Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Transformation Parameters' details in ICP Registration #9590

Closed
yiyangd opened this issue Aug 9, 2021 · 7 comments
Closed

Transformation Parameters' details in ICP Registration #9590

yiyangd opened this issue Aug 9, 2021 · 7 comments

Comments

@yiyangd
Copy link

yiyangd commented Aug 9, 2021

Required Info
Camera Model D435i
Firmware Version 5.12.13.50
Operating System & Version Linux (Ubuntu 18.04)
Kernel Version (Linux Only) 5.4.0-77-generic
Platform PC
SDK Version 2.44.0
Language open3d/python
Segment others

Issue Description

Hi,

I am trying to simulate the example below with my RealSense D435i Depth Camera, I would like to know how could I understand the parameters for initial transformation's parameters and how to define a set of transformation on my own experiment?

My own experiments (a chair with wall) is a little bit mess:
Screenshot from 2021-08-09 10-15-12

Thanks!

@MartyG-RealSense
Copy link
Collaborator

Hi @yiyangd I must emphasize that I am not an Open3D specialist. I have researched your question carefully though and will do my best to help you with the information that you require.

ICP registration and its use with Python is introduced in a tutorial guide (linked to below) in the Basics section of the Open3D documentation.

http://www.open3d.org/docs/0.9.0/tutorial/Basic/icp_registration.html

An extract from this guide about initial alignment is shown below.

image

In regard to using Open3D and ICP with a RealSense camera, the discussion below should be a good entry point into the subject.

isl-org/Open3D#362

As mentioned at the beginning of this case, my knowledge of Open3D is limited but I will be happy to work through any further questions that you have that arise from the information resources highlighted above.

@yiyangd
Copy link
Author

yiyangd commented Aug 11, 2021

Hi MartyG,

Thanks, so now my biggest problem is I didn't know the parameters of init_trans arrays, I guess I need to set one since I try to capture my two point clouds in different positions?

@MartyG-RealSense
Copy link
Collaborator

You could try using the trans_init values from the earlier example.

image

The Open3D tutorial in the link below suggests though a method for calculating trans_init values with user-provided correspondences.

http://www.open3d.org/docs/latest/tutorial/visualization/interactive_visualization.html#registration-using-user-correspondences

image

@MartyG-RealSense
Copy link
Collaborator

Hi @yiyangd Do you require further assistance with this case, please? Thanks!

@yiyangd
Copy link
Author

yiyangd commented Aug 18, 2021

Hi @MartyG-RealSense,I just got some important information from here, and thank you for your information.

Hi @yiyangd Do you require further assistance with this case, please? Thanks!

@MartyG-RealSense
Copy link
Collaborator

Thanks very much @yiyangd for the update. If you are satisifed with the outcome of this case and do not require further assistance, please feel free to use the Close issue button beneath the comment writing box to close the case. Thanks!

@yiyangd yiyangd closed this as completed Aug 19, 2021
@yiyangd yiyangd reopened this Aug 19, 2021
@yiyangd
Copy link
Author

yiyangd commented Aug 19, 2021

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants