This is the current implementation for CIELAB color matching which takes gaussian distance in 3D CIELAB space as a feature and creates a bounding sphere for per-object color values. We found that this approach is flawed, and can be improved by using the Mahalanobis distance to create a bounding range. More details are available in the write-up, located here: https://docs.google.com/document/d/1cTtY67oCg_8EX1lfU6srxbjPAPtC1NnpAx5RVXyFRlI/edit?usp=sharing
Closes #1128
This is the current implementation for CIELAB color matching which takes gaussian distance in 3D CIELAB space as a feature and creates a bounding sphere for per-object color values. We found that this approach is flawed, and can be improved by using the Mahalanobis distance to create a bounding range. More details are available in the write-up, located here: https://docs.google.com/document/d/1cTtY67oCg_8EX1lfU6srxbjPAPtC1NnpAx5RVXyFRlI/edit?usp=sharing