Funding
Funding not disclosed
Founders
Product
Problem
Accurate eye‑tracking typically requires specialized infrared cameras, depth sensors, or GPU‑accelerated processing, which limits deployment to controlled lab environments and raises privacy concerns due to data transmission.
Solution
1gaze delivers real‑time 3D gaze estimation at 100 Hz using only a standard RGB webcam and CPU‑only algorithms. By applying temporal fusion across video frames and enforcing geometric constraints, the system produces robust full‑vector gaze data, head pose, confidence scores, and ocular events without external sensors or GPU hardware. All processing occurs locally on the device, ensuring that no video data leaves the user’s machine and supporting privacy‑sensitive applications. The lightweight pipeline includes rapid calibration and operates reliably under varying lighting and motion conditions, making professional‑grade eye‑tracking accessible on commodity hardware.
Target Audience
The primary customers are developers of remote collaboration tools, telehealth platforms, and behavioral research applications that require high‑frequency, privacy‑preserving eye‑tracking on standard consumer hardware.
Features
- CPU‑only architecture achieving 100 Hz sampling on standard laptops and desktops
- Full 3D gaze vector output with off‑axis angle estimation for precise attention modeling
- Temporal fusion across multiple frames to reduce noise and handle low‑light or rapid movement
- On‑device processing with no data transmission, supporting GDPR‑compliant deployments
- Compatibility with any conventional RGB webcam; no infrared or depth sensors required
- Rapid calibration workflow enabling immediate use across diverse user groups
- Simultaneous extraction of confidence scores, head pose parameters, and ocular event detection
- Integrated acoustic event classification for contextual awareness in remote settings