Seg Preview¶

Check what blocks you can connect to Seg Preview in version v1.

• inputs: Heatmap Visualization, Pixelate Visualization, Roboflow Custom Metadata, Halo Visualization, Anthropic Claude, OpenAI, Halo Visualization, Contrast Equalization, Trace Visualization, Grid Visualization, Ellipse Visualization, Model Monitoring Inference Aggregator, Local File Sink, Webhook Sink, Dynamic Crop, Image Convert Grayscale, Corner Visualization, Polygon Zone Visualization, Circle Visualization, Model Comparison Visualization, Florence-2 Model, Image Slicer, Stability AI Outpainting, Dynamic Zone, Keypoint Visualization, Text Display, Google Vision OCR, EasyOCR, OCR Model, Anthropic Claude, Size Measurement, Clip Comparison, Twilio SMS Notification, Stability AI Inpainting, Roboflow Dataset Upload, SIFT Comparison, Reference Path Visualization, Image Blur, Stitch OCR Detections, VLM As Classifier, QR Code Generator, CogVLM, Cosine Similarity, LMM For Classification, Image Slicer, Icon Visualization, Background Color Visualization, Absolute Static Crop, Google Gemini, Label Visualization, Image Preprocessing, Classification Label Visualization, SIFT, OpenAI, Mask Visualization, Single-Label Classification Model, Bounding Box Visualization, OpenAI, Slack Notification, Roboflow Dataset Upload, Triangle Visualization, Dimension Collapse, Detections List Roll-Up, Dot Visualization, Keypoint Detection Model, Email Notification, Twilio SMS/MMS Notification, Google Gemini, Image Contours, Anthropic Claude, Instance Segmentation Model, Stitch OCR Detections, Email Notification, LMM, Florence-2 Model, Background Subtraction, Camera Calibration, Image Threshold, VLM As Detector, Clip Comparison, Relative Static Crop, Gaze Detection, Identify Changes, Crop Visualization, Stitch Images, Blur Visualization, Perspective Correction, Motion Detection, CSV Formatter, Camera Focus, Camera Focus, Line Counter Visualization, Color Visualization, Morphological Transformation, Llama 3.2 Vision, Google Gemini, Buffer, Multi-Label Classification Model, OpenAI, Qwen3.5-VL, Stability AI Image Generation, Depth Estimation, Polygon Visualization, Object Detection Model, Polygon Visualization
• outputs: Path Deviation, Roboflow Custom Metadata, Heatmap Visualization, Line Counter, Detections Stitch, Halo Visualization, Pixelate Visualization, Byte Tracker, Mask Visualization, Time in Zone, Bounding Box Visualization, Detections Consensus, Distance Measurement, PTZ Tracking (ONVIF), Mask Area Measurement, Halo Visualization, Detections Filter, Detection Offset, Triangle Visualization, Trace Visualization, Detections List Roll-Up, Dot Visualization, Ellipse Visualization, Model Monitoring Inference Aggregator, Segment Anything 2 Model, Dynamic Crop, Detection Event Log, Byte Tracker, Corner Visualization, Circle Visualization, Model Comparison Visualization, Florence-2 Model, Florence-2 Model, Bounding Rectangle, Dynamic Zone, Overlap Filter, Detections Merge, Detections Combine, Byte Tracker, Crop Visualization, Size Measurement, Detections Stabilizer, Detections Classes Replacement, Velocity, Stability AI Inpainting, Roboflow Dataset Upload, Blur Visualization, Perspective Correction, Path Deviation, Camera Focus, Detections Transformation, Color Visualization, Line Counter, Time in Zone, Icon Visualization, Background Color Visualization, Label Visualization, Time in Zone, Roboflow Dataset Upload, Polygon Visualization, Polygon Visualization

• input

  • images (image): The image to infer on..
  • class_names (Union[list_of_values, string]): List of classes to recognise.
  • threshold (float): Threshold for predicted mask scores.

• output

  • predictions (instance_segmentation_prediction): Prediction with detected bounding boxes and segmentation masks in form of sv.Detections(...) object.