Privacy Policy

Effective Date: May 22, 2026

StyleMate ("we," "our," or "us") operates the StyleMate mobile application (the "App"). This Privacy Policy explains how we collect, use, disclose, and safeguard your information when you use our App. By using StyleMate, you agree to the collection and use of information in accordance with this policy.

1. Information We Collect

1.1 Information You Provide Directly

Account registration data: phone number or email address (used for sign-in verification), and password (stored as a bcrypt hash).
Third-party login identifiers: if you sign in via WeChat, Alipay, or Apple, we receive your OpenID/User ID from the respective platform.
Profile information: nickname, avatar image, gender, birth year, height, weight, and style preferences (up to 5 tags selected from preset options).
Wardrobe data: clothing items you add (manually or via AI recognition), including category, color, material, style tags, season, pattern, fit, and associated photos.
Outfit preferences: preferred occasions, budget level (affordable/mid-range/luxury), and free-text styling instructions you provide for AI outfit generation.

1.2 Information Collected Automatically

Clothing photos: when you use the AI recognition feature, photos you take with your camera or select from your photo library are first processed on your device to detect any human faces. Detected faces are automatically blurred using on-device machine learning before the photo is transmitted. The original (unblurred) photo never leaves your device. See Section 3 for details.
Usage & analytics data: page views, feature interactions, and action events within the App. This data is used solely for improving the App experience and is not linked to your identity for advertising purposes.
Device & network information: standard HTTP request data (IP address, user agent) processed by our servers for request handling and abuse prevention.
Location data (optional): if you enable the location setting, we use your city name to fetch weather data for contextual outfit recommendations. You may disable this at any time in Privacy Settings. We do not collect or store precise GPS coordinates.

2. How We Use Your Information

To provide core App functionality: AI clothing recognition, AI outfit generation, wardrobe management, and cloud sync for paid users.
To personalize recommendations: using your style preferences, body measurements, and wardrobe data to generate relevant outfit suggestions.
To authenticate your identity: managing login sessions via JWT tokens and verifying ownership of your account.
To communicate with you: sending verification codes via SMS or email during account registration and login.
To improve our AI models and services: analyzing aggregated, anonymized usage patterns to enhance recognition accuracy and recommendation quality.
To enforce our Terms of Service: detecting and preventing abuse, fraud, or unauthorized access.

3. How We Process Your Photos

3.1 On-Device Face Anonymization

Before any photo leaves your device, StyleMate automatically detects and blurs human faces using Google ML Kit's on-device face detection model. This processing happens entirely on your phone's CPU — the original, unblurred photo is never transmitted off your device.

Detection: ML Kit identifies all human faces in the photo, along with their bounding box coordinates.
Blurring: A 25% margin is added around each detected face region, and a Gaussian blur (radius 20) is applied — rendering the face permanently unidentifiable.
Result: Only the anonymized (face-blurred) version of the photo proceeds to compression and upload. If no faces are detected, the photo is used as-is.
Privacy by design: This is an automatic, non-optional step. You do not need to enable it, and you cannot disable it — it ensures that no human face is ever sent to our servers or any third-party AI provider, whether intentionally or accidentally.

Photos you upload for clothing recognition are then handled as follows:

Photos are compressed (to a maximum of 768px, JPEG quality 80) and transmitted to our servers hosted on Microsoft Azure (United States region).
The compressed photo is base64-encoded and sent to our third-party AI providers for analysis (see Section 4).
Recognition photos are NOT permanently stored on our servers. They are held in memory only for the duration of processing and are discarded immediately after the recognition result is returned.
If you choose to save a recognized item to your wardrobe, only the AI-generated metadata (category, color, material, etc.) and a compressed copy of the photo are stored. Paid users may sync this stored photo to our cloud; free users' wardrobe photos remain only on their device.

Photos used for AI outfit visualization (generating a styled look image) are similarly transmitted to our AI provider and are not permanently stored by us. Generated outfit images are cached on your device and may be synced to our cloud if you are a paid user.

4. Third-Party Service Providers

We share certain data with the following third-party services to deliver core App functionality. All data transmission is encrypted via HTTPS.

Service Provider	Purpose	Data Shared	Location
Alibaba Cloud (qwen3-vl-flash)	AI clothing recognition from photos	Compressed & base64-encoded clothing photos	United States
Black Forest Labs / BFL (Flux 2 Klein 4B)	AI outfit visualization image generation	Text descriptions of outfits; optional reference clothing photos	United States
Alibaba Cloud / DashScope (qwen3-vl-flash)	AI outfit combination recommendations (text generation)	Clothing item attributes (category, color, material, style tags); user preferences (occasion, budget, instructions). No photos are sent.	United States
Pexels	Stock fashion photography for outfit inspiration	Search queries derived from clothing attributes (e.g., "white linen shirt casual")	United States
QWeather	Weather-based outfit recommendations	City name (not precise location)	China
Alibaba Cloud SMS	SMS verification code delivery	Phone number (solely for code delivery)	China
Firebase Authentication (Google)	Phone number sign-in verification	Phone number, SMS verification code	United States
WeChat (Tencent)	OAuth social login	Authorization code exchanged for OpenID & public profile info	China
Alipay (Ant Group)	OAuth social login	Authorization code exchanged for User ID	China
Apple	Sign in with Apple	Apple User ID	United States
126.com (NetEase) SMTP	Email verification code delivery	Email address (solely for code delivery)	China
Google ML Kit	On-device face detection for photo anonymization	No user data is sent to Google. The face detection model runs entirely on your device. Only the ML model file itself may be downloaded from Google's servers on first use.	On-device

We do not sell your personal information to third parties. We do not use your data for advertising or ad targeting. Data sharing with the above providers is strictly for providing the App's core features.

5. Data Storage & Security

Server location: All backend services are hosted on Microsoft Azure in the United States.
Data at rest: User profile data and wardrobe metadata are stored in a PostgreSQL database. Uploaded files (avatars, wallpapers, synced wardrobe photos) are stored on the server filesystem. Passwords are hashed with bcrypt.
Data in transit: All communication between the App and our servers, and between our servers and third-party providers, is encrypted using HTTPS/TLS.
Local device storage: Wardrobe data and generated outfit images are stored locally on your device. You may use the App in guest mode or as a free user without any data leaving your device (except for AI processing as described above).
Retention: We retain your account information and synced wardrobe data for as long as your account is active. Recognition photos are not retained. SMS/email verification codes expire after 5 minutes.

6. Your Privacy Choices

The App includes a Privacy Settings page where you can control:

Location access: Enable/disable city-level location for weather-based recommendations. Default: enabled.
Usage analytics: Enable/disable collection of usage data for product improvement. Default: enabled.
Personalized recommendations: Enable/disable AI personalization based on your preferences. Default: enabled.
Anonymous data sharing: Enable/disable sharing of anonymized data to improve recommendation quality. Default: disabled.

7. Your Rights

Depending on your jurisdiction, you may have the following rights regarding your personal data:

Access: You may request a copy of the personal data we hold about you.
Correction: You may update your profile information at any time within the App.
Deletion: You may request deletion of your account and all associated data by contacting us (see Section 9). Upon deletion, all your personal information will be permanently removed from our servers within 30 days.
Data portability: You may request your data in a structured, machine-readable format.
Withdraw consent: You may withdraw consent for optional data processing at any time via the Privacy Settings page. Withdrawing consent does not affect the lawfulness of processing based on consent before its withdrawal.

To exercise any of these rights, please contact us using the information in Section 9. We will respond within 30 days.

8. Children's Privacy

StyleMate is not intended for use by children under the age of 13 (or the applicable age of digital consent in your jurisdiction). We do not knowingly collect personal information from children. If we become aware that we have collected personal data from a child, we will take steps to delete it promptly.

9. Contact Us

If you have questions, concerns, or requests regarding this Privacy Policy or your personal data, please contact us:

Email: stylemate@126.com
Response time: We aim to respond to all inquiries within 5 business days.

10. Changes to This Privacy Policy

We may update this Privacy Policy from time to time. We will notify you of material changes by posting a notice within the App or by sending an email to the address associated with your account. Continued use of the App after the effective date constitutes acceptance of the updated policy.