{
 "id": 69823,
 "slug": "realtime-human-pose-estimation-with-tensorflow-and-posenet",
 "url": "https://agents.willem.com/en/snippets/realtime-human-pose-estimation-with-tensorflow-and-posenet/",
 "language": "en",
 "title": "Realtime human pose estimation with TensorFlow and PoseNet",
 "subtitle": "How a pretrained model finds body joints in a video feed.",
 "summary": "PoseNet is a TensorFlow vision model that estimates a person's pose by locating key body joints, like elbows, hands, hips and knees. When I explored it in 2019, it ran in realtime on ordinary hardware because the model was already trained.",
 "topics": [
  "art",
  "data",
  "experiment",
  "game",
  "people",
  "sensors",
  "work"
 ],
 "date": "2019-12-01",
 "published": "2019-12-01T00:00:00+01:00",
 "created": "2026-08-30T07:13:38.014000+02:00",
 "updated": "2026-08-30T07:17:01.401000+02:00",
 "full_article_url": "https://willem.com/en/2019-12-01_realtime-human-pose-recognition-through-computer-vision/",
 "author": {
  "name": "Willem L. Middelkoop",
  "url": "https://willem.com"
 },
 "note": "A short distillation of a longer willem.com post. Not currently published on willem.com itself.",
 "answers_question": "How does realtime human pose recognition with TensorFlow and PoseNet work?",
 "text": "TensorFlow is an open source machine learning platform originally developed by Google. A tensor is a multi-dimensional array of numeric values, and images flow through a chain of steps, cropping, resizing, removing colour, comparing contours, from unanalysed to recognised.\nThe heavy lifting happens when models are trained, so running a trained model does not need very expensive hardware. PoseNet is such a pretrained vision model: it estimates a person's pose by finding key body joints, elbows, hands, hips, knees and ankles.\nBody detection first uses the Single Shot MultiBox Detector (SSD), a fast algorithm that shrinks the image into simplified feature maps and predicts early, which suits large objects like human bodies. With poses detected live, a person can quite literally become the game controller or drive an interactive art installation.",
 "markdown_url": "https://agents.willem.com/en/snippets/realtime-human-pose-estimation-with-tensorflow-and-posenet/snippet.md"
}