Markus Vieth

> Does it, in C++? Or just in some compilers? I am basing my knowledge mostly on this article: https://acodersjourney.com/20-abi-breaking-changes/ , but for gcc I have confirmed it experimentally

I believe the fourth entry is set to zero/one so the behaviour is consistent with the SSE/AVX implementations of so3/se3. > But I used a customized point type where I...

Hi, this sounds like an interesting and useful addition to `UniformSampling`. However, I have a few comments: - please use `unsigned int` for `count` in `Leaf` - in the end,...

By the way (in case you did not know this): to change a pull request, you can push more commits to the branch, no need to close the pull request...

The normal distributions of the target cloud are computed in `setInputTarget`. So if you call `setInputTarget` in the initial setup, and then alternately call `setInputSource` and `align`, the normal distributions...

> I am using pcl 1.10 version now, is this version doing same as well? I am looking at the code on the `master` branch, but you can check if...

> That means "some NDT grid be able to be ignored if the NDT grid includes a few point"? I am not sure I understand what you mean > and...

Thanks for the pull request, however I am unsure how much benefit the proposed changes bring. As far as I can tell, the added casts only make the widening explicit,...

> I believe that the casting is to prevent the _product_ from overflowing, But if the multiplication is performed first, then the casting, and the overflow would happen during the...

@DingLei627 I found this issue which seems to be related: https://github.com/PointCloudLibrary/pcl/issues/3484 I would also recommend to use a newer PCL version, if possible.