ResUNet-a: a deep learning framework for semantic segmentation of remotely sensed data

By Foivos I. Diakogiannis, François Waldner, Peter A. Caccetta, Chen Wu. Isprs Journal of Photogrammetry and Remote Sensing, 2019

1,958 citations recorded by the source corpus.

Scene understanding of high resolution aerial images is of great importance for the task of automated monitoring in various remote sensing applications. Due to the large within-class and small between-class variance in pixel values of objects of interest, this remains a challenging task. In recent years, deep convolutional neural networks have started being used in remote sensing applications and demonstrate state of the art performance for pixel level classification of objects. Here we present a novel deep learning architecture, \resuneta, that combines ideas from various state of the art modules used in computer vision for semantic segmentation tasks. We analyse the performance of several flavours of the Generalized Dice loss for semantic segmentation, and we introduce a novel variant loss function for semantic segmentation of objects that has better convergence properties and behaves well even under the presence of highly imbalanced classes. The performance of our modeling framework is evaluated on the ISPRS 2D Potsdam dataset. Results show state-of-the-art performance with an average F1 score of 92.9% over all classes for our best model.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/9c6eca31f311eae935e84efa6966c0165bc4e14a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

UNetFormer: A UNet-like transformer for efficient semantic segmentation of remote sensing urban scene imagery (2021, positive): “Based on encoder-decoder structure, (Diakogiannis et al., 2020; Yue et al., 2019; Zhou et al., 2018) designed different skip connections to capture more abundant context, while (Liu et al., 2018; Zhao et al., 2017b) (Shen et al., 2019) developed various decoders to retain semantic information.”
Cross-City Matters: A Multimodal Remote Sensing Benchmark Dataset for Cross-City Semantic Segmentation using High-Resolution Domain Adaptation Networks (2023, methodology): “Further, Diakogiannis et al. [7] designed an enhanced ResUNet version, ResUNet-a, with atrous convolutions for semantic segmentation of RS images.”
Multiattention Network for Semantic Segmentation of Fine-Resolution Remote Sensing Images (2020, methodology): “Meanwhile, the structural development based on residual connections is also a promising direction [25, 66-69].”
A review of deep learning methods for semantic segmentation of remote sensing imagery (2021, background): “Diakogiannis, Waldner, Caccetta, and Wu (2020) proposed ResUNet-a that uses a UNet encoder/ decoder backbone that combines residual connections, atrous convolu-tions, pyramid scene parsing pooling, and multi-tasking inference.”
Multistage Attention ResU-Net for Semantic Segmentation of Fine-Resolution Remote Sensing Images (2020, neutral): “362 ResUNet-a [2] 86.”
RS-Mamba for Large Remote Sensing Image Dense Prediction (2024, methodology): “…SS task, the compared CNN-based models include FCN [40], SegNet [41], U-Net [19], PSPNet [42], HRNet [43], MA-FCN [44], Deeplabv3 + [45], ResUNet [46], MAP-Net [47], D-LinkNet [48], and SIINet [49], the compared transformer-based models include Segformer [50] and Road-Former [10], and the…”
TransAttUnet: Multi-Level Attention-Guided U-Net With Transformer for Medical Image Segmentation (2021, neutral): “(cid:2) Multi-scalecontextapproaches: Meanwhile, severalmulti-scale context models are used as the major contenders, including Unet++ [9], R2U-Net [38] ResUNet [41], ResUNet++ [42], BCDU-Net [43], KiU-Net [37], and DoubleU-Net [36].”
A Multilevel Multimodal Fusion Transformer for Remote Sensing Semantic Segmentation (2024, methodology): “For instance, ResUNet-a was developed in [48] by stacking Red-Green-Blue (RGB) images and the DSM into synthesized inputs of four channels.”
MARE: Self-Supervised Multi-Attention REsu-Net for Semantic Segmentation in Remote Sensing (2021, methodology): “Particularly, ResUNet-a [7] uses a U-Net encoder-decoder structure, in combination with residual connections, dilated convolutions, pyramid scene parsing pooling and multitasking inference.”
Unsupervised Single-Scene Semantic Segmentation for Earth Observation (2022, background): “Research toward supervised image segmentation methods has received signiﬁcant attention in the era of deep learning that has outperformed previous methods [9]–[11].”