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