AI Pose Detector

Detect 33 body keypoints from a photo or your camera and draw the pose skeleton, on your device.

AI Private · runs on your device Free
Runs locally in your browser AI processing happens on your device. Your files are not uploaded for inference.
AI model
…
Download size
… ·
Device
Checking browser support…

    Analyse human pose instantly

    Pose detection finds the joints of the human body in an image. This tool uses Google's MediaPipe Pose Landmarker to find up to four people and 33 keypoints each: nose, eyes, shoulders, elbows, wrists, hips, knees, ankles and more, each with a visibility score.

    Use cases

    • Fitness and yoga posture checks
    • Animation and 3D character reference
    • Sports movement analysis
    • Computer vision research and student projects
    • Interactive art and game prototypes

    Photo or live camera

    Upload a photo or detect live with your camera. Download the image with the skeleton drawn or all keypoint coordinates as JSON. The camera feed is analysed in your browser and is never recorded or uploaded.

    How to use AI Pose Detector

    1. Upload a photo or start the camera.
    2. The AI detects body keypoints and draws the skeleton.
    3. Review the visibility scores and points.
    4. Download the annotated image or the JSON coordinates.

    Why use this tool?

    Get pose data from an ordinary camera or photo without expensive motion capture systems.

    FAQ

    Is the camera feed recorded?

    No. Frames are analysed in your browser and discarded immediately; nothing is recorded or uploaded.

    How many people can be detected?

    Up to four people per image.

    Does it work on partly visible bodies?

    Yes, points that aren't visible get a low visibility score.

    Related tools