Unverified50% confidenceFactExact time
The FaceRuler app performs face inference on the browser side, using MediaPipe FaceLandmarker (478 keypoints) compiled to WASM, with a model of about 3.6 MB; the inference stage generates no third-party requests, and photos never leave the browser tab.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/14/2026
First Seen
Valid until: 1/12/2027
Sources
Related Entities
Related Claims
UnverifiedFace detection nodes based on InsightFace or MediaPipe first locate face bounding boxes, then perform img2img redrawing on each face individually, and finally merge the results back into the original image.71% similarUnverifiedWith the maturation of open-source projects such as FaceSwap and DeepFaceLab, and the emergence of face-swapping tools based on Stable Diffusion, some tools can generate highly credible dynamic face-swapping videos with just a single photo.69% similarUnverifiedPopular face feature extraction models include FaceNet, ArcFace, and Dlib's ResNet model67% similarUnverifiedQuietGlass uses Apple's Vision framework to run a face detection model locally in real time to identify the number of faces in view.67% similarUnverifiedThe complete dataset is available through the Hugging Face dataset page datasocial/tiktok-5.6B-videos.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1029583API
curl https://kongchang.com/api/v1/knowledge/claims/1029583MCP
get_claim(id=1029583)