Researchers at the University of North Carolina have discovered a way that hackers could use photos from your Facebook account to bypass facial recognition technology.

The team of researchers collected photos online from 20 volunteer subjects, then created 3D models of the volunteers’ faces based off the photos. They then added facial animations and manipulated the subjects’ eyes in the photos so they stared straight ahead. The researchers even filled in parts of the volunteers’ faces that may have been hidden in the photo, plus added shadow and texture. Using these 3D models, the researchers were able to fool four out of five security systems between 55 percent and 85 percent of the time.

“Such VR-based spoofing attacks constitute a fundamentally new class of attacks that point to a serious weakness in camera-based authentication systems,” the researchers wrote. “Unless they incorporate other sources of verifiable data, systems relying on color image data and camera motion are prone to attacks via virtual realism.”

Perhaps most disturbingly of all, it only took a few low-resolution photos for the researchers to create these detailed facial replications. You can help protect yourself by hiding your photos from public view — but even that isn’t guaranteed to beat such an advanced hack.