Gaze Detection

Class: GazeBlockV1

Source: inference.core.workflows.core_steps.models.foundation.gaze.v1.GazeBlockV1

Run L2CS Gaze detection model on faces in images.

This block can: 1. Detect faces in images and estimate their gaze direction 2. Estimate gaze direction on pre-cropped face images

The gaze direction is represented by yaw and pitch angles in degrees.

Type identifier

Use the following identifier in step "type" field: roboflow_core/gaze@v1to add the block as as step in your workflow.

Properties

Name	Type	Description	Refs
name	str	Enter a unique identifier for this step..	❌
do_run_face_detection	bool	Whether to run face detection. Set to False if input images are pre-cropped face images..	✅

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 Gaze Detection in version v1.

• inputs: Roboflow Dataset Upload, Email Notification, VLM as Detector, Mask Visualization, SIFT, Model Comparison Visualization, Stability AI Outpainting, Contrast Equalization, Identify Changes, Absolute Static Crop, PTZ Tracking (ONVIF).md), Polygon Visualization, Dynamic Zone, SIFT Comparison, VLM as Detector, Image Slicer, Dynamic Crop, Background Subtraction, Image Threshold, Local File Sink, Icon Visualization, Reference Path Visualization, Depth Estimation, Circle Visualization, Stitch Images, Trace Visualization, Email Notification, Stability AI Image Generation, Pixelate Visualization, Twilio SMS/MMS Notification, Triangle Visualization, Line Counter Visualization, Label Visualization, Camera Calibration, Detections Consensus, Camera Focus, SIFT Comparison, Webhook Sink, Halo Visualization, Text Display, Blur Visualization, Twilio SMS Notification, Relative Static Crop, QR Code Generator, Ellipse Visualization, Dot Visualization, Keypoint Visualization, Crop Visualization, Polygon Zone Visualization, Perspective Correction, Identify Outliers, JSON Parser, Image Convert Grayscale, VLM as Classifier, Color Visualization, Grid Visualization, Stability AI Inpainting, Background Color Visualization, Slack Notification, Image Slicer, Image Contours, Motion Detection, Image Preprocessing, Classification Label Visualization, Roboflow Dataset Upload, VLM as Classifier, Bounding Box Visualization, Roboflow Custom Metadata, Morphological Transformation, Camera Focus, Model Monitoring Inference Aggregator, Image Blur, Corner Visualization
• outputs: Roboflow Dataset Upload, Byte Tracker, Velocity, Model Comparison Visualization, Google Gemini, Dynamic Zone, Dynamic Crop, Icon Visualization, OpenAI, Circle Visualization, Trace Visualization, Pixelate Visualization, Line Counter Visualization, Camera Calibration, Webhook Sink, Blur Visualization, OpenAI, Ellipse Visualization, Crop Visualization, Seg Preview, OpenAI, Detections Transformation, Florence-2 Model, SAM 3, Roboflow Custom Metadata, Model Monitoring Inference Aggregator, Corner Visualization, Anthropic Claude, Detections Merge, Anthropic Claude, SAM 3, SAM 3, PTZ Tracking (ONVIF).md), Detections List Roll-Up, Segment Anything 2 Model, Google Gemini, Template Matching, Florence-2 Model, Triangle Visualization, Label Visualization, Detections Consensus, Detections Filter, Detection Offset, Text Display, Keypoint Visualization, Dot Visualization, Distance Measurement, Color Visualization, Background Color Visualization, Roboflow Dataset Upload, Anthropic Claude, Bounding Box Visualization, Google Gemini, Detections Classes Replacement, Llama 3.2 Vision, Continue If

Input and Output Bindings

The available connections depend on its binding kinds. Check what binding kinds Gaze Detection in version v1 has.

Bindings

• input

  • images (image): The image to infer on..
  • do_run_face_detection (boolean): Whether to run face detection. Set to False if input images are pre-cropped face images..

• output

  • face_predictions (keypoint_detection_prediction): Prediction with detected bounding boxes and detected keypoints in form of sv.Detections(...) object.
  • yaw_degrees (float): Float value.
  • pitch_degrees (float): Float value.

Example JSON definition of step Gaze Detection in version v1

{
    "name": "<your_step_name_here>",
    "type": "roboflow_core/gaze@v1",
    "images": "$inputs.image",
    "do_run_face_detection": "<block_does_not_provide_example>"
}