FASA: Feature Augmentation and Sampling Adaptation for Long-Tailed Instance Segmentation
This addresses data scarcity in rare object classes for instance segmentation, offering a fast and generic solution without elaborate loss design or costly inter-class transfer learning.
The paper tackles the problem of long-tailed instance segmentation by proposing FASA, a method that augments feature space for rare classes and adapts sampling to avoid over-fitting, achieving consistent performance gains with little added cost and state-of-the-art results in tasks like long-tailed classification.
Recent methods for long-tailed instance segmentation still struggle on rare object classes with few training data. We propose a simple yet effective method, Feature Augmentation and Sampling Adaptation (FASA), that addresses the data scarcity issue by augmenting the feature space especially for rare classes. Both the Feature Augmentation (FA) and feature sampling components are adaptive to the actual training status -- FA is informed by the feature mean and variance of observed real samples from past iterations, and we sample the generated virtual features in a loss-adapted manner to avoid over-fitting. FASA does not require any elaborate loss design, and removes the need for inter-class transfer learning that often involves large cost and manually-defined head/tail class groups. We show FASA is a fast, generic method that can be easily plugged into standard or long-tailed segmentation frameworks, with consistent performance gains and little added cost. FASA is also applicable to other tasks like long-tailed classification with state-of-the-art performance.