左圖是從 ARCore ML Kit 範例中取得，以 Android Kotlin 編寫而成。這個範例應用程式使用機器學習模型將相機分類的物件分類，並將標籤附加到虛擬場景中的物件。
ML Kit API 提供 Android 和 iOS 開發服務，而 Google Cloud Vision API 同時具備 REST 和 RPC 介面，因此您可以自己使用 Android NDK (C)、iOS 或 Unity (AR Foundation) 建構的應用程式中，取得與 ARCore ML Kit 範例相同的結果。
[null,null,["上次更新時間：2025-07-26 (世界標準時間)。"],[[["\u003cp\u003eARCore's camera feed can be used with ML Kit and Google Cloud Vision API for identifying real-world objects and creating intelligent AR experiences.\u003c/p\u003e\n"],["\u003cp\u003eThe provided sample app demonstrates object classification by attaching virtual labels to identified objects.\u003c/p\u003e\n"],["\u003cp\u003eML Kit offers cross-platform support for Android and iOS, while Google Cloud Vision API provides REST and RPC interfaces for broader integration.\u003c/p\u003e\n"],["\u003cp\u003eDevelopers can utilize ARCore's data as input for their own machine learning models for object recognition.\u003c/p\u003e\n"],["\u003cp\u003eThese functionalities extend to apps built with Android NDK (C), iOS, and Unity (AR Foundation), offering flexibility in development environments.\u003c/p\u003e\n"]]],[],null,["# Machine learning with ARCore\n\nYou can use the camera feed that ARCore captures in a machine learning pipeline\nwith the [ML Kit](https://developers.google.com/ml-kit) and the [Google Cloud Vision API](https://cloud.google.com/vision) to identify real-world objects, and create an\nintelligent augmented reality experience.\nYour browser does not support the video tag.\n\nThe image at left is taken from the [ARCore ML Kit sample](https://github.com/googlesamples/arcore-ml-sample),\nwritten in Kotlin for Android. This sample app uses a machine learning\nmodel to classify objects in the camera's view and attaches a label to the object\nin the virtual scene.\n\nThe [ML Kit](https://developers.google.com/ml-kit) API provides for both Android\nand iOS development, and the [Google Cloud Vision API](https://cloud.google.com/vision)\nhas both REST and RPC interfaces, so you can achieve the same results as the\nARCore ML Kit sample in your own app built with the Android NDK (C), with iOS, or\nwith Unity (AR Foundation).\n\nSee [Use ARCore as input for Machine Learning models](/ar/develop/java/machine-learning)\nfor an overview of the patterns you need to implement. Then apply these to your\napp built with the Android NDK (C), with iOS, or with Unity (AR Foundation)."]]
