2
$\begingroup$

I am developing a Body Measurement extraction application, my current stage is able to extract the point clouds of human body in a standing posture, from every angles.

Now, to be able to recognize shoulders, neck point etc, my research seems to fall into following flows:

Method A:

  1. Obtain a lot of data, with labeled landmark points (shoulder left, shoulder right, neck line).
  2. Use PointCNN / PointNet++ to perform segmentation for each landmark.
  3. Once the landmarks are extracted, use Open3D / Point Cloud Library convex hull to obtain the measurement along point clouds.
    • Method A seems straight forward, but might depends on the quality of point cloud, especially 3rd step.

Method B:

  1. Obtain a lot of data, with labeled landmarks and also measurement of shoulder length, chest circumference etc.
  2. We first train the network to identify landmarks,
  3. Then from landmark, we record the distance to the next nearest point, train the network to obtain the measurement that we want.

Method C:

  1. Obtain a lot of data, with measurement of shoulder length, chest circumference etc ONLY.
  2. Pick a random point, and we record the distance to the next nearest point, train the network to obtain the measurement that we want.

My questions:

  1. How much data needed for this kind of learning?
  2. If I obtain training data from somewhere online, and later validate using my own scanned data, will that valid?
  3. Which method makes more sense?
  4. Which existing problem with solution is similar to my case? (Facial recognition?) That I can refer to it to solve my problem.

This is technically my first machine learning project, so please bear with me if my questions seems too silly.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.