CORDIS OpenAire Datacite Sygma Publications Open Repository TOrino Crossref arXiv.org e-Print Archive European Union Open Data Portal https://iris.polito.it/bitstream/11583/2962571/1/ICIAP_2021_arnaudo_contrastive_distillation_camera_ready.pdf https://iris.polito.it/bitstream/11583/2962571/3/978-3-031-06430-2_62.pdf https://link.springer.com/content/pdf/10.1007/978-3-031-06430-2 https://link.springer.com/content/pdf/10.1007/978-3-031-06430-2_62 https://doi.org/10.48550/arxiv.2112.03814 Open Access 2022-05-16 2021-01-01 2022-05-03 Open Access12 pages, ICIAP 2021 Open Access12 pages, ICIAP 2021 Incremental learning represents a crucial task in aerial image processing, especially given the limited availability of large-scale annotated datasets. A major issue concerning current deep neural architectures is known as catastrophic forgetting, namely the inability to faithfully maintain past knowledge once a new set of data is provided for retraining. Over the years, several techniques have been proposed to mitigate this problem for image classification and object detection. However, only recently the focus has shifted towards more complex downstream tasks such as instance or semantic segmentation. Starting from incremental-class learning for semantic segmentation tasks, our goal is to adapt this strategy to the aerial domain, exploiting a peculiar feature that differentiates it from natural images, namely the orientation. In addition to the standard knowledge distillation approach, we propose a contrastive regularization, where any given input is compared with its augmented version (i.e. flipping and rotations) in order to minimize the difference between the segmentation features produced by both inputs. We show the effectiveness of our solution on the Potsdam dataset, outperforming the incremental baseline in every test. Code available at: https://github.com/edornd/contrastive-distillation. FOS: Computer and information sciences computer vision, semantic segmentation, aerial images, incremental learning Computer Vision and Pattern Recognition (cs.CV) Image and Video Processing (eess.IV) Computer Science - Computer Vision and Pattern Recognition FOS: Electrical engineering, electronic engineering, information engineering 0211 other engineering and technologies 0202 electrical engineering, electronic engineering, information engineering 02 engineering and technology Electrical Engineering and Systems Science - Image and Video Processing Edoardo Arnaudo, Fabio Cermelli, Antonio Tavera, Claudio Rossi, Barbara Caputo, 2022-01-01 Open Access12 pages, ICIAP 2021 Open Access12 pages, ICIAP 2021 Incremental learning represents a crucial task in aerial image processing, especially given the limited availability of large-scale annotated datasets. A major issue concerning current deep neural architectures is known as catastrophic forgetting, namely the inability to faithfully maintain past knowledge once a new set of data is provided for retraining. Over the years, several techniques have been proposed to mitigate this problem for image classification and object detection. However, only recently the focus has shifted towards more complex downstream tasks such as instance or semantic segmentation. Starting from incremental-class learning for semantic segmentation tasks, our goal is to adapt this strategy to the aerial domain, exploiting a peculiar feature that differentiates it from natural images, namely the orientation. In addition to the standard knowledge distillation approach, we propose a contrastive regularization, where any given input is compared with its augmented version (i.e. flipping and rotations) in order to minimize the difference between the segmentation features produced by both inputs. We show the effectiveness of our solution on the Potsdam dataset, outperforming the incremental baseline in every test. Code available at: https://github.com/edornd/contrastive-distillation. A Contrastive Distillation Approach for Incremental Semantic Segmentation in Aerial Images 10.48550/arxiv.2112.03814 publication