In a recent online webinar session, Intel mentioned the possibility of using Open3D with 400 Series cameras to combine point clouds into a "mega point cloud". A transcription of that webinar is available at the link below:
yes.. thank you. I saw this statement.
" Discussing the Open3D open source point cloud library, the possibility of combining point clouds into a "mega point cloud" was mentioned."
But at the moment, the consecutive images does not align well using open3d for us.
We get a result using plain rgb images extracted from a rosbag but no chance with depth + rgb images with open3d.
Then I think it has to be with the parameters.
There is a forum discussion about peoples; experiences of using the 400 Series with Open3D. They mention that although public datasets taken with Asus Xion seem to work fine, data that they recorded with D435 had problems.