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: Crop Visualization, Keypoint Detection Model, Detections Filter, Ellipse Visualization, VLM as Classifier, Stability AI Inpainting, Stability AI Image Generation, Blur Visualization, Keypoint Detection Model, Circle Visualization, Pixelate Visualization, YOLO-World Model, Model Comparison Visualization, Webhook Sink, Stability AI Outpainting, Detections Transformation, Detections Consensus, Bounding Box Visualization, Time in Zone, Time in Zone, Path Deviation, Identify Changes, Grid Visualization, Cosine Similarity, VLM as Classifier, Identify Outliers, Background Color Visualization, Llama 3.2 Vision, Image Contours, LMM, Object Detection Model, Twilio SMS Notification, Label Visualization, Triangle Visualization, Dynamic Crop, Detections Stitch, VLM as Detector, Google Vision OCR, Florence-2 Model, Instance Segmentation Model, Florence-2 Model, Keypoint Visualization, OCR Model, Halo Visualization, PTZ Tracking (ONVIF).md), Corner Visualization, Line Counter Visualization, LMM For Classification, Time in Zone, Template Matching, Dot Visualization, Anthropic Claude, Slack Notification, Roboflow Dataset Upload, Icon Visualization, Roboflow Custom Metadata, Depth Estimation, CogVLM, Byte Tracker, Polygon Zone Visualization, Stitch Images, Image Slicer, OpenAI, Email Notification, Relative Static Crop, Segment Anything 2 Model, Detections Classes Replacement, Google Gemini, SIFT Comparison, SIFT Comparison, Line Counter, Byte Tracker, Image Slicer, Gaze Detection, Image Threshold, Perspective Correction, Byte Tracker, Camera Focus, JSON Parser, Camera Calibration, Classification Label Visualization, Reference Path Visualization, Color Visualization, Overlap Filter, Path Deviation, Instance Segmentation Model, Model Monitoring Inference Aggregator, Local File Sink, OpenAI, Clip Comparison, Velocity, Mask Visualization, QR Code Generator, Bounding Rectangle, Roboflow Dataset Upload, OpenAI, CSV Formatter, Stitch OCR Detections, SIFT, Polygon Visualization, Image Convert Grayscale, VLM as Detector, Detections Stabilizer, Dynamic Zone, Moondream2, Detection Offset, Image Blur, Multi-Label Classification Model, Trace Visualization, Detections Merge, Absolute Static Crop, Image Preprocessing, Single-Label Classification Model, Object Detection Model
• outputs: Crop Visualization, Detections Filter, Ellipse Visualization, Stability AI Inpainting, Segment Anything 2 Model, Blur Visualization, Circle Visualization, Pixelate Visualization, Detections Classes Replacement, Model Comparison Visualization, Detections Transformation, Detections Consensus, Bounding Box Visualization, Line Counter, Time in Zone, Path Deviation, Time in Zone, Background Color Visualization, Byte Tracker, Perspective Correction, Label Visualization, Byte Tracker, Triangle Visualization, Detections Stitch, Florence-2 Model, Dynamic Crop, Overlap Filter, Florence-2 Model, Path Deviation, Model Monitoring Inference Aggregator, Color Visualization, Halo Visualization, PTZ Tracking (ONVIF).md), Corner Visualization, Roboflow Dataset Upload, Mask Visualization, Bounding Rectangle, Velocity, Time in Zone, Polygon Visualization, Dot Visualization, Size Measurement, Detections Stabilizer, Roboflow Dataset Upload, Icon Visualization, Roboflow Custom Metadata, Detection Offset, Distance Measurement, Dynamic Zone, Trace Visualization, Line Counter, Detections Merge, Byte Tracker

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[object_detection_prediction, keypoint_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
}