Face⁺⁺ (Megvii Technology Artificial Intelligence Open Platform) is a computer vision capability platform for developers and enterprise users, providing multiple types of recognition and processing services for faces, bodies, text, and images, helping applications integrate visual AI capabilities.
I. Main Capabilities The platform's capabilities cover the following categories: Face recognition and related analysis: face detection (supports detecting one or multiple faces, and supports analysis of 7 types of facial attributes such as age, gender, and emotion), face comparison (determines whether two face images belong to the same person, supports face feature comparison across all age groups), face search (finds similar faces within a face database, supports databases with millions of faces), beauty rating, emotion recognition, facial attributes, skin analysis (detects skin tone, wrinkles, acne, and other skin conditions, with up to 23 dimensions of analysis), facial feature analysis, 3D face reconstruction, facial landmarks, dense landmarks, album clustering, and aging prediction. Portrait processing: face fusion (one-click fusion of a face image and a template image, with no limit on the number of templates), beautification (skin smoothing, whitening), face shaping (multi-dimensional enhancement such as bigger eyes, slimmer face, and smaller face), filters (up to 35 filter options), stickers, and virtual makeup try-on. Body recognition: body detection (supports attribute queries such as gender and upper/lower garment color), body attributes, body skeleton points (single or multiple people, up to 22 points), body matting (hair-strand-level segmentation), and gesture recognition (supports up to 19 gestures). Text recognition: custom template text recognition, ID card quality detection, ID card recognition, driver's license recognition, vehicle registration certificate recognition, bank card recognition, business license recognition, and general text recognition, supporting formatted text recognition for cards and certificates, express delivery slips, bills, and more. Image recognition: license plate recognition (including license plate number and color, supports detection in nighttime environments), dog nose print recognition (dog identity recognition and comparison), and scene and object recognition.
II. Solutions Typical solutions include FaceID face verification, Beauty Cloud, FaceStyle virtual makeup try-on, FaceStyle intelligent skin testing, and FaceID Enterprise Edition identity management. Industry solutions cover the financial industry, online marketing, the mobile phone industry, the automotive industry, and the education industry.
III. Usage and Integration Methods Public cloud APIs can call Face⁺⁺ capabilities online and support integration into various applications such as mini programs, mobile H5, PC H5, and apps. Online/offline licensed SDK services can invoke capabilities in real time for business scenarios involving video streams on mobile and PC. In addition, product services such as private deployment and server-side SDK integration are also provided. The platform offers diverse services via Web API and SDK, supports cloud integration and offline deployment, with a minimum of 1The service can be integrated and used in 0 minutes. The platform also provides a free testing service, as well as resources such as pre-sales support, help documentation, and technical documentation.
IV. Applicable Scenarios Based on the above capabilities, it can be used in business scenarios such as identity verification, facial feature comparison, portrait beautification and shaping, human pose analysis, and card and document recognition across finance, marketing, education, and other fields.
V. Technical and Compliance Information The platform adopts the self-developed deep learning framework MegEngine, and is stated to have won 40 world-class computer vision competition championships and set a record of three consecutive COCO Challenge wins; it has millisecond-level processing speed, supports high-concurrency capacity and high-availability guarantees; it serves developers and enterprise users in more than 220 countries and regions worldwide. The platform provides documents such as the "Megvii SDK/API Compliance and Security Guide" and the "Initiative for the Correct Use of Artificial Intelligence Products."
VI. Contact Information Address: S1/N6, Jinyu Intelligent Manufacturing Factory, No. 27 Jiancaicheng Middle Road, Haidian District, Beijing; Telephone: 15001180685 (weekdays 9:00-18:00); consultation and suggestions email: zixunjianyi@megvii.com.
VII. Limitations and Information Not Currently Provided Regarding specific charging methods and prices, the page only lists a "Price" entry and does not provide specific price information; limits such as the number of calls and quotas for each capability are not currently provided; the platform is not suitable for tasks outside the scope of its capabilities.
