Suppose a rainy image I can be decomposed into 3 parts: background B which is what we need, a residual layer (rain layer) R and an edge layer E to store details. Mathmatically, it can be formulated below:
To better decompose, as it's the case in many papers, we construct four constraints:
- Consistency:
- Smooth constraint for background:
- constraints for rain layer:
- constraints for edge layer:
In DiG-CoM, the parameter is set adaptively according to a metric for rain streaks below
An overflow of optimization can be seen below
In matlab, running demo.m will provide derained results in folder results/, the correspond rainy images are in folder imgs
RAINY
CLEAN
