3

I have a set of .ply files containing 3D points of an image.

The images are frame with particular postures. Each .ply file represent a frame.

By using those .ply files, I want to train a ML algorithm in order to predict any image as one of the given in the train data.

I tried using KNN algorithm but was not able to train it.

I do not have any labelling on data such as these points resemble hand, face, etc.

I am currently using plyfile library and Python to perform these analysis.

I came to know about PCL library as well but was not able to install it for Python.

I am stuck as how to proceed with the current dataset. Can someone guide me.

I will provide further details, if required.

David de la Iglesia
  • 2,436
  • 14
  • 29
mrsan22
  • 727
  • 2
  • 11
  • 28
  • 1
    Could you give me a little more information about the content of each `ply` file?. Like what do you mean with particular postures? The content of each file is a point cloud that represents a human in a certain posture?. Where do this 3D points come from? Wich kind of sensor did you use it to capture?. I would be happy to help you as I'm working on a python library for managing point clouds but I need more information to understand what you are trying to accomplish here – David de la Iglesia Apr 28 '16 at 18:42

0 Answers0