首页 > AI前沿 > Driver Gaze Target Identification via Gaze Estimation and Structured Driving Scene Understanding

Driver Gaze Target Identification via Gaze Estimation and Structured Driving Scene Understanding

arXiv机器学习 2026-11-19 17:09 11 阅读 查看原文
用户要求将一段英文新闻/论文摘要排版成语义化的HTML片段。需要遵循以下规则: 1. 只输出HTML片段本身,不要Markdown围栏,不要解释。 2. 每个自然段落用`

`包裹,长文用`

`/`

`划分小节。 3. 列表内容用`
  • `,引用用`
    `,代码用`
    `,重要句子可用``。
    4. 保留原文所有信息,不得增删改内容。
    5. 不要输出``/``/``等文档级标签。
    
    原文内容分析:
    第一句到"...forward-view road images." 是引言/背景和提出的框架概述。
    第二句到"...class-level semantic priors." 详细描述了框架的四个组件,适合用列表或者分段,但原文是一句话带四个组件,为了符合列表规则,可以将其拆分或者保留原句。不过规则说“保留原文所有信息,不得增删改内容”,如果原句是一句话,我可以用`

    `包裹,或者如果它明显是列举,可以用列表但要注意不要改变原意和字词。原文:"The framework comprises four components: MFGaze, which combines a Vision Transformer pretrained with a masked autoencoder (MAE) and head-pose conditioning through feature-wise linear modulation (FiLM); panoptic scene understanding, which constructs a structured candidate set; polynomial ridge regression, which maps gaze features to 2D scene-image coordinates through individualized calibration; and entropy-driven target inference, which adaptively fuses boundary-distance geometry with class-level semantic priors." 这里用`

    `包裹整个句子是最安全的,避免增删改。或者可以拆分成列表,但拆分可能会被视为“修改内容”或改变句子结构。为了严格遵守“不得增删改内容”,最好保持原段落结构,用`

    `。 不过,规则提到“列表内容用`

    • `”,如果原文有列表才用。这里没有明显的列表,只是冒号后的并列。保持为段落即可。 第三句到"...self-collected gaze data, and the Look Both Ways (LBW) dataset." 实验评估部分。 第四句到"...in the laboratory experiment." 实验室实验结果。 第五句到"...on LBW." LBW数据集结果。 第六句到"...demonstrates strong cross-dataset generalization." MFGaze角度误差结果。 第七句到结尾"...for driver monitoring." 敏感性分析和结论。 可以将原文分为几个自然段落,或者保持原文的段落结构。原文实际上是一整段。我需要把它合理地划分为几个段落,或者用`

      `/`

      `来划分小节,或者就用几个`

      `。 原文没有明确的换行,是一整段。我可以按照语义拆分为几个`

      `,或者加小标题。 比如:

      研究背景与框架提出

      Identifying a driver's gaze target provides object-level information for visual attention monitoring in intelligent vehicles. Gaze-zone classification and continuous attention localization provide limited information about the identity of the attended scene target. This paper proposes a modular framework that identifies object- or region-level gaze targets from synchronized in-cabin driver images and forward-view road images.

      框架组成

      The framework comprises four components: MFGaze, which combines a Vision Transformer pretrained with a masked autoencoder (MAE) and head-pose conditioning through feature-wise linear modulation (FiLM); panoptic scene understanding, which constructs a structured candidate set; polynomial ridge regression, which maps gaze features to 2D scene-image coordinates through individualized calibration; and entropy-driven target inference, which adaptively fuses boundary-distance geometry with class-level semantic priors.

      实验评估与结果

      Experiments evaluate the framework using three public gaze datasets, a laboratory test using Cityscapes road scenes and self-collected gaze data, and the Look Both Ways (LBW) dataset. The complete framework achieves an overall Top-1 gaze target identification accuracy of 96.27% in the laboratory experiment. After sequence-specific calibration, it achieves Top-1 accuracies of 79.92% for task-relevant objects and 85.58% for the road class on LBW. MFGaze achieves the lowest mean angular errors among the compared methods, with 4.06$^\circ$, 4.23$^\circ$, and 10.52$^\circ$ on MPIIFaceGaze, ETH-XGaze, and Gaze360, respectively, and demonstrates strong cross-dataset generalization. Sensitivity analyses show stable performance under hyperparameter perturbations. These results support the framework as a source of object-level gaze information for driver monitoring.

      检查规则: 1. 只输出HTML片段本身,不要Markdown围栏、不要解释。 -> 满足 2. 每个自然段落用`

      `包裹,长文用`

      `/`

      `划分小节。 -> 满足,使用了`

      `和`

      `。 3. 列表内容用`

      • `,引用用`
        `,代码用`
        `,重要句子可用``。 -> 原文没有列表、引用、代码,可以不使用。重要句子可以用``,比如结果部分。为了自然,可以加几个``。
        4. 保留原文所有信息,不得增删改内容。 -> 小标题是我加的,这算不算“增删改内容”?规则说“长文用

        /

        划分小节”,这暗示需要加小标题来划分小节。小标题属于排版结构,不是正文内容。为了保险,小标题尽量概括,不改变正文。或者不加小标题,直接用`

        `拆分段落?“每个自然段落用

        包裹,长文用

        /

        划分小节”,说明可以使用小标题。 5. 不要输出``/``/``等文档级标签。 -> 满足。 再审视一下“不得增删改内容”。如果加了`

        `标题,标题文字本身是新增的。为了严格避免“增”内容,我可以把原文按逻辑拆分成几个`

        `,而不加`

        `,或者加`

        `但确保它只是结构标记。通常“内容”指正文文本。加小标题是排版的一部分。不过,如果完全不加小标题,只用`

        `,也是符合“每个自然段落用`

        `包裹”的。原文是一整段,我可以把它拆分成几个自然段,用`

        `包裹。 例如:

        Identifying a driver's gaze target provides object-level information for visual attention monitoring in intelligent vehicles. Gaze-zone classification and continuous attention localization provide limited information about the identity of the attended scene target. This paper proposes a modular framework that identifies object- or region-level gaze targets from synchronized in-cabin driver images and forward-view road images.

        The framework comprises four components: MFGaze, which combines a Vision Transformer pretrained with a masked autoencoder (MAE) and head-pose conditioning through feature-wise linear modulation (FiLM); panoptic scene understanding, which constructs a structured candidate set; polynomial ridge regression, which maps gaze features to 2D scene-image coordinates through individualized calibration; and entropy-driven target inference, which adaptively fuses boundary-distance geometry with class-level semantic priors.

        Experiments evaluate the framework using three public gaze datasets, a laboratory test using Cityscapes road scenes and self-collected gaze data, and the Look Both Ways (LBW) dataset. The complete framework achieves an overall Top-1 gaze target identification accuracy of 96.27% in the laboratory experiment. After sequence-specific calibration, it achieves Top-1 accuracies of 79.92% for task-relevant objects and 85.58% for the road class on LBW. MFGaze achieves the lowest mean angular errors among the compared methods, with 4.06$^\circ$, 4.23$^\circ$, and 10.52$^\