Ruse : Mobile Camera-Based Application That Attempts To Alter Photos To Preserve Their Utility To Humans While Making Them Unusable For Facial Recognition Systems

Ruse is a mobile camera-based application that attempts to alter photos to preserve their utility to humans while making them unusable for facial recognition systems.

Installation

  • Easy Method: Wait and download app from appropriate app store.
  • Download and run ios app via XCode (see Development setup for more detail)

Usage Example

App is developed as a camera-based app, allowing for the modification of faces on new camera capture or current photos on camera roll with the goal of keeping them useful for social media and human consumption while making it difficult for facial recognition systems to utilize them accurately and effectively.

This is done through a variety of methods based on previous research. Due to the limits of mobile and TensorFlow Lite, learning on the device itself is not possible—so some of the more advanced techniques are not yet possible (but research and development may yield future results.)

Instructions on usage and a full video to come with first release.

The Jupyter notebook illustrates the “arbitrary fast style” adversarial technique that is possible on mobile: 

In the long term, this technique will be applied selectively (likey to segments of the photographs), along with perlin/simplex noise generated on a per image basis, a la https://github.com/kieranbrowne/camera-adversaria.

A variety of methods are used to conceal the faces from commerical recognition systems (e.g. arbitrary file transfer, perlin noise introduction). Before saving to the camera roll or being used for online purposes, an onboard facility checks to see if faces can be detected.

The effect of these adversarial approaches may then be checked wihtout needing to have network access.

(Future versions plan on including a similar onboard estimation of how a sample recognition system fairs against the modified image (classification as opposed to merely detection.))

Development Setup

Requirements: Xcode 12

Pods installed as part of the process below: TensorFlow Lite (Swift nightly build) // GoogleMLKit // GPUImage3

  • Download ios and model (tflite models) directory
  • run “pod install” in the downloaded directory
  • Open Ruse.xcworkspace

Installation to device is left as an excercise for the reader.

R K

Recent Posts

git fetch vs git pull: How They Work and When to Use Each

Both git fetch and git pull talk to a remote repository, but they do very different things to your…

5 days ago

git cherry-pick Command: Apply Commits from Another Branch

Sometimes the change you need already exists, just on the wrong branch. A hotfix lands…

5 days ago

Best Email APIs for Secure Business Email: Why Developers Are Moving Beyond SMTP

Email is still one of the most important communication channels inside modern applications. Password resets,…

6 days ago

Nginx Commands in Linux: Start, Stop, Reload, Test, and Log

Nginx is a high-performance web server and reverse proxy trusted by some of the largest…

1 week ago

ufw Command in Linux: Manage Firewall Rules with Examples

ufw (Uncomplicated Firewall) sits on top of iptables (or nftables on newer systems) and replaces…

1 week ago

who Command in Linux: Show All Logged-In Users and Sessions

When you share a server with a team or investigate unexpected activity, the first question…

1 week ago