Pose Detection

Creating models to detect key points on an object, or perform pose detection.

Written by Mohamed Traore

Last published at: May 18th, 2022

What is Pose Detection?

Pose detection is the combination of position and orientation relative to a specified object.

A person standing while holding a book, with an overlayed output of the inferred detections from a human pose (keypoint) model.A person standing while holding a book, with an overlayed output of the inferred detections from a human pose (keypoint) model. Source

Identifying the Position of Features with Keypoint Detection

Pose detection tasks are commonly accomplished by creating keypoint detection models.

Identifying the Position of Features with Object Detection

Alternatively, if keypoint detection is not an option for you, object detection may do the trick if you are more focused on identifying the locations (position) of specific features, rather than the orientation.

  1. Create unique classes for each feature you would like to identify
  2. Consistently label the features of interest, and all other classes of interest for your model to detect.
  3. Train your model with Roboflow Train or with one of our other custom object detection model training options
  4. Deploy your model: Roboflow Deploy