How to Deploy Cloud-Based Face Recognition with Alibaba AI Services

Introduction

Face recognition is a widely used biometric technology with strong potential across many industries. Traditional self‑hosted solutions are costly and complex, but cloud services now allow rapid integration without training or deploying deep‑learning models.

1. Deployment Tutorial

Overall Architecture

The solution uses Function Compute to build a web application, integrates the Vision AI Face‑Body service for face matching, and stores images in an OSS bucket. When a user uploads a face image, it is saved to OSS; a second image is then compared against the stored one, and a similarity score is returned.

The technical components are:

Vision AI Open Platform – Face‑Body service

One Function Compute application to call the Face‑Body API

One OSS bucket for storing uploaded images

Application Deployment

1. Enable Vision AI Face‑Body Service

Log in to the Vision AI platform, accept the service agreement, and activate the free trial.

2. Create an OSS Bucket

Create a bucket to store static resources, following the OSS console steps.

2. Application Experience

After integration, the API automatically detects the largest face in each image and returns a similarity score, enabling fast and accurate face matching for scenarios such as identity verification, customer identification, access control, and security monitoring.

Access Example Application

Find the demo site URL in the environment details and open it.

Perform Face Matching

Upload a reference face image, which is stored in OSS, then upload a second image to compare. The service calculates the similarity and displays the result.

For more details, refer to the official Vision AI and OSS documentation links.