Privacy Policy

This privacy policy applies to FaceBlur app (hereby referred to as "Application") for mobile devices that was created by Basa Studios (hereby referred to as "Service Provider") as a Commercial service. This service is intended for use "AS IS".

Information Collection and Use

When you use the Application, your files stay on your device. Your photos and videos are processed entirely on your device. Face detection, face matching, and blurring — including Gaussian, Pixelate, Emoji, and Motion Blur modes for photos, and Gaussian, Pixelate, and Emoji modes for videos — all happen locally, and your photos and videos are never uploaded to Basa Studios servers.

The types of data the app may use and why:

• Basic anonymous usage metrics (such as feature usage or processing time) to improve reliability and performance.
• Crash diagnostics to identify and fix bugs.

The Application may collect limited technical information when you download and use it. This may include:

• Your device's Internet Protocol address (e.g. IP address)
• The pages of the Application that you visit, the time and date of your visit, the time spent on those pages
• The time spent on the Application
• Crash report if the application crashes
• The operating system you use on your mobile device

Permissions We Request

To provide core features, the app may request access to:

• Photos / Media Library: so you can select photos and videos for face blurring and save the blurred result back to your gallery.

Permissions are used only for app functionality and do not change our offline file-processing model.

Face Data

What face data does the app collect?

When you open a photo or video, the app uses on-device machine learning to locate faces: Google ML Kit for photos and for videos on Android, and Apple's Vision framework for videos on iOS. The following face data is created solely to enable the blurring features:

• Photos: face bounding box coordinates — the position and dimensions of each detected face region within the photo.
• Photos: face contour points — a set of geometric outline coordinates tracing the boundary of each face, used to position blur effects accurately.
• Photos: a 96×96 pixel thumbnail crop of each detected face, shown in the in-app face selector so you can identify which faces to blur.
• Videos: the position and size of each detected face over time (sampled about 10 times per second), together with a detection quality score.
• Videos: a small cropped thumbnail image of each person detected in the video, shown in the in-app face list so you can choose which people to blur.
• Videos: a face embedding — a list of numbers computed by an on-device model (MobileFaceNet) for each face track. It is used only to recognise when the same person appears more than once in the same video, so they can be shown as a single row. Embeddings are kept in memory only, are never saved to storage, and are discarded as soon as this grouping finishes.
• The app does not identify who anyone is. Faces are never compared against faces from other photos or videos, or against any external database.

How is face data used?

• Face positions and contour points are used solely to place your chosen blur or emoji overlay on the photo or video.
• Face thumbnails are shown in the face selector only to help you identify and select faces for blurring.
• Face embeddings are used only to group appearances of the same person within the video you are editing. You can split a grouped row if the match is wrong.
• All face detection, matching, and blurring run on your device — the app never transmits face data over a network.

Will face data be shared with any third parties? Where is it stored?

No. Face data — including face positions, contour points, thumbnails, and embeddings — is never shared with any third party or sent over the internet. The face detection models (Google ML Kit and Apple's Vision framework) and the face matching model run entirely on your device and do not upload face data to Google, Apple, or any other server.

Storage: For photos, face data exists only in the app's memory (RAM) and is never written to disk. For videos, so that you can reopen the same video without analysing it again, the app saves the face positions, person groupings, and face thumbnails for each analysed video in its private cache storage on your device. This storage is not accessible to other apps, is never uploaded by the app, and is excluded from device backups (such as iCloud Backup and Android backup) and device-to-device transfers. Face embeddings are never saved. When you preview or export a blurred video, the app also writes the positions of the faces you selected to a temporary file in your device's temporary folder.

How long is face data retained?

For photos, face data is discarded when you select a new photo, clear the current photo, or close the app, and does not persist between sessions. For videos, face embeddings are discarded as soon as analysis finishes. Saved face positions and thumbnails remain in the app's private cache so analysed videos can be reopened quickly. They are deleted when you uninstall the app, and your operating system may also clear them earlier to free up space. Temporary files are cleared by your operating system from time to time.

Third Party Access

Please note that the Application utilizes third-party services that have their own Privacy Policy about handling data. Below are the links to the Privacy Policy of the third-party service providers used by the Application:

• Google Play Services
• Google Analytics for Firebase
• Firebase Crashlytics
• RevenueCat - Used for purchases and entitlement management.
• Mixpanel - May be used for anonymous event analytics.
• AppsFlyer - May be used for attribution and marketing analytics.

Contact Us

If you have any questions regarding privacy while using the Application, or have questions about the practices, please contact the Service Provider via email at thebasastudios@gmail.com.