Efficient Curriculum based Continual Learning with Informative Subset Selection for Remote Sensing Scene Classification

TL;DR

This paper introduces a novel curriculum-based replay memory approach for class incremental learning in remote sensing scene classification, effectively reducing catastrophic forgetting and noise influence, and demonstrating superior performance on benchmark datasets.

Contribution

It proposes a curriculum learning strategy based on class similarity and a confident sample selection method for replay memory in CIL, addressing key limitations of existing approaches.

Findings

Significant reduction in catastrophic forgetting.

Improved stability-plasticity trade-off.

Enhanced performance on NWPU-RESISC45, PatternNet, EuroSAT datasets.

Abstract

We tackle the problem of class incremental learning (CIL) in the realm of landcover classification from optical remote sensing (RS) images in this paper. The paradigm of CIL has recently gained much prominence given the fact that data are generally obtained in a sequential manner for real-world phenomenon. However, CIL has not been extensively considered yet in the domain of RS irrespective of the fact that the satellites tend to discover new classes at different geographical locations temporally. With this motivation, we propose a novel CIL framework inspired by the recent success of replay-memory based approaches and tackling two of their shortcomings. In order to reduce the effect of catastrophic forgetting of the old classes when a new stream arrives, we learn a curriculum of the new classes based on their similarity with the old classes. This is found to limit the degree of forgetting substantially. Next while constructing the replay memory, instead of randomly selecting samples from the old streams, we propose a sample selection strategy which ensures the selection of highly confident samples so as to reduce the effects of noise. We observe a sharp improvement in the CIL performance with the proposed components. Experimental results on the benchmark NWPU-RESISC45, PatternNet, and EuroSAT datasets confirm that our method offers improved stability-plasticity trade-off than the literature.