Actor Richa Chadha recently suggested a makeup trick for dodging cameras: draw eyes and lips on the face and conceal the eyebrows. The idea is playful, but it raises a serious question that many people are now asking: does makeup beat facial recognition? Short answer: not reliably, and not in a way you should bank on.
What Richa Chadha Suggested and Why It Went Viral
According to the Indian Express explainer, Chadha's tip involves painting eyes and lips onto the face and hiding the real eyebrows. The logic is to give a camera something confusing to look at, so that the features a system expects to find are in the wrong place or missing.
It spread quickly because it is simple, visual and cheap, and because many people feel uneasy about cameras that can identify them in public. A makeup tutorial feels like a personal fix for a problem that usually seems out of individual hands.
The approach belongs to a broader family of ideas often called dazzle or anti-surveillance makeup, where bold patterns are meant to disrupt automated face detection. These ideas have circulated for years, and they tend to get more attention than they deserve as practical defenses.
How Facial Recognition Reads a Face, and Why Makeup Rarely Fools It
Facial recognition generally works in stages. First, software detects that a face is present in an image or video frame. Next, it measures the face, mapping the relative positions and proportions of features such as the eyes, nose, mouth and jawline. Finally, it converts those measurements into a numerical template and compares it with templates stored in a database.
That pipeline explains why makeup is an unreliable shield:
- Systems differ. A pattern that confuses one algorithm may do nothing to another. Tools are updated constantly, so a trick that worked against an older model can fail against a newer one.
- Faces are read as a whole. Modern systems look at overall structure, not a single feature. Hiding eyebrows or adding painted features does not change bone structure, face shape or the spacing between key points.
- Detection can still succeed. Even if a system struggles to match you, a camera may still flag that a face is present. Unusual makeup can also draw human attention, which defeats the purpose in a crowd.
- Conditions matter. Lighting, camera angle and image quality all affect results, so a look that seems to work once says little about the next camera.
In short, makeup might sometimes degrade a match, but it cannot be counted on to prevent one. Treating it as protection could give a false sense of security.
Where Your Face Data Comes From: Photos, Metadata and Public Databases
The bigger issue is that most facial recognition does not depend on catching you at a lucky angle on the street. It depends on having good reference images of you in the first place.
Those images often come from places you control, at least partly:
- Social media photos. Public profile pictures and tagged images are easy sources for building or matching face templates.
- Photo metadata. Image files can carry embedded details such as the time a picture was taken and, in some cases, the location. Sharing originals without stripping that data can reveal more than the image itself.
- Public and shared databases. Images that appear in public records, event galleries or other open collections can be gathered and reused.
Once a template exists in a database, makeup at the point of capture matters much less. The data that identifies you has already been collected and stored. That is why controlling collection and storage tends to do more than changing how you look.
The handling of personal data by organizations is a recurring theme in privacy news. For a sense of how regulators respond when large amounts of personal information are exposed, see our report on how Prodhab opened its own probe into a Costa Rica data leak.
Practical Ways to Limit Biometric Tracking, Including Legal Opt-Out Rights
No single step removes you from facial recognition, but several reduce your exposure.
- Tighten social media privacy settings. Limit who can see your photos, turn off automatic face tagging where the option exists, and review which images are public.
- Strip metadata before sharing. Many phones and photo apps let you remove location and other embedded details when you share an image. Make it a habit.
- Audit your old photos. Delete or restrict images you no longer want circulating, especially clear, front-facing shots.
- Check app permissions. Disable camera and photo library access for apps that do not need it, and be cautious about services that ask for face scans.
- Use your legal rights. Depending on where you live, data protection law may let you ask what biometric data a company holds, object to its use or request deletion. In the European Union, for example, biometric data used to identify a person receives heightened protection under data protection rules. Look for the privacy contact or data protection officer listed in a company's privacy policy.
- Question face-based verification. Identity checks that rely on face scans are expanding. Our coverage of the EU age assurance report and its privacy red flags shows how debates over verifying users can touch on sensitive personal data.
What This Means For You
If you enjoy experimenting with makeup, there is no harm in it, but do not assume it hides you from a modern biometric system. Chadha's tip is a good conversation starter, not a privacy strategy.
Your stronger levers are upstream: what photos exist of you, who can access them, what they reveal through metadata and which companies are storing your face data. Those are things you can change today, and several can be enforced through the law.
Takeaways
So, does makeup beat facial recognition? Not dependably. Systems vary, they keep improving, and your reference images are often already in someone's database. Spend your effort on locking down social accounts, removing photo metadata, trimming public images, limiting app permissions and using any opt-out or deletion rights available to you.
For more on how data-protection authorities and regulators deal with personal information, start with our reports on the Prodhab probe and the EU age assurance report.
