Is it possible to train a convolutional neural network (CNN) to predict the dimensions of primitive objects such as (spheres, cylinders, cuboids, etc.) from point clouds?

The input to the CNN will be the point cloud of a single object and the output will be the dimensions of the object (for example, radius and height of the cylinder). The training data will be the point cloud of the object with the ground truth dimensions in a regression final layer?

I think it is possible for images since it will similar to a bounding box detection, but I am not sure with point clouds.

  • $\begingroup$ How exactly do the points relate to the object? Are they merely all contained within the object? Are they only around the surface of the object? Do they define the surface of the object? Are they random/approximate, or exact? $\endgroup$ – The Guy with The Hat Nov 15 '19 at 20:37

I'm working on a similar problem. I'm using a 2D point cloud of an object, for example, X and Y coordinates for height, and with that more simple data set I will train a regression model (currently working on that). In my opinion, this approach with dissecting complex point cloud into cross sections that contain wanted dimension and feeding that to the model will be more simple and easier for the training.

|improve this answer|||||
  • $\begingroup$ I know that this doesn't answers your question directly, but I hope it helps and I don't have enough points to comment on your question. $\endgroup$ – Ermin Podrug Nov 15 '19 at 20:28

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.