GazeEval：用眼动追踪评估 AI 生成视频的视觉真实感
Evaluating Visual Realism of AI-Generated Videos with Eye Tracking
AI 导读
一项研究将眼动追踪作为 AI 生成视频人类评价的补充信号，开展了两次视觉图灵测试（VTT）实验，共 90 名参与者，在 438 个视频上收集了 4,286 个有效眼动试验数据。研究发现注视模式与显式人类反馈一致，并据此提出首个计算方法 GazeEval，从眼动数据预测感知视觉真实感，准确率达 71.3%。研究数据已公开以支持后续研究。
正文 · 原文
Abstract:Human evaluation of AI-generated videos has so far relied on explicit human feedback, such as judgments or ratings. While these signals capture the conclusions viewers reach, they fail to offer insights into the perception that leads to them. This paper investigates eye tracking as a complementary human evaluation signal for AI-generated videos. We conducted two Visual Turing Test (VTT) experiments with a total of 90 participants on videos generated by latest generative models, in which participants watched each video twice and then gave their feedback. Across the experiments, we collected 4,286 valid eye-tracking trials on 438 videos. Our analysis indicates that gaze patterns offer informative evaluation signals that align with explicit human feedback. Building on the findings, we propose the first computational method GazeEval to predict the perceived visual realism of AI-generated videos from eye-tracking data. It achieves 71.3% accuracy in prediction, providing evidence that eye tracking data carries informative signals about how people evaluate a video. We release the study data to support further research on eye tracking for evaluating AI-generated video.
| Subjects: | Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV) |
| Cite as: | arXiv:2610.38825 [cs.HC] |
| (or arXiv:2610.38825v1 [cs.HC] for this version) | |
| https://doi.org/10.48550/arXiv.2610.38825 arXiv-issued DOI via DataCite |
Submission history
From: Danqing Shi [view email]
[v1]
Wed, 28 Oct 2026 20:20:43 UTC (15,355 KB)
来源：arXiv cs.AI · arxiv.org