboflow_core/roboflow_instance_segmentation_model@v1",
"images": "$inputs.image",
"model_id": "my_project/3",
"confidence": 0.3,
"class_filter": [
"a",
"b",
"c"
],
"iou_threshold": 0.4,
"max_detections": 300,
"class_agnostic_nms": true,
"max_candidates": 3000,
"mask_decode_mode": "accurate",
"tradeoff_factor": 0.3,
"disable_active_learning": true,
"active_learning_target_dataset": "my_project"
}