How to send image for processing (Object Detection) to firebase Ml-kit and retrieve output?

Question

我正在尝试使用 firebase 提供的用于 ML-kit android 开发的应用程序，GitHub 链接https://github.com/firebase/mlkit-material-android在这个项目中，他们说您必须修改 SearchEngine 类实现，但是我无法确定我必须更改的所有内容以及代码的外观。

Answer 1

Here are something you can do:

1. In the ML Kit sample app's SearchEngine class, you can get the imageBytes of the detected object: https://github.com/firebase/mlkit-material-android/blob/master/app/src/main/java/com/google/firebase/ml/md/java/productsearch/SearchEngine.java#L66

2. Then you can send that image to your own product search backend. One way to build such a backend is using https://cloud.google.com/vision/product-search/docs You will be able to find more details in the Product-Search API documentation and tutorials. There is also a google group for that https://groups.google.com/g/cloud-vision-discuss , where you can ask Product-Search related questions and find answers.