Continual learning for surface defect segmentation by subnetwork creation and selection
This addresses incremental learning challenges in industrial inspection by enabling segmentation of multiple defect types without retraining, though it is incremental as it builds on existing continual learning techniques.
The paper tackles the problem of catastrophic forgetting in continual learning for surface defect segmentation by introducing LDA-CP&S, which creates subnetworks for each defect type and uses LDA for selection, resulting in a mean Intersection over Union improvement by a factor of two compared to existing methods on two datasets.
We introduce a new continual (or lifelong) learning algorithm called LDA-CP&S that performs segmentation tasks without undergoing catastrophic forgetting. The method is applied to two different surface defect segmentation problems that are learned incrementally, i.e. providing data about one type of defect at a time, while still being capable of predicting every defect that was seen previously. Our method creates a defect-related subnetwork for each defect type via iterative pruning and trains a classifier based on linear discriminant analysis (LDA). At the inference stage, we first predict the defect type with LDA and then predict the surface defects using the selected subnetwork. We compare our method with other continual learning methods showing a significant improvement -- mean Intersection over Union better by a factor of two when compared to existing methods on both datasets. Importantly, our approach shows comparable results with joint training when all the training data (all defects) are seen simultaneously