# 01Spatial Documentation

> 01Spatial visual positioning documentation for spatial capture, mapping, localization, and AR integration.

## English documentation (draft)

> English content is available for preview and AI evaluation, but is not yet the production default.

- [Get Started](https://portal.01spatial.ai/ai-docs/en/overview.md): 01Spatial is a Space-as-a-Service platform built on our distinctive UX-first visual positioning technology (UX-first VPS). Space-as-a-Service is an innovative concept introduced by 01Spatial: visual positioning turns a museum, stadium, mall, factory, or other real venue into a locatable, programmable, and operable high-value spatial product.
- [How it works](https://portal.01spatial.ai/ai-docs/en/concepts/platform-workflow.md): 01Spatial first turns a real place into a persistent Spatial Map. Two roles then use that map for different jobs: a Space User obtains positioning, navigation, or AR / MR experiences, while a Space Owner manages maps, content, and venue operations. A Space User can be a person or a robot.
- [For AR developers](https://portal.01spatial.ai/ai-docs/en/concepts/for-ar-developers.md): AR projects span table-scale, room-scale, building-scale, campus-scale, and city-scale environments. Device SLAM can track motion at 30–60 FPS and is usually sufficient for table- and room-scale experiences. Over longer routes, small frame-to-frame errors accumulate into visible drift, so building- and campus-scale projects need VPS.
- [For robotics developers](https://portal.01spatial.ai/ai-docs/en/concepts/for-robotics-developers.md): Robotics projects usually focus on localization and monitoring rather than AR rendering. A robot can use an onboard camera or an externally mounted camera. With the intrinsics that correspond to the submitted pixels, 01Spatial returns a 6-DoF pose in the Spatial Map coordinate system.
- [Scan a space with a standard camera](https://portal.01spatial.ai/ai-docs/en/mapping/standard-camera-video/capture.md): Standard camera video is a fast way to scan a space with a phone, action camera, or pinhole camera. The goal is not to make a polished walk-through video; it is to give the mapping system a continuous sequence of sharp, overlapping views.
- [Configure standard-camera scan data](https://portal.01spatial.ai/ai-docs/en/mapping/standard-camera-video/map-configurator.md): Map Configurator loads frames extracted from the video and preserves their recorded orientation. Mark surveyed, fixed corners across several frames to give the map the correct real-world size, origin, and orientation. The video does not need to be uploaded again.
- [Build a map from standard-camera video](https://portal.01spatial.ai/ai-docs/en/mapping/standard-camera-video/build.md): After configuration, 01Spatial turns the standard-camera video into a spatial map for localization, content authoring, and AR delivery. The cloud build continues after you leave the page.
- [Scan a space with a 360° camera](https://portal.01spatial.ai/ai-docs/en/mapping/panorama-camera-video/capture.md): 01Spatial is not tied to a dedicated brand or model. Any 360° camera that can export a standard 2:1 equirectangular video through its vendor software can be used, including any Insta360 model, any GoPro MAX model, any RICOH THETA model, and cameras from other brands.
- [Use 01Spatial Reference Boards](https://portal.01spatial.ai/ai-docs/en/mapping/panorama-camera-video/spatial-reference-board.md): The 01Spatial Reference Board is a printable ChArUco pattern. It tells the system how long one meter is in the real site, where the map origin is, and which direction is up. You may use no boards or place several according to the space; the current official kit contains ten uniquely numbered boards. If boards cannot be placed, scan normally and configure surveyed GCPs (Ground Control Points) in Map Configurator after parsing. Despite the name, a GCP does not have to be on the floor.
- [Configure scan data](https://portal.01spatial.ai/ai-docs/en/mapping/panorama-camera-video/map-configurator.md): Map Configurator associates visual observations from the scan video with measured site data, giving the map a trustworthy real-world size, origin, and axis orientation. Portal loads the parsed panorama frames directly; the video is not uploaded again.
- [Build a map from 360° video](https://portal.01spatial.ai/ai-docs/en/mapping/panorama-camera-video/build.md): After scan data and spatial constraints are configured, 01Spatial turns the 360° video into a spatial map for localization, content authoring, and AR delivery. The process runs in the cloud and continues even after you leave the page.
- [Scan with Polycam](https://portal.01spatial.ai/ai-docs/en/mapping/mobile-3d-scanning/polycam.md): Polycam records both spatial imagery and device motion on a phone or tablet. 01Spatial builds a high-quality visual map from the complete Raw Data package and further refines the recorded camera trajectory. You do not need to configure GCPs manually, and a surface-model file alone is not sufficient.
- [Scan with KIRI Engine](https://portal.01spatial.ai/ai-docs/en/mapping/mobile-3d-scanning/kiri-engine.md): KIRI Engine can record scene images, camera positions, and depth information on a LiDAR-equipped iPhone or iPad. 01Spatial builds a visual map from the complete Scene Scan Raw Data ZIP and further refines the camera trajectory recorded by KIRI Engine.
- [LiDAR scanning overview](https://portal.01spatial.ai/ai-docs/en/mapping/lidar.md): A valid input must include visual panoramas, camera positions and orientations, and metric 3D geometry. A point cloud without imagery cannot produce a visual-localization map. Register and clean the project in the vendor software before export.
- [Scan with Matterport](https://portal.01spatial.ai/ai-docs/en/mapping/lidar/matterport.md): 01Spatial converts a self-contained Matterport E57—scan panoramas, camera poses, and 3D data—into a localizable spatial map.
- [Scan with FARO](https://portal.01spatial.ai/ai-docs/en/mapping/lidar/faro.md): FARO projects commonly contain multiple fixed stations. 01Spatial accepts a ZIP with one E57 per station and normalizes them into one spatial dataset.
- [Scan with NavVis](https://portal.01spatial.ai/ai-docs/en/mapping/lidar/navvis.md): NavVis supports continuous capture over large indoor spaces. 01Spatial reads spherical imagery, poses, and point-cloud data from one self-contained E57.
- [Scan with Leica RTC](https://portal.01spatial.ai/ai-docs/en/mapping/lidar/leica-rtc.md): The Leica path supports Leica RTC series projects that export a self-contained E57. 01Spatial converts each station's cube images into a panorama while preserving station poses and 3D data.
- [Scan with Hexagon RealSLAM](https://portal.01spatial.ai/ai-docs/en/mapping/lidar/hexagon-realslam.md): Hexagon RealSLAM supports mobile capture across large spaces. 01Spatial reads panoramas, poses, and point-cloud data from E57 and produces localization assets through the same quality gates.
- [Scan with XGRIDS](https://portal.01spatial.ai/ai-docs/en/mapping/lidar/xgrids.md): 01Spatial supports exported data from all XGRIDS handheld scanners, including K1, K2, and L2, and reads spherical images, poses, and point clouds from E57.
- [Scan with FJD](https://portal.01spatial.ai/ai-docs/en/mapping/lidar/fjd.md): An FJD input uses panoramas plus pano-pose-info.csv, with an optional E57 point cloud. 01Spatial validates and normalizes them as one map input.
- [Scan with CHCNAV](https://portal.01spatial.ai/ai-docs/en/mapping/lidar/chcnav.md): A CHCNAV input contains panoramas, one text pose table, and an optional LAS point cloud. Portal matches image filenames against pose records before mapping.
- [Manage maps](https://portal.01spatial.ai/ai-docs/en/maps/manage.md): Spatial Maps is the control center for creating maps, uploading scan data, monitoring builds, and opening, moving, or transferring completed maps.
- [Explore a map](https://portal.01spatial.ai/ai-docs/en/maps/explore.md): The left Map Panel in Space Studio is for inspecting the map itself: points, mesh, origin, and mapping-camera poses. For query-image validation, continue to Run a localization test in Space Studio.
- [Add content](https://portal.01spatial.ai/ai-docs/en/maps/add-content.md): The right Content Panel in Space Studio places models, primitives, lights, and annotations in map coordinates. The map describes the physical space; Content describes the digital experience anchored to it.
- [Run a localization test in Space Studio](https://portal.01spatial.ai/ai-docs/en/localization/space-studio-test.md): Use Space Studio to validate a ready map with representative camera images before integrating the localization API into an application. The test runs the same production localization service and visualizes each returned pose directly against the map.
- [Launch AR on Android](https://portal.01spatial.ai/ai-docs/en/localization/launch-ar/android-webxr.md): The Android client uses Chrome WebXR and on-device ARCore, so no separate app is required. Space Studio generates an AR link for the current map and Content.
- [Launch AR on iOS](https://portal.01spatial.ai/ai-docs/en/localization/launch-ar/ios-app-clip.md): 01Spatial uses an App Clip on iPhone. Scanning the QR code launches a lightweight ARKit client without requiring the full app, and loads the current map's web Content.
- [Launch AR in WeChat](https://portal.01spatial.ai/ai-docs/en/localization/launch-ar/wechat-mini-program.md): The WeChat Mini Program entry is intended for users in mainland China. Space Studio asks the server to generate a Mini Program code for the current map.
- [Localization Guide for AR Apps](https://portal.01spatial.ai/ai-docs/en/localization/localization-api.md): This tutorial shows how to send one camera image—or a coherent Rig.v2 frame set—to 01Spatial, read the returned camera pose, and use that pose to align AR content with the real world.
- [Build an AR app with the Unity template](https://portal.01spatial.ai/ai-docs/en/localization/templates/unity.md): The Unity integration connects 01Spatial absolute localization to AR Foundation. ARKit or ARCore provides smooth frame-to-frame tracking; 01Spatial aligns that local session with persistent map coordinates.
- [Build an AR app with the WeChat Mini Program template](https://portal.01spatial.ai/ai-docs/en/localization/templates/wechat-mini-program.md): The Mini Program integration uses XR-FRAME for local tracking and 01Spatial for an absolute camera pose in map coordinates. Persistent Content remains under a map root aligned to the current trackerSpace.
- [Credentials and keys](https://portal.01spatial.ai/ai-docs/en/reference/credentials.md): 01Spatial separates who is calling from which resource is targeted.
- [Coordinate and pose conventions](https://portal.01spatial.ai/ai-docs/en/reference/coordinate-pose.md): 01Spatial returns the camera pose in map coordinates. Let TAB transform points from frame B to frame A: M is Map, A is the current AR session, and C is Camera. API cameratoworld is TMC.
- [Errors and troubleshooting](https://portal.01spatial.ai/ai-docs/en/reference/troubleshooting.md): First isolate the failing layer: upload, build, viewing, localization request, or AR alignment. Avoid changing scan data, intrinsics, and coordinate conversion at the same time.

## 简体中文文档

- [快速开始](https://portal.01spatial.cn/ai-docs/zh-Hans/overview.md): 01Spatial 是一个基于我们独特的 UX-first VPS（以用户体验为先的视觉定位技术） 打造的 Space-as-a-Service（空间即服务）平台。Space-as-a-Service 是 01Spatial 提出的创新性概念：利用视觉定位技术，把博物馆、体育场、商场、工厂等真实空间转化为可定位、可编程、可运营的高价值空间商品。也就是说，同一个场所可以像软件一样持续增加功能、发布体验并创造运营价值。
- [它是如何工作的？](https://portal.01spatial.cn/ai-docs/zh-Hans/concepts/platform-workflow.md): 01Spatial 先把真实空间转换为一张可长期复用的空间地图，再让两类角色基于同一张地图完成不同任务：空间使用者在现场获得定位、导航或 AR / MR 体验；空间运营方管理地图与内容，并持续运营场所、设备和服务。
- [写给 AR 开发者](https://portal.01spatial.cn/ai-docs/zh-Hans/concepts/for-ar-developers.md): AR 项目的技术难度首先由空间尺度决定：
- [写给机器人开发者](https://portal.01spatial.cn/ai-docs/zh-Hans/concepts/for-robotics-developers.md): 机器人项目的重点通常不是显示 AR 内容，而是让设备在大空间中持续知道“我在哪里、朝向哪里、刚才经过哪里”。01Spatial 为机器人提供统一空间地图中的视觉定位，并为运营方保留定位记录与运动轨迹，用于导航、任务执行、监控和复盘。
- [使用普通相机进行空间扫描](https://portal.01spatial.cn/ai-docs/zh-Hans/mapping/standard-camera-video/capture.md): 普通相机视频适合使用手机、运动相机或其他非全景相机快速扫描空间。目标不是拍摄一段好看的参观视频，而是让建图系统持续看到清晰、连续且相互重叠的视角。
- [配置普通相机扫描数据](https://portal.01spatial.cn/ai-docs/zh-Hans/mapping/standard-camera-video/map-configurator.md): Map Configurator 会直接加载从视频中解析出的画面，竖屏和横屏视频都会保留原始方向。你需要在多个画面中标记现场测量的固定角点，让地图获得正确的真实大小、原点和方向。视频不需要再次上传。
- [使用普通相机视频建图](https://portal.01spatial.cn/ai-docs/zh-Hans/mapping/standard-camera-video/build.md): 完成扫描数据配置后，01Spatial 会把普通相机视频转化为可用于定位、内容编辑和 AR 发布的空间地图。建图在云端运行，关闭页面不会中断任务。
- [使用全景相机进行空间扫描](https://portal.01spatial.cn/ai-docs/zh-Hans/mapping/panorama-camera-video/capture.md): 01Spatial 不绑定专用品牌或型号。只要相机能够通过厂商软件导出标准的 2:1 等距柱状投影视频，即可用于空间扫描。支持范围包括 Insta360 任意型号、GoPro MAX 任意型号、RICOH THETA 任意型号，以及其他品牌的全景相机。
- [使用 01Spatial 空间基准板](https://portal.01spatial.cn/ai-docs/zh-Hans/mapping/panorama-camera-video/spatial-reference-board.md): 01Spatial 空间基准板（01Spatial Reference Board） 是一套可打印的定位图案，由棋盘格角点和唯一编码组合而成，技术名称为 ChArUco。它帮助系统确定地图的真实大小、原点和朝上方向。扫描前可以不放置，也可以按空间规模放置多张；当前官方套件包含 10 张唯一编号的板。
- [配置扫描数据](https://portal.01spatial.cn/ai-docs/zh-Hans/mapping/panorama-camera-video/map-configurator.md): Map Configurator 用现场实测坐标标记视频中的固定角点，从而确定地图的真实大小、原点和坐标方向。Portal 会直接加载已经解析好的全景画面，不需要再次上传视频。
- [使用全景视频建图](https://portal.01spatial.cn/ai-docs/zh-Hans/mapping/panorama-camera-video/build.md): 当扫描数据、真实大小和坐标基准准备完成后，01Spatial 会将全景视频转化为可用于定位、内容编辑和 AR 发布的空间地图。整个过程在云端完成，关闭页面不会中断任务。
- [使用 Polycam 扫描](https://portal.01spatial.cn/ai-docs/zh-Hans/mapping/mobile-3d-scanning/polycam.md): Polycam 可以在手机或平板上同时记录空间画面与设备的移动路径（相机轨迹）。01Spatial 使用完整的 Raw Data（原始数据） 数据包建立高质量视觉地图，并进一步优化这条轨迹。你不需要手工设置控制点，也不要只上传一个三维模型文件。
- [使用 KIRI Engine 扫描](https://portal.01spatial.cn/ai-docs/zh-Hans/mapping/mobile-3d-scanning/kiri-engine.md): KIRI Engine 可以在支持 LiDAR 的 iPhone 或 iPad 上记录空间画面、设备的移动路径（相机轨迹）和深度信息。01Spatial 使用 Scene Scan 的 Raw Data（原始数据）ZIP 建立视觉地图，并在建图过程中进一步优化这条轨迹。
- [LiDAR 扫描概述](https://portal.01spatial.cn/ai-docs/zh-Hans/mapping/lidar.md): 与只依赖普通视频的扫描相比，LiDAR 数据通常具备更稳定的尺度、更完整的几何和更可靠的设备轨迹。它尤其适合：
- [使用 Matterport 扫描](https://portal.01spatial.cn/ai-docs/zh-Hans/mapping/lidar/matterport.md): 01Spatial 可以直接读取 Matterport 导出的自包含 E57。E57 是扫描设备常用的交换文件，可以在一个文件中保存扫描站图像、位置和三维点云；01Spatial 会把这些数据转换为可定位的空间地图。
- [使用 FARO 扫描](https://portal.01spatial.cn/ai-docs/zh-Hans/mapping/lidar/faro.md): FARO 静态扫描项目通常包含多个扫描站。01Spatial 接受“每个扫描站一个 E57”的 ZIP，并在建图前把它们统一为同一空间数据集。E57 是扫描设备常用的交换文件，可保存站点图像、位置和三维点云。
- [使用 NavVis 扫描](https://portal.01spatial.cn/ai-docs/zh-Hans/mapping/lidar/navvis.md): NavVis 适合连续采集大范围室内空间。01Spatial 从一个自包含 E57 中读取球面全景图、相机位置和点云，并生成视觉定位地图。E57 是扫描设备常用的交换文件，可在一个文件中保存图像、位置和三维点云。
- [使用 Leica RTC 扫描](https://portal.01spatial.cn/ai-docs/zh-Hans/mapping/lidar/leica-rtc.md): 01Spatial 的 Leica 路径面向 Leica RTC 系列 等能够导出自包含 E57 的静态扫描项目。E57 是扫描设备常用的交换文件；系统会把其中每个扫描站的立方体图像合成为全景图，并保留站点位置与三维数据。
- [使用 Hexagon RealSLAM 扫描](https://portal.01spatial.cn/ai-docs/zh-Hans/mapping/lidar/hexagon-realslam.md): Hexagon RealSLAM 适合移动式大范围采集。01Spatial 读取 E57 中的全景图、相机位置和点云，并使用同一套地图质量检查生成定位资产。E57 是扫描设备常用的交换文件，可在一个文件中保存图像、位置和三维点云。
- [使用 XGRIDS 扫描](https://portal.01spatial.cn/ai-docs/zh-Hans/mapping/lidar/xgrids.md): 01Spatial 支持 XGRIDS K1、K2、L2 等各类手持式扫描仪的导出数据，并从项目 E57 中读取球面图像、相机位置和点云，适合移动式扫描形成的大范围室内地图。E57 是扫描设备常用的交换文件，可在一个文件中保存图像、位置和三维点云。
- [使用 FJD 扫描](https://portal.01spatial.cn/ai-docs/zh-Hans/mapping/lidar/fjd.md): FJD 项目使用全景图和 pano-pose-info.csv 描述扫描轨迹，并可附带 E57 点云。E57 是扫描设备常用的交换文件；在本流程中，它用于提供可选的三维点云。01Spatial 会在上传时校验这些文件，并将它们统一为一个地图输入。
- [使用 CHCNAV 扫描](https://portal.01spatial.cn/ai-docs/zh-Hans/mapping/lidar/chcnav.md): CHCNAV 输入由全景图、一个记录每张图片位置和方向的文本表，以及可选的 LAS 点云文件组成。Portal 会核对图像文件名与位置记录，防止错配的数据进入建图。
- [管理地图](https://portal.01spatial.cn/ai-docs/zh-Hans/maps/manage.md): Spatial Maps 是地图的管理入口。你可以在这里创建地图、上传扫描数据、查看建图进度，并在地图就绪后打开 Space Studio、移动或转让地图。
- [探索地图](https://portal.01spatial.cn/ai-docs/zh-Hans/maps/explore.md): Space Studio 左侧的地图面板用于查看地图本身，包括点云、网格、坐标原点，以及建图时相机的位置和方向。本篇只介绍地图浏览。定位验收请阅读使用 Space Studio 执行定位测试。
- [添加内容](https://portal.01spatial.cn/ai-docs/zh-Hans/maps/add-content.md): Space Studio 右侧的 Content Panel 用于把 3D 模型、基础几何、灯光和标注放入地图坐标系。地图负责描述真实空间，Content 描述用户将在这个空间中看到和交互的数字内容。
- [使用 Space Studio 执行定位测试](https://portal.01spatial.cn/ai-docs/zh-Hans/localization/space-studio-test.md): 在应用接入定位 API 之前，可以先用 Space Studio 和现场测试照片验证已就绪地图。测试使用与 API 相同的定位服务，并将每次定位得到的相机位置和朝向直接显示在地图中。
- [在 Android 设备开启 AR](https://portal.01spatial.cn/ai-docs/zh-Hans/localization/launch-ar/android-webxr.md): 01Spatial 的 Android 客户端使用 Chrome 的 WebXR 能力和设备上的 ARCore，不需要安装独立 App。Space Studio 会为当前地图和 Content 生成可直接打开的 AR 链接。
- [在 iOS 设备开启 AR](https://portal.01spatial.cn/ai-docs/zh-Hans/localization/launch-ar/ios-app-clip.md): iPhone 上的 01Spatial AR 体验通过 App Clip 启动。用户扫描二维码后无需先安装完整 App；App Clip 使用 ARKit 跟踪设备，并加载当前地图的 Web 内容。
- [在微信小程序开启 AR](https://portal.01spatial.cn/ai-docs/zh-Hans/localization/launch-ar/wechat-mini-program.md): 微信小程序入口适合中国大陆用户。Space Studio 会通过服务器生成当前地图的小程序码，用户用微信扫码后直接进入对应的 AR 场景。
- [AR 应用定位指南](https://portal.01spatial.cn/ai-docs/zh-Hans/localization/localization-api.md): 这篇教程介绍如何把一张相机图片，或一组来自同一段连续 AR 跟踪过程的 Rig.v2 相机帧（下文简称“帧”）发送给 01Spatial。系统会返回相机位姿，也就是相机在地图中的位置和朝向；应用再用这个结果把 AR 内容对齐到真实世界。
- [使用 Unity 模板开发 AR 应用](https://portal.01spatial.cn/ai-docs/zh-Hans/localization/templates/unity.md): Unity 模板用于把 01Spatial 绝对定位接入 AR Foundation：ARKit 或 ARCore 负责帧间平滑跟踪，01Spatial 把这段本地 AR Session 对齐到持久地图坐标。
- [使用微信小程序模板开发 AR 应用](https://portal.01spatial.cn/ai-docs/zh-Hans/localization/templates/wechat-mini-program.md): 微信小程序模板使用 XR-FRAME 提供本地跟踪，并调用 01Spatial 获取地图中的绝对相机位姿。所有长期存在的数字内容都应保存在地图坐标中，再整体对齐到当前 trackerSpace。
- [凭证与 Key](https://portal.01spatial.cn/ai-docs/zh-Hans/reference/credentials.md): 01Spatial 使用不同凭证表达“谁在请求”和“请求哪个资源”。不要把 API Key、Map Key、Project Key 和 Portal 登录令牌混为一谈。
- [坐标系与位姿约定](https://portal.01spatial.cn/ai-docs/zh-Hans/reference/coordinate-pose.md): 定位成功后，01Spatial 返回相机在地图坐标系中的位姿。正确理解矩阵方向和相机轴，是避免内容镜像、旋转或固定偏移的关键。
- [错误与故障排查](https://portal.01spatial.cn/ai-docs/zh-Hans/reference/troubleshooting.md): 先确定问题发生在哪一层：上传、建图、地图浏览、定位请求，还是 AR 对齐。不要在没有证据时同时修改扫描数据、内参和坐标转换，否则很难判断真正原因。

## Additional resources

- [Structured documentation manifest](/docs-manifest.json)
- [Complete bilingual documentation](/llms-full.txt)
- [XML sitemap](/sitemap.xml)
- [OpenAPI schema](/openapi.json)
- [Interactive API reference](/docs)
- [Human-readable Chinese documentation](https://portal.01spatial.cn/tutorials/zh-Hans)