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

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>"
}