Fix underflow issue when computing normalization#1344
Merged
alecjacobson merged 1 commit intodevfrom Nov 8, 2019
Merged
Conversation
jdumas
reviewed
Nov 8, 2019
| if(norm == 0) | ||
| norm = ma*sqrt(norm); | ||
| // These are probably over kill; ma==0 should be enough | ||
| if(ma == 0 || norm == 0 || norm!=norm) |
Collaborator
There was a problem hiding this comment.
Maybe use std::isnan() or std::isfinite() instead of checking if norm != norm?
Collaborator
|
Could this be written as a one-liner using a combination of |
Contributor
Author
|
yes, it could. |
4 tasks
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Heat Geodesics first computes a scalar potential field. Values far from the source will be extremely small, approaching the limits of floating point numbers. Then the method takes the per triangle gradient and normalizes this vector (divides the vector by the norm).
The gradient vector will have tiny values. (e.g., [1e-300 2e-300 3e-300]). Computing the vector norm in the obvious way will fail. It's more robust to first divide by the coordinate with the max absolute and then take the norm.