I am planning a project which compares the different techniques of 3D mapping of a setup. I understand that there are many algorithms to accomplish the same, but due to a non-existent hardware budget, I have to compare 3D techniques/algorithms that can be accomplished with the algorithm and datasets only. I have done a fair share of research that is still ongoing, but I was looking for assistance by people who may already have knowledge of this and can help me narrow down what I'm searching.
In summary, I am looking for 3D mapping techniques/algorithms that can be accomplished using only software and datasets in order to compare them on various performance and application parameters. Since I have to compare performance, all techniques should be applied on the same dataset.
I would appreciate any information, research papers, links, otherwise that could direct me.


You need no hardware for this :)

Check out OpenSLAM website, you'll find plenty of options for slam algorithms and dataset. A quick Google search gives you things such as the KITTI Dataset.

I particularly like this webpage to find datasets.

Most of those can be used with ROS so I would advise you to implement your comparison method in ROS. Then testing different slam algorithm and/or datasets is mainly a matter of (almost) plug and play.

| improve this answer | |
  • $\begingroup$ Thank you @Malcom. The information you provided was very helpful in creating my proposal. However, as I dive into actually implementing the algorithms on ROS, I am only concerned about how to match the algorithm to the dataset. I'm hoping I get the hang of it as I go on. $\endgroup$ – NidTiw Feb 7 '18 at 22:24
  • $\begingroup$ If that answer answered your question, don't forget to accept it :) $\endgroup$ – Malcolm Feb 7 '18 at 22:25

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.