Terms-Visual Intelligence API(VIAPI)-阿里云帮助中心

This topic describes the basic concepts of the Alibaba Cloud Visual Intelligence API to help you understand and use the service.

Consulting Service

If you have questions about the Alibaba Cloud Visual Intelligence API, you can join our DingTalk support group. Search for group ID 23109592 to connect with our team. The group is open to all enterprise users, service providers, and developers.

Region

A region is the physical location where a service is hosted. You can select a region based on factors such as cost and the location of your users. A region that is closer to your users generally provides faster access speeds. The recommended region for the Visual Intelligence API is China (Shanghai). Additional regions will be supported in the future.

Endpoint

An endpoint is the public domain name for accessing the Visual Intelligence API. The service is provided through HTTP RESTful APIs. Each region has a unique endpoint, and endpoints also vary based on the AI capability. For more information, see Endpoints.

AccessKey

An AccessKey is an AccessKey pair used for identity verification. It consists of an AccessKey ID and an AccessKey secret. The Visual Intelligence API uses the AccessKey ID and AccessKey secret with symmetric encryption to authenticate the request sender.

The AccessKey ID identifies the user.
The AccessKey secret is the key used to encrypt and verify the signature string.

Important Keep your AccessKey pair secure. If an AccessKey pair is compromised, delete it and create a new one immediately.

You can obtain an AccessKey for the Visual Intelligence API from the following sources:

An AccessKey that belongs to the service owner.
An AccessKey granted to a third party by the service owner through RAM.
An AccessKey granted to a third party by the service owner through Security Token Service (STS).

For more information, see Create an AccessKey.

Frequently asked questions

How do I check the Visual Intelligence services I have activated?

Log on to the Alibaba Cloud console and go to the Visual Intelligence API Service Management page to view the list of services you have purchased or activated, including services such as face comparison.

Do enterprise accounts need to provide additional materials to activate Visual Intelligence services?

No additional materials are required. Both enterprise-authenticated and personal-authenticated accounts can directly activate and use the services. No supplementary documentation is needed for activating services such as face recognition.

What alternatives are available now that the Image Recognition feature has been discontinued?

The original Image Recognition feature has been discontinued and has no dedicated replacement API. You can use Alibaba Cloud Model Studio to obtain visual recognition results by crafting prompts. Note that you need to independently evaluate and verify the accuracy of the results for your specific use case.

SDK and development integration

UniApp SDK Support

Financial-grade face detection and similar services currently do not provide a UniApp SDK. Only server-side APIs and native Android/iOS SDKs are supported. If you are developing with UniApp, you can integrate the official SDK by implementing a native plugin, or use the H5 liveness detection solution as an alternative.

Resolving Java SDK Dependency Conflicts

If you encounter compilation errors (package or method not found) after adding the facebody20191230 SDK dependency, the issue is typically caused by an outdated SDK version. To resolve it:

Upgrade the SDK to the latest version and update your code to align with the latest sample code.
If errors persist, clean the Maven build cache by running mvn clean and rebuild the project.
Check and update the underlying dependencies, including okhttp, okio, and tea-openapi.

Algorithm Model Clarification

The Image Recognition and other capabilities provided by the Visual Intelligence API use dedicated product algorithm interfaces, not general-purpose AI models. These interfaces are purpose-built for specific visual recognition tasks.