A keypoint is one labelled coordinate the model locates, such as a wrist or an eye corner. Collections of keypoints form skeletons and face meshes; their movement over frames is what fitness counters and AR effects read.

Book cover: AI for Video, Volume 6 of the Fora Soft Video Engineering Handbook

The book · Volume 6 of 7

AI for Video: A Complete Guide to Intelligence in Video Products: From Models and Inference to Computer Vision, Speech, Search, and Generative Video

Keypoint accuracy decides whether a fitness counter counts reps correctly or an AR effect sticks to a face. This volume covers keypoint models in its computer vision section: pose estimation with MediaPipe, RTMPose and ViTPose, face landmarks, tracking keypoints over time, and on-device deployment. Written by Nikolay Sapunov, CEO at Fora Soft.

Read it on Amazon →Kindle & paperback · Fora Soft Video Engineering Handbook

Need keypoint tracking that stays accurate during fast motion?

Send us sample footage, and we will compare keypoint and pose models on accuracy, jitter and frame rate on your target devices. 250+ video products shipped since 2005.

Book a 30-min call →WhatsApp →Email us →