English(EN) GazePrior: Zero-Shot AR/VR Eye Tracking via Learned 3D Gaze Reconstruction

GazePrior通过合成AR/VR数据实现零样本眼球追踪

作者 PulseAugur 编辑部 · [2 个来源] · 2026-05-21 11:54

研究人员开发了GazePrior，一种用于AR/VR应用零样本眼球追踪的新方法。该方法利用学习到的3D先验来模拟人类眼球分布，从而能够合成逼真且多样化的训练数据。通过利用这些合成数据，眼球追踪模型与现有的零样本技术相比，实现了更高的准确性和鲁棒性，无需为新设备进行昂贵的真实世界数据收集。 AI

影响通过改进眼球追踪技术，实现更逼真且更具成本效益的AR/VR体验。

排序理由该集群包含一篇详细介绍眼球追踪新方法的学术论文。

AI 生成摘要 · Google Gemini · 来自 2 个来源。我们如何撰写摘要 →

报道来源 [2]

arXiv cs.CV TIER_1 English(EN) · Corentin Dumery, David Colmenares, Alexander Fix, Pascal Fua, Ali Behrooz, Jogendra Kundu · 2026-05-22 04:00

GazePrior: Zero-Shot AR/VR Eye Tracking via Learned 3D Gaze Reconstruction

arXiv:2605.22359v1 Announce Type: new Abstract: Eye tracking (ET) is a foundational technology for advanced AR/VR applications. However, training ET models for every new ET device is challenging: real data collection is costly and time-consuming, while existing synthetic data gen…
arXiv cs.CV TIER_1 English(EN) · Jogendra Kundu · 2026-05-21 11:54

GazePrior: Zero-Shot AR/VR Eye Tracking via Learned 3D Gaze Reconstruction

Eye tracking (ET) is a foundational technology for advanced AR/VR applications. However, training ET models for every new ET device is challenging: real data collection is costly and time-consuming, while existing synthetic data generation methods lack realism. To remove the need…