New dataset tackles pupil segmentation with ambient light

Researchers have introduced AmbientEye, a new dataset designed to advance eye-tracking technology for applications like smart glasses. This dataset focuses on pupil segmentation using only ambient infrared illumination, unlike previous systems that required active IR light sources. AmbientEye comprises over 2.6 million eye images captured outdoors under natural sunlight, featuring annotations refined by both automatic and human annotators. Initial benchmarking revealed a significant performance drop for state-of-the-art algorithms in this ambient-light setting, highlighting the challenges and the need for further research in this practical scenario. AI

IMPACT Enables development of more practical eye-tracking for AR/VR by removing reliance on active IR illumination.

RANK_REASON The cluster contains a new academic paper introducing a dataset for a specific research problem. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• AmbientEye
• arXiv

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →