Multi-class segmentation under severe class imbalance: A case study in roof damage assessment

10/14/2020
by   Jean-Baptiste Boin, et al.
14

The task of roof damage classification and segmentation from overhead imagery presents unique challenges. In this work we choose to address the challenge posed due to strong class imbalance. We propose four distinct techniques that aim at mitigating this problem. Through a new scheme that feeds the data to the network by oversampling the minority classes, and three other network architectural improvements, we manage to boost the macro-averaged F1-score of a model by 39.9 percentage points, thus achieving improved segmentation performance, especially on the minority classes.

READ FULL TEXT

Please sign up or login with your details

Forgot password? Click here to reset