参考文档
使用集合让一切井井有条
根据您的偏好保存内容并对其进行分类。
本页面提供了有关用于训练模型的 BigQuery 表的指南,包括其架构和字段说明。您还将看到数据结构的核心概念(快照、资产和观测结果)的细分,以及有关如何在相关情况下使用表格的详细信息。
snapshots
快照是数据集在特定时间点的固定且不可更改的副本。快照表提供与快照关联的元数据,以便您从宏观层面了解数据的时态。
列名 |
类型 |
说明 |
snapshot_id |
STRING |
快照的唯一标识符。用作联接表的键。 |
subscription_id |
STRING |
相应订阅的唯一标识符。 |
creation_time |
TIMESTAMP |
采用 ISO 8601 格式的时间戳,例如 2019-09-25 17:26:27.757171 UTC 。 |
all_observations
“观测”是指在现实世界中看到资产。观测数据与素材资源的不同之处在于,前者包含用于检测素材资源的元数据,即拍摄图片的时间和相机的位置。
all_observations
表提供来自所有快照的观测结果。您可以使用此表来检测快照之间观测结果的差异。
列名 |
类型 |
说明 |
snapshot_id |
STRING |
快照的唯一标识符。用作联接表的键。 |
asset_id |
STRING |
相应资产的唯一标识符。 |
asset_type |
STRING |
资产的主要分类,例如ASSET_CLASS_ROAD_SIGN 。 |
location |
STRUCT |
包含纬度/经度坐标(以浮点数表示)的结构体。 |
detection_time |
TIMESTAMP |
采用 ISO 8601 格式的时间戳,例如 2019-09-25 17:26:27.757171 UTC 。 |
observation_id |
STRING |
唯一标识观测结果的字符串。 |
bbox |
STRUCT |
包含用于对齐素材资源边界框的 x/y 坐标的结构体。 |
camera_pose |
STRUCT |
包含纬度/经度、海拔(以米为单位)、俯仰角、航向和横滚角的浮点数的结构。 |
capture_time |
TIMESTAMP |
采用 ISO 8601 格式的时间戳,例如 2019-09-25 17:26:27.757171 UTC 。 |
gcs_uri |
STRING |
托管图片的 Google Cloud Storage URI。 |
map_url |
STRING |
显示观测位置的 Google 地图网址。 |
all_assets
“资产”是指现实世界中的对象。all_assets
表提供来自所有快照的资源。您可以使用此表检测快照之间资源存在的差异。
列名 |
类型 |
说明 |
asset_id |
STRING |
相应资产的唯一标识符。 |
snapshot_id |
STRING |
快照的唯一标识符。用作联接表的键。 |
asset_type |
STRING |
资产的主要分类,例如ASSET_CLASS_ROAD_SIGN 。 |
observation_id |
STRING |
唯一标识观测结果的字符串。 |
location |
STRUCT |
包含纬度/经度坐标(以浮点数表示)的结构体。 |
detection_time |
TIMESTAMP |
采用 ISO 8601 格式的时间戳,例如 2019-09-25 17:26:27.757171 UTC 。 |
latest_observations
latest_observations
表仅提供来自最新快照的观测结果。
列名 |
类型 |
说明 |
snapshot_id |
STRING |
快照的唯一标识符。用作联接表的键。 |
asset_id |
STRING |
相应资产的唯一标识符。 |
asset_type |
STRING |
资产的主要分类,例如ASSET_CLASS_ROAD_SIGN 。 |
location |
STRUCT |
包含纬度/经度坐标(以浮点数表示)的结构体。 |
detection_time |
TIMESTAMP |
采用 ISO 8601 格式的时间戳,例如 2019-09-25 17:26:27.757171 UTC 。 |
observation_id |
STRING |
唯一标识观测结果的字符串。 |
bbox |
STRUCT |
包含用于对齐素材资源边界框的 x/y 坐标的结构体。 |
camera_pose |
STRUCT |
包含纬度/经度、海拔(以米为单位)、俯仰角、航向和横滚角的浮点数的结构。 |
capture_time |
TIMESTAMP |
采用 ISO 8601 格式的时间戳,例如 2019-09-25 17:26:27.757171 UTC 。 |
gcs_uri |
STRING |
托管图片的 Google Cloud Storage URI。 |
map_url |
STRING |
显示观测位置的 Google 地图网址。 |
latest_assets
latest_assets
表仅提供最新快照中的资源。
列名 |
类型 |
说明 |
asset_id |
STRING |
相应资产的唯一标识符。 |
snapshot_id |
STRING |
快照的唯一标识符。用作联接表的键。 |
asset_type |
STRING |
资产的主要分类,例如ASSET_CLASS_ROAD_SIGN 。 |
observation_id |
STRING |
唯一标识观测结果的字符串。 |
location |
STRUCT |
包含纬度/经度坐标(以浮点数表示)的结构体。 |
detection_time |
TIMESTAMP |
采用 ISO 8601 格式的时间戳,例如 2019-09-25 17:26:27.757171 UTC 。 |
如未另行说明,那么本页面中的内容已根据知识共享署名 4.0 许可获得了许可,并且代码示例已根据 Apache 2.0 许可获得了许可。有关详情,请参阅 Google 开发者网站政策。Java 是 Oracle 和/或其关联公司的注册商标。
最后更新时间 (UTC):2025-08-31。
[null,null,["最后更新时间 (UTC):2025-08-31。"],[],[],null,["# Reference\n\nThis page provides a guide to the BigQuery tables used for training models,\nincluding their schemas and field descriptions. You'll also find a breakdown of\nthe core concepts---snapshots, assets, and observations---that structure the data,\nalong with details on how to use the tables where relevant.\n\n`snapshots`\n-----------\n\nA snapshot is a fixed, unchangeable copy of a dataset at a specific moment in\ntime. The snapshot table provides metadata associated with snapshots, to\nallow you to understand the temporal state of the data at a high level.\n\n| Column name | Type | Description |\n|-------------------|-----------|----------------------------------------------------------------------|\n| `snapshot_id` | STRING | Unique identifier for the snapshot. Used as a key to join tables. |\n| `subscription_id` | STRING | Unique identifier for the subscription. |\n| `creation_time` | TIMESTAMP | ISO 8601 formatted timestamp e.g., `2019-09-25 17:26:27.757171 UTC`. |\n\n`all_observations`\n------------------\n\nAn \"observation\" is a sighting of an asset in the real world. Observations\ndiffer from assets in that they contain metadata for the detection of an asset,\ni.e., the time the image was captured and the position of the camera.\n\nThe `all_observations` table provides observations from all snapshots. You can\nuse this table to detect differences in observations between snapshots.\n\n| Column name | Type | Description |\n|------------------|-----------|---------------------------------------------------------------------------------------|\n| `snapshot_id` | STRING | Unique identifier for the snapshot. Used as a key to join tables. |\n| `asset_id` | STRING | Unique identifier for the asset. |\n| `asset_type` | STRING | Major classification of the asset e.g., `ASSET_CLASS_ROAD_SIGN`. |\n| `location` | STRUCT | Struct containing lat/lng coordinates as floats. |\n| `detection_time` | TIMESTAMP | ISO 8601 formatted timestamp e.g., `2019-09-25 17:26:27.757171 UTC`. |\n| `observation_id` | STRING | String that uniquely identifies the observation. |\n| `bbox` | STRUCT | Struct of structs containing the x/y coordinates aligning the asset's bounding box. |\n| `camera_pose` | STRUCT | Struct containing floats for lat/lng, altitude (in meters), pitch, heading, and roll. |\n| `capture_time` | TIMESTAMP | ISO 8601 formatted timestamp e.g., `2019-09-25 17:26:27.757171 UTC`. |\n| `gcs_uri` | STRING | Google Cloud Storage URI where the image is hosted. |\n| `map_url` | STRING | Google Maps URL that shows the location of the observation. |\n\n`all_assets`\n------------\n\nAn \"asset\" is an object in the real world. The `all_assets` table provides\nassets from all snapshots. You can use this table to detect differences in\nassets between snapshots.\n\n| Column name | Type | Description |\n|------------------|-----------|----------------------------------------------------------------------|\n| `asset_id` | STRING | Unique identifier for the asset. |\n| `snapshot_id` | STRING | Unique identifier for the snapshot. Used as a key to join tables. |\n| `asset_type` | STRING | Major classification of the asset e.g., `ASSET_CLASS_ROAD_SIGN`. |\n| `observation_id` | STRING | String that uniquely identifies the observation. |\n| `location` | STRUCT | Struct containing lat/lng coordinates as floats. |\n| `detection_time` | TIMESTAMP | ISO 8601 formatted timestamp e.g., `2019-09-25 17:26:27.757171 UTC`. |\n\n`latest_observations`\n---------------------\n\nThe `latest_observations` table provides observations only from the most recent\nsnapshot.\n\n| Column name | Type | Description |\n|------------------|-----------|---------------------------------------------------------------------------------------|\n| `snapshot_id` | STRING | Unique identifier for the snapshot. Used as a key to join tables. |\n| `asset_id` | STRING | Unique identifier for the asset. |\n| `asset_type` | STRING | Major classification of the asset e.g., `ASSET_CLASS_ROAD_SIGN`. |\n| `location` | STRUCT | Struct containing lat/lng coordinates as floats. |\n| `detection_time` | TIMESTAMP | ISO 8601 formatted timestamp e.g., `2019-09-25 17:26:27.757171 UTC`. |\n| `observation_id` | STRING | String that uniquely identifies the observation. |\n| `bbox` | STRUCT | Struct of structs containing the x/y coordinates aligning the asset's bounding box. |\n| `camera_pose` | STRUCT | Struct containing floats for lat/lng, altitude (in meters), pitch, heading, and roll. |\n| `capture_time` | TIMESTAMP | ISO 8601 formatted timestamp e.g., `2019-09-25 17:26:27.757171 UTC`. |\n| `gcs_uri` | STRING | Google Cloud Storage URI where the image is hosted. |\n| `map_url` | STRING | Google Maps URL that shows the location of the observation. |\n\n`latest_assets`\n---------------\n\nThe `latest_assets` table provides assets only from the most recent snapshot.\n\n| Column name | Type | Description |\n|------------------|-----------|----------------------------------------------------------------------|\n| `asset_id` | STRING | Unique identifier for the asset. |\n| `snapshot_id` | STRING | Unique identifier for the snapshot. Used as a key to join tables. |\n| `asset_type` | STRING | Major classification of the asset e.g., `ASSET_CLASS_ROAD_SIGN`. |\n| `observation_id` | STRING | String that uniquely identifies the observation. |\n| `location` | STRUCT | Struct containing lat/lng coordinates as floats. |\n| `detection_time` | TIMESTAMP | ISO 8601 formatted timestamp e.g., `2019-09-25 17:26:27.757171 UTC`. |"]]