Isometric drawing in aqua of a face swap, one face laid over another on a plate, riding a belt from a webcam into the Moveris machine. Three cards leave the back of the machine for three stations: real or fake, the human-likelihood score, and the explanation.

A face swap can look like a person. It still has to work like one.

Moveris reads one second of standard webcam video and decides whether a living face was in front of the camera. It is trained and tested on face swaps. Use it at onboarding, on video calls, and on the video files people send you.

Book a demo

What is a face swap?

A face swap is a deepfake that lays another person's face over real video. Software redraws the face frame by frame, while a real person underneath supplies the movement and the voice. The tools now run on an ordinary laptop, so a swap can happen during the call itself.

A head with another face lifted onto it.

Where face swaps show up

At the selfie step

Someone opens an account under another person's face, fed into the check as if it came from a webcam.

On a video call

A stand-in joins a job interview, a help desk call, or a payment approval wearing someone else's face.

In a recorded file

A video sent with a claim or an application shows a face that was swapped before anyone pressed send.

A face swap fed in through a virtual camera is an injection attack. Played on a phone held up to a webcam, it is a screen replay. Moveris is trained and tested on both.

Why a face swap is hard to spot

A swap is built on a real person. It blinks when that person blinks and turns when that person turns, so it can do what a selfie screen asks. The face looks right to a reviewer, and deepfake tools get a new version every few days.

What a face swap has to copy

A brain and a body run a living face, so its blood flow, muscle movement, and emotion work together. We call that cognitive coherence. A deepfake can copy any one signal, but not how they all work together. Even high-quality face swaps fall out of sync.

Moveris reads one second of video. It reads each of a hundred regions of the face four ways: facial muscle movement, blood flow, estimated depth, and neighbor contrast. Muscle movement goes down to the microexpressions that carry emotion. Moveris checks how all of those signs work together.

Isometric drawing in cyan of one region of a face lifted as a small chip with an orange outline, its place in the face's grid below it. Lines run from the chip to four tiles: a moving sheet for facial muscle movement, an orange ramp for blood flow, stacked purple layers for estimated depth, and a grid with an orange center for neighbor contrast.

Isometric drawing in cyan of three short tracks, each bringing a card to a station: the living-face mark for the decision, a score bar against a threshold, and a table of pathway tests for what affected the score.

What your team gets back

Each check returns a decision, living face or not, and a human-likelihood score from 0 to 100. A breakdown shows what affected each score, so your fraud and review teams see more than the decision.

Built on our own data

Moveris is trained and tested on face swaps, lip-sync, and AI-generated video. We generate, collect, and curate 90% of this data in-house, so it matches what happens in the real world.