{
  "paper_id": "a2152deb895270b6c4c34fe1ea6fbab7",
  "paper_title": "Person Re-identification Dataset with RGB-D Camera in a Top-View Configuration",
  "paper_authors": [
    "Daniele Liciotti",
    "Marina Paolanti",
    "Emanuele Frontoni",
    "Adriano Mancini",
    "Primo Zingaretti"
  ],
  "publication_date": "",
  "publication_venue": "Video Analytics. Face and Facial Expression Recognition and Audience Measurement",
  "abstract": "Video analytics, involves a variety of techniques to monitor, analyse, and extract meaningful information from video streams. In this light, person re-identification is an important topic in scene monitoring, human computer interaction, retail, people counting, ambient assisted living and many other computer vision research. The existing datasets are not suitable for activity monitoring and human behaviour analysis. For this reason we build a novel dataset for person re-identification that uses an RGB-D camera in a top-view configuration. This setup choice is primarily due to the reduction of occlusions and it has also the advantage of being privacy preserving, because faces are not recorded by the camera. The use of an RGB-D camera allows to extract anthropometric features for the recognition of people passing under the camera. The paper describes in details the collection and construction modalities of the dataset TVPR. This is composed by 100 people and for each video frame nine depth and colour features are computed and provided together with key descriptive statistics.",
  "keywords": [
    "Person re-identification",
    "Top-view dataset",
    "RGB-D camera",
    "TVPR"
  ],
  "paper_institutions": [
    "Dipartimento di Ingegneria dell'Informazione, Università Politecnica delle Marche"
  ],
  "output_language": "zh",
  "provider": "doubao",
  "model": "ark-code-latest",
  "prompt_template": "simple",
  "extracted_at": "2026-08-16T19:46:53.489746Z",
  "template_tag": "simple",
  "summary": "### 摘要\n本文提出了一种基于RGB-D摄像头顶视配置的行人重识别数据集TVPR，旨在解决现有数据集不适用于活动监控和行为分析的问题。研究背景是视频分析中对行人重识别的需求，其应用场景包括监控、零售、环境辅助生活等。所提出的解决方案是使用顶视安装的Asus Xtion Pro Live摄像头，采集深度和彩色图像，从每帧中提取九个特征（七个基于深度的人体测量特征和两个基于颜色的HSV直方图特征），并定义TVH、TVD和TVDH描述符。数据集包含100名受试者，在23次采集会话中记录于8天，总时长约2000秒，室内场景涵盖不同光照条件。实验采用标准距离度量（L1、欧氏距离、余弦距离）并使用CMC曲线评估性能，结果表明深度描述符在排序较高时优于颜色描述符，而两者组合（TVDH）性能最佳。未来工作包括更复杂特征的融合、音频集成以及使用飞行时间传感器等。该设置减少了遮挡并保护隐私，适用于人员计数、跌倒检测等场景，其核心机制是基于深度和颜色特征的描述符组合用于匹配。\n\n### 关键要点\n- **问题与应用场景**：行人重识别在视频监控、零售分析、环境辅助生活等领域有重要应用，但现有数据集未针对顶视配置和行为分析设计。\n- **顶视配置优势**：摄像头安装在离地4米的顶部，覆盖约14.66平方米区域，能减少遮挡且不拍摄人脸，保护隐私。\n- **数据集构成**：TVPR包含100名受试者（43女57男，年龄19-36岁），在23个会话中采集，总时长约2004.894秒，覆盖8天，含不同光照条件。\n- **特征提取**：从深度图像提取7个特征（头部到地面距离、肩部到地面距离、头部表面积、头围、肩围、肩宽、胸廓前后径），从颜色图像提取2个HSV直方图特征（头部/头发和外套）。\n- **描述符定义**：TVH仅包含颜色特征，TVD仅包含深度特征，TVDH同时包含深度和颜色特征（9个特征）。\n- **数据采集流程**：受试者以平均步态沿单向行走，转身后返回，以支持划分训练集（第一次通过）和测试集（第二次通过）。\n- **性能度量**：使用CMC曲线评估识别性能，采用L1距离、欧氏距离和余弦距离三种度量。\n- **实验结果**：深度描述符（TVD）在排序大于15时优于颜色描述符（TVH），组合描述符（TVDH）的平均性能最佳，表明深度信息更具区分力。\n- **隐私与遮挡**：顶视配置避免了面部记录，减少遮挡问题，适用于人群计数和跌倒检测等任务。\n- **未来方向**：计划融合更多特征、集成音频框架、使用飞行时间传感器，以及在环境智能场景中集成系统。\n- **数据格式**：视频以原生.oni格式保存，可转换，帧率约30fps，彩色流非压缩。"
}