Title | ||
---|---|---|
Class-Incremental Learning Network for Small Objects Enhancing of Semantic Segmentation in Aerial Imagery |
Abstract | ||
---|---|---|
Due to the differences in the feature distribution between classes, when the model learns in a continuous data stream, it will encounter catastrophic forgetting. The incremental learning methods have shown great potential to solve this problem. However, most existing methods based on task-incremental learning are difficult to adapt to characteristics of remote sensing scenes with few differences in appearance but large differences in features, which is not conducive to artificially distinguish task-identity document (ID). Thus, we propose a class-incremental learning (CIL) network for small objects enhancing semantic segmentation in aerial imagery. Specifically, considering the superior accuracy of the binary classifier, we propose a twin-auxiliary (TA) model that adds an auxiliary binary classification task. Then, for expansion and contraction at the edge and small object confusion problems, we introduce a diversity distillation loss, using the results of binary-classifier to constrain the multiclass segmentation results and strengthen the attention to the locations of the segmentation results that have changed. Finally, we design a conflict reduction mechanism for multihead classifier to achieve single-head prediction for CIL. Experiments demonstrate that our method has good performance on the Vaihingen and Potsdam datasets by the International Society for Photogrammetry and Remote Sensing (ISPRS), outperforming state-of-the-art (SOTA) incremental learning methods. The code will be available soon. |
Year | DOI | Venue |
---|---|---|
2022 | 10.1109/TGRS.2021.3124303 | IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING |
Keywords | DocType | Volume |
Task analysis, Data models, Remote sensing, Semantics, Learning systems, Image segmentation, Pipelines, Catastrophic forgetting, edge contraction, incremental learning, knowledge distillation, small-target confusion | Journal | 60 |
ISSN | Citations | PageRank |
0196-2892 | 0 | 0.34 |
References | Authors | |
0 | 7 |
Name | Order | Citations | PageRank |
---|---|---|---|
Junxi Li | 1 | 0 | 0.68 |
Xian Sun | 2 | 0 | 8.45 |
Wenhui Diao | 3 | 0 | 4.73 |
Peijin Wang | 4 | 4 | 2.08 |
Yingchao Feng | 5 | 0 | 1.35 |
Xiaonan Lu | 6 | 0 | 1.01 |
Guangluan Xu | 7 | 0 | 1.35 |