Meet segmentation from MediaPipe.

Sample and Computation Redistribution for Efficient Face Detection.

Single Shot Detector Lite for general object detection.

Robust video matting for alpha compositing.

MoveNet Lightning architecture for single body pose detection.

Realtime single-stage face detector.

Pose landmark prediction from Google MediaPipe.

High performance object detector based on YOLO series.

Lightweight object classification model optimized for mobile applications.

Hair matte segmentation.

Anchor-free ultra-fast object detection.

Ultra-high performance face detection architecture.

CenterFace anchor-free face detection.

Ultra-light fast generic face detector.

Realtime pose detection in 3D with augmented reality.

Hand pose detection from MediaPipe.

Mobile line segment detector from NAVER Vision.

Lightweight face detection from Google MediaPipe.

Pose detection from Google MediaPipe.