Binocular Eye-Gaze Tracking Data
120Hz binocular gaze vector streams tracking operator visual attention, fixation points, and task intent during complex bimanual manipulation.
Quality & Precision Benchmarks
SAMPLING FREQUENCY
120 Hz Binocular
GAZE ACCURACY
< 0.5° Visual Angle
FIXATION PRECISION
0.1° Resolution
HARDWARE RIG
Tobii Pro Glasses 3
Dataset Taxonomy & Output Structure
gaze_direction_3d (x, y, z Unit Ray)
pupil_diameter_mm (Left & Right Eyes)
fixation_duration_ms (Object Attention Lock)
gaze_target_object_id (Semantic Object Latch)
Specific Type Tasks & Applications
- • Visual Attention Intent Prediction
- • Human-Robot Skill Transfer Optimization
- • AR/VR Gaze-Directed UI Interfaces
What Is Right vs What Is Wrong
| COMMON COMPETITOR ERRORS (WRONG) | BLUE PROJECTS GROUND TRUTH (RIGHT) |
|---|---|
| FAIL: Monocular low-frequency gaze estimation with > 2.5° angular drift | PASS: 120Hz binocular IR eye tracking with < 0.5° angular accuracy |
| FAIL: Uncalibrated visual fixation points missing target objects | PASS: PTP timestamped gaze vectors aligned with 4K video frames |
Files & Telemetry Data Example (Python)
import json
import numpy as np
# Load Blue Projects Egocentric Data Type: Binocular Eye-Gaze Tracking Data
with open("binocular-eye-gaze-tracking_sample.json", "r") as f:
data = json.load(f)
print("Loaded Frame Keys:", list(data.keys()))
Why Blue Projects for Binocular Eye-Gaze Tracking Data?
Request a free matched 10-sequence sample batch formatted to your exact VLA foundation model requirements.
Request Free Sample Batch →