arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2607.04860cs.CVcs.HC
收藏

PAGE:迈向实用的人类水平注视目标估计

PAGE: Towards Practical Human-level Gaze Target Estimation

Tsinghua University(清华大学);Jinan University(暨南大学)

Zhoutong Ye, Chengwen Zhang, Zhaibin Cui, Mingze Sun, Jiaqi Liu, Xiangwu Li, Qingyang Wan, Chang Liu, Xutong Wang, Huan-ang Gao, Yu Mei, Chun Yu, Yuanchun Shi

AI总结:

提出PaGE模型明确建模场景与头部特征交互,以大ViT-H+骨干模型为教师,在更大更多样未标注数据集上蒸馏轻量级学生模型,提升注视估计性能,缩小人机差距且便于实际部署。

摘要:

注视目标估计对于理解人类注意力和意图至关重要,现有模型难以达到人类水平。为此提出PaGE模型,明确建模场景与头部特征的复杂交互。以大ViT-H+骨干的PaGE模型为教师,在更大更多样未标注数据集上蒸馏学生模型。改进架构和训练方法使PaGE在多个任务中达先进性能,缩小人机差距,蒸馏学生模型性能好且便于实际部署。代码和模型检查点可在项目页面获取。

英文摘要:

Gaze target estimation, the task of predicting where a person is looking in a scene, is crucial to understanding human attention and intent. It is a challenging task that combines high-level understanding of global scene semantics and precise spatial reasoning using human appearance (e.g. pose, eye orientation). As a result, human-level performance remains elusive for existing models, limiting their practical application. To this end, we propose PaGE (Practical Gaze Estimator), a gaze estimation model that explicitly models the complex interaction between scene and head features. Using a PaGE model with a large ViT-H+ backbone as the teacher, we further distill student models with lighter backbones on a much larger and more diverse unlabeled dataset. The architectural improvements and novel training recipe allow PaGE to achieve state-of-the-art performance on several gaze estimation tasks, outperforming humans in 7 out of 9 metrics while reducing the human-AI gap by at least 60% in the remaining 2. The distilled student models retain most of the teacher's performance while being lightweight enough for practical deployment on robots and consumer devices. The code and model checkpoints are available at our project page.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑