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

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>"
}