遇见数据集

Outerview/global-potholes-dataset

收藏
Hugging Face2026-04-09 更新2026-04-12 收录
官方服务:

资源简介:

--- license: cc-by-4.0 pretty_name: Outerview Global Potholes Dataset task_categories: - image-classification - object-detection task_ids: - multi-class-image-classification - vehicle-detection tags: - computer-vision - image-classification - geospatial - mapping - earth-observation - urban-analytics - infrastructure - open-data - street-view - potholes - road-damage - transportation - road-safety annotations_creators: - Paul Wynter source_datasets: - original language: - en size_categories: - 10K<n<100K --- # Outerview Global Potholes Dataset A large-scale geospatial dataset of potholes with latitude and longitude. This dataset is part of Outerview’s mission to organize the world’s physical infrastructure and make it searchable. --- ## 🌍 Overview - **Feature:** Potholes - **Scope:** Global - **Entries:** 29,000 - **Total Dataset:** 3M+ potholes (full dataset) - **Formats:** Parquet / CSV / GeoJSON This dataset represents a **subset of a larger 3M+ pothole dataset** generated and maintained by Outerview. Each row corresponds to a real-world location. --- ## 📊 Dataset Schema | Column | Description | |--------------|------------| | id | Unique identifier | | latitude | Latitude coordinate | | longitude | Longitude coordinate | | region | Administrative region | | source | Data source | --- ## 🧠 Data Sources & Labeling - **Imagery Source:** Mapillary - **Label Generation:** Outerview AI models All pothole detections were generated using Outerview’s API. --- ## 🧪 Example Use Cases - Train computer vision models for road damage detection - Build geospatial search and indexing systems - Urban infrastructure monitoring - Autonomous navigation and mapping systems - Large-scale earth observation modeling --- ## 🚀 About Outerview Outerview is a research lab focused on building world models that help solve some of the world’s toughest challenges. We’ve built a world model trained on billions of images, videos, and location data that allows anyone to search physical features of the world. --- ## 🔗 API & Full Dataset Access This is a sample dataset. The full platform provides: - Millions of additional locations - **Timestamps / dates for each capture** - Access to **billions of real-world images and videos** - Real-time querying of physical features Access the full dataset and API: 👉 https://outerview.ai View API documentation: 👉 https://outerview.ai/developers/docs --- ## 🔄 Updates This dataset is actively maintained and updated **biweekly** with new data and improvements. --- ## 💬 Feedback We’d love feedback on this dataset. If there are other infrastructure features you’d like to see (e.g. drainage, signage, utilities, hazards), let us know. --- ## 📜 License This dataset is released under the **CC-BY-4.0 license**. Free for research and commercial use with attribution. --- ## ⚠️ Notes - This dataset is a sampled subset of a much larger system - The full dataset contains additional attributes such as timestamps - Coverage and density may vary by region ---

--- 许可证:CC BY 4.0 友好名称:Outerview全球路面坑洼数据集 任务类别: - 图像分类(image classification) - 目标检测(object detection) 任务子项: - 多类别图像分类(multi-class image classification) - 车辆检测(vehicle detection) 标签: - 计算机视觉(computer vision) - 图像分类(image classification) - 地理空间(geospatial) - 测绘(mapping) - 地球观测(earth observation) - 城市分析(urban analytics) - 基础设施(infrastructure) - 开放数据(open data) - 街景(street view) - 路面坑洼(potholes) - 路面破损(road damage) - 交通(transportation) - 道路安全(road safety) 标注创作者: - 保罗·温特(Paul Wynter) 源数据集: - 原创数据集 语言: - 英语 规模类别: - 10K < n < 100K --- # Outerview全球路面坑洼数据集 这是一个包含经纬度信息的大规模地理空间路面坑洼数据集。 本数据集隶属于Outerview的愿景:整理全球实体基础设施并实现其可检索性。 --- ## 🌍 数据集概览 - **特征:** 路面坑洼(potholes) - **覆盖范围:** 全球 - **条目数:** 29000条 - **完整数据集规模:** 超300万个路面坑洼(完整数据集) - **数据格式:** Parquet / CSV / GeoJSON 本数据集为Outerview生成并维护的超300万条路面坑洼完整数据集的子集。 每一行对应一个真实世界的地理位置。 --- ## 📊 数据集架构 | 列名 | 描述 | |--------------|------------| | id | 唯一标识符 | | latitude | 纬度坐标 | | longitude | 经度坐标 | | region | 行政区域 | | source | 数据来源 | --- ## 🧠 数据来源与标注 - **影像来源:** Mapillary - **标注生成:** Outerview人工智能模型 所有路面坑洼检测结果均通过Outerview的应用程序编程接口(API)生成。 --- ## 🧪 典型应用场景 - 训练用于路面破损检测的计算机视觉模型 - 构建地理空间检索与索引系统 - 城市基础设施监测 - 自主导航与测绘系统 - 大规模地球观测建模 --- ## 🚀 关于Outerview Outerview是一家专注于构建世界模型以解决全球最棘手挑战的研究实验室。 我们构建了一个基于数十亿张图像、视频与位置数据训练而成的世界模型,可让任何人检索全球实体地理特征。 --- ## 🔗 API与完整数据集访问 本数据集为样本数据集。 完整平台提供: - 数百万个额外地理位置 - 每条采集记录的时间戳与日期 - 数十亿张真实世界图像与视频的访问权限 - 实体地理特征的实时查询 访问完整数据集与API:👉 https://outerview.ai 查看API文档:👉 https://outerview.ai/developers/docs --- ## 🔄 数据更新 本数据集得到积极维护与更新,每两周发布一次新数据并进行功能改进。 --- ## 💬 反馈建议 我们期待收到关于本数据集的反馈意见。 若您希望看到其他基础设施特征相关数据(例如排水设施、标识标牌、公用设施、安全隐患等),欢迎告知我们。 --- ## 📜 许可证 本数据集采用**CC BY 4.0许可证**发布。 可免费用于研究与商业用途,但需注明原作者。 --- ## ⚠️ 注意事项 - 本数据集为更大规模系统的采样子集 - 完整数据集包含时间戳等额外属性 - 各区域的数据集覆盖范围与密度可能存在差异

提供机构:
Outerview
二维码
社区交流群
二维码
科研交流群
商业服务