Segment Anything 2 Model¶
Class: SegmentAnything2BlockV1
Source: inference.core.workflows.core_steps.models.foundation.segment_anything2.v1.SegmentAnything2BlockV1
Run Segment Anything 2, a zero-shot instance segmentation model, on an image.
** Dedicated inference server required (GPU recomended) **
You can use pass in boxes/predictions from other models to Segment Anything 2 to use as prompts for the model. If you pass in box detections from another model, the class names of the boxes will be forwarded to the predicted masks. If using the model unprompted, the model will assign integers as class names / ids.
Type identifier¶
Use the following identifier in step "type" field: roboflow_core/segment_anything@v1to add the block as
as step in your workflow.
Properties¶
| Name | Type | Description | Refs |
|---|---|---|---|
name |
str |
Enter a unique identifier for this step.. | ❌ |
version |
str |
Model to be used. One of hiera_large, hiera_small, hiera_tiny, hiera_b_plus. | ✅ |
threshold |
float |
Threshold for predicted masks scores. | ✅ |
multimask_output |
bool |
Flag to determine whether to use sam2 internal multimask or single mask mode. For ambiguous prompts setting to True is recomended.. | ✅ |
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 Segment Anything 2 Model in version v1.
- inputs:
Ellipse Visualization,Bounding Rectangle,Label Visualization,Llama 3.2 Vision,Polygon Zone Visualization,Detections Stabilizer,Time in Zone,Image Contours,Background Color Visualization,Color Visualization,Image Blur,Stitch Images,LMM,Segment Anything 2 Model,Dynamic Crop,Pixelate Visualization,Instance Segmentation Model,Keypoint Detection Model,Template Matching,VLM as Detector,SIFT Comparison,Detection Offset,Object Detection Model,Reference Path Visualization,VLM as Classifier,Classification Label Visualization,Clip Comparison,Perspective Correction,OCR Model,Byte Tracker,Roboflow Dataset Upload,Instance Segmentation Model,Roboflow Dataset Upload,Image Slicer,Trace Visualization,YOLO-World Model,Model Monitoring Inference Aggregator,Twilio SMS Notification,Detections Filter,Line Counter Visualization,Google Vision OCR,VLM as Detector,Image Convert Grayscale,Byte Tracker,Corner Visualization,Google Gemini,Blur Visualization,Detections Stitch,Polygon Visualization,Email Notification,Halo Visualization,Single-Label Classification Model,Local File Sink,CogVLM,JSON Parser,Florence-2 Model,Grid Visualization,Gaze Detection,Florence-2 Model,Image Slicer,LMM For Classification,Detections Consensus,Anthropic Claude,Image Threshold,Camera Focus,Keypoint Visualization,Dynamic Zone,Byte Tracker,Stitch OCR Detections,Keypoint Detection Model,Roboflow Custom Metadata,Crop Visualization,Absolute Static Crop,Detections Transformation,Stability AI Image Generation,Path Deviation,Line Counter,Bounding Box Visualization,Multi-Label Classification Model,Identify Changes,Slack Notification,Time in Zone,Detections Classes Replacement,OpenAI,Dot Visualization,Mask Visualization,Stability AI Inpainting,Webhook Sink,SIFT Comparison,Velocity,CSV Formatter,Model Comparison Visualization,Path Deviation,Identify Outliers,Circle Visualization,Image Preprocessing,SIFT,Cosine Similarity,VLM as Classifier,Object Detection Model,Relative Static Crop,Triangle Visualization,OpenAI - outputs:
Bounding Rectangle,Ellipse Visualization,Line Counter,Florence-2 Model,Florence-2 Model,Label Visualization,Distance Measurement,Detections Stabilizer,Size Measurement,Time in Zone,Background Color Visualization,Detections Consensus,Color Visualization,Segment Anything 2 Model,Dynamic Crop,Pixelate Visualization,Dynamic Zone,Byte Tracker,Roboflow Custom Metadata,Crop Visualization,Detection Offset,Detections Transformation,Path Deviation,Line Counter,Bounding Box Visualization,Time in Zone,Perspective Correction,Byte Tracker,Roboflow Dataset Upload,Detections Classes Replacement,Roboflow Dataset Upload,Mask Visualization,Dot Visualization,Stability AI Inpainting,Velocity,Trace Visualization,Model Monitoring Inference Aggregator,Model Comparison Visualization,Detections Filter,Path Deviation,Circle Visualization,Byte Tracker,Corner Visualization,Blur Visualization,Detections Stitch,Polygon Visualization,Triangle Visualization,Halo Visualization
Input and Output Bindings¶
The available connections depend on its binding kinds. Check what binding kinds
Segment Anything 2 Model in version v1 has.
Bindings
-
input
images(image): The image to infer on..boxes(Union[keypoint_detection_prediction,object_detection_prediction,instance_segmentation_prediction]): Bounding boxes (from another model) to convert to polygons.version(string): Model to be used. One of hiera_large, hiera_small, hiera_tiny, hiera_b_plus.threshold(float): Threshold for predicted masks scores.multimask_output(boolean): Flag to determine whether to use sam2 internal multimask or single mask mode. For ambiguous prompts setting to True is recomended..
-
output
predictions(instance_segmentation_prediction): Prediction with detected bounding boxes and segmentation masks in form of sv.Detections(...) object.
Example JSON definition of step Segment Anything 2 Model in version v1
{
"name": "<your_step_name_here>",
"type": "roboflow_core/segment_anything@v1",
"images": "$inputs.image",
"boxes": "$steps.object_detection_model.predictions",
"version": "hiera_large",
"threshold": 0.3,
"multimask_output": true
}