tencent cloud

Face Recognition
Accurate and real-time face detection, analysis, recognition, and search services

Powered by Tencent YouTu's world-leading facial analysis technology, Tencent Cloud Face Recognition features facial detection and analysis, feature localization, search, comparison, verification, deduplication and face liveness detection. Tencent Cloud Face Recognition offers high-performance facial recognition services to developers and companies in various use cases such as smart retail, smart communities, online entertainment, smart building, online identity verification and more, meeting diverse facial recognition and customer identity verification demands across multiple industries.

Accurate Recognition

Face Recognition has industry-leading accuracy and is record-breaking in international competitions. It achieved 99.80% accuracy in LFW 2017 and ranked 1st in terms of first-attempt recognition rate (83.29%) in the MegaFace million-scale dataset.

Leading Algorithms

Based on the 3rd generation Tencent YouTu Grandmother Model, Face Recognition was optimized by various training methods such as metric learning, transfer learning, and multi-task learning. Its customized fine-tuning or distilling models can meet performance and latency requirements in different use cases.

High Stability and Reliability

Face Recognition was tested by Tencent products' massive user base and complex use cases and is proven to be highly reliable and robust with an over 99.9% availability rate.

Real-Time Response

Face Recognition features high concurrence, high throughput, and low latency. It processes millions of faces in just a few hundred milliseconds, satisfying your real-time usage demands.

Ease of Use

Face Recognition is equipped with a diverse set of online APIs and offline SDKs for easy application. You can easily call the APIs to use it on the cloud or integrate the SDK in your applications and hardware devices. You can also use the offline SDKs and online APIs together to customize a device + cloud solution to meet your needs in different scenarios.

Wide Variety of Use Cases

Face Recognition is used in a wide variety of use cases, such as online photo albums, smart retail, face recognition access control and attendance, face recognition login, face effects, online proctoring and more.

Face Recognition

Face Recognition features a rich set of face recognition capabilities, including face detection, face attribute extraction, facial key point localization, face comparison, large-scale face search, and face deduplication.

Flexible Usage Options

Face Recognition provides online APIs and offline SDKs options to meet your diverse business needs.


Call volume and QPS statistics are collected and can be queried for easy data analysis.


Facial recognition-based access control can be installed in places requiring higher security levels, where unauthorized intruders will trigger alarms upon entry.

Face analysis technology can quickly and accurately analyze people attributes such as age and gender in images, which helps targeted advertising.

With the face detection and search technologies, Face Recognition can sort similar faces in an album and easily categorize photos by face.

With the facial feature localization technology, Face Recognition can accurately locate facial features for use in beauty filters, face reshaping, cartoon stylization, face switching, and face stickers.


Face Recognition is billed by the number of calls using a tiered pricing model. If the monthly total number of API calls reaches a tier, all calls will be billed at the unit price in the tier. The higher the tier, the lower the unit price. The monthly bill will be generated between the 1st to 3rd day of the next month, and fees will be automatically deducted from your account balance. View more.