-
Notifications
You must be signed in to change notification settings - Fork 964
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
order changed after decode? #40
Comments
Yes you are right, the order of the points is not preserved for the options that you selected. Try setting the speed options to: anything below 10 currently uses kD-tree compression method which provides better compression but is not order preserving. 10 will use a simple delta + entropy coding on the input points, which is an order preserving approach but the compression ratio is going to be worse. |
Thanks. In fact I do not care about the order in practice, so maybe I could try nn search to test the accuracy. |
It would be really useful if Draco can export vertices order correspondence between uncompressed mesh and origin one. For example, skinning weight data will be useless if the order is lost. |
@mlenta |
off-topic: @saedrna can you say something about compression rate differences between Draco Pointcloud and LAZ format? |
@themmes Draco wins, but is lossy. |
Hi,
I am intended to test the accuracy loss of draco encoding and decoding. That is encode a point cloud to buffer and then decode the buffer to point cloud. Then check the difference between the two point clouds (both are saved in col-major Eigen matrix). There should be some accuracy loss, because I have used the qunatization to 14 bits.
However, it seems the result are not correct. Maybe the order has changed after decoding?
Below is my code,
the main function
and here is the related functions
The text was updated successfully, but these errors were encountered: