Zihao Sun

Recent advancements in Earth observation technologies have accelerated remote sensing (RS) data acquisition, yet cross-domain semantic segmentation remains challenged by domain shifts. Traditional unsupervised domain adaptation (UDA) methods often rely on computationally intensive and unstable generative adversarial networks (GANs). This study introduces elevation-aware domain adaptation (EADA), a multi-task framework that integrates elevation estimation (via digital surface models) with semantic segmentation to address distribution discrepancies. EADA employs a shared encoder and task-specific decoders, enhanced by a spatial attention-based feature fusion module. Experiments on Potsdam and Vaihingen datasets under cross-domain settings (e.g., Potsdam IRRG → Vaihingen IRRG) show that EADA achieves state-of-the-art performance, with a mean IoU of 54.62% and an F1-score of 65.47%, outperforming single-stage baselines. Elevation awareness significantly improves the segmentation of height-sensitive classes, such as buildings, while maintaining computational efficiency. Compared to multi-stage approaches, EADA’s end-to-end design reduces training complexity without sacrificing accuracy. These results demonstrate that incorporating elevation data effectively mitigates domain shifts in RS imagery. However, lower accuracy for elevation-insensitive classes suggests the need for further refinement to enhance overall generalizability.

The performance of a semantic segmentation model for remote sensing (RS) images pre-trained on an annotated dataset greatly decreases when testing on another unannotated dataset because of the domain gap. Adversarial generative methods, e.g., DualGAN, are utilized for unpaired image-to-image translation to minimize the pixel-level domain gap, which is one of the common approaches for unsupervised domain adaptation (UDA). However, the existing image translation methods face two problems when performing RS image translation: (1) ignoring the scale discrepancy between two RS datasets, which greatly affects the accuracy performance of scale-invariant objects; (2) ignoring the characteristic of real-to-real translation of RS images, which brings an unstable factor for the training of the models. In this paper, ResiDualGAN is proposed for RS image translation, where an in-network resizer module is used for addressing the scale discrepancy of RS datasets and a residual connection is used for strengthening the stability of real-to-real images translation and improving the performance in cross-domain semantic segmentation tasks. Combined with an output space adaptation method, the proposed method greatly improves the accuracy performance on common benchmarks, which demonstrates the superiority and reliability of ResiDualGAN. At the end of the paper, a thorough discussion is conducted to provide a reasonable explanation for the improvement of ResiDualGAN. Our source code is also available.

The marine environment is becoming increasingly complex due to the various marine vehicles, and the diversity of maritime objects poses a challenge to marine environmental governance. Maritime object detection technology plays an important role in this segment. In the field of computer vision, there is no sufficiently comprehensive public dataset for maritime objects in the contrast to the automotive application domain. The existing maritime datasets either have no bounding boxes (which are made for object classification) or cover limited varieties of maritime objects. To fulfil the vacancy, this paper proposed the Multi-Category Large-Scale Dataset for Maritime Object Detection (MCMOD) which is collected by 3 onshore video cameras that capture data under various environmental conditions such as fog, rain, evening, etc. The whole dataset consists of 16,166 labelled images along with 98,590 maritime objects which are classified into 10 classes. Compared with the existing maritime datasets, MCMOD contains a relatively balanced quantity of objects of different sizes (in the view). To evaluate MCMOD, this paper applied several state-of-the-art object detection approaches from computer vision research on it and compared their performances. Moreover, a comparison between MCMOD and an existing maritime dataset was conducted. Experimental results indicate that the proposed dataset classifies more types of maritime objects and covers more small-scale objects, which can facilitate the trained detectors to recognize more types of maritime objects and detect maritime objects over a relatively long distance. The obtained results also show that the adopted approaches need to be further improved to enhance their capabilities in the maritime domain.

The performance of a semantic segmentation model for remote sensing (RS) images pretrained on an annotated dataset would greatly decrease when testing on another unannotated dataset because of the domain gap. Adversarial generative methods, e.g., DualGAN, are utilized for unpaired image-to-image translation to minimize the pixel-level domain gap, which is one of the common approaches for unsupervised domain adaptation (UDA). However, the existing image translation methods are facing two problems when performing RS images translation: 1) ignoring the scale discrepancy between two RS datasets which greatly affects the accuracy performance of scale-invariant objects, 2) ignoring the characteristic of real-to-real translation of RS images which brings an unstable factor for the training of the models. In this paper, ResiDualGAN is proposed for RS images translation, where an in-network resizer module is used for addressing the scale discrepancy of RS datasets, and a residual connection is used for strengthening the stability of real-to-real images translation and improving the performance in cross-domain semantic segmentation tasks. Combined with an output space adaptation method, the proposed method greatly improves the accuracy performance on common benchmarks, which demonstrates the superiority and reliability of ResiDuanGAN. At the end of the paper, a thorough discussion is also conducted to give a reasonable explanation for the improvement of ResiDualGAN. Our source code is available at https://github.com/miemieyanga/ResiDualGAN-DRDG.

Semantic segmentation is a critical problem for many remote sensing (RS) image applications. Benefiting from large-scale pixel-level labeled data and the continuous evolution of deep neural network architectures, the performance of semantic segmentation approaches has been constantly improved. However, deploying a well-trained model on unseen and diverse testing environments remains a major challenge: a large gap between data distributions in train and test domains results in severe performance loss, while manual dense labeling is costly and not scalable. To this end, we proposed an unsupervised domain adaptation framework for RS image semantic segmentation that is both practical and effective. The framework is supported by the consistency principle, including the cycle consistency in the input space and self-supervised consistency in the training stage. Specifically, we introduce cycle-consistent generative adversarial networks to reduce the discrepancy between source and target distributions by translating one into the other. The translated source data then drive a pipeline of supervised semantic segmentation model training. We enforce consistency of model predictions across target image transformations in order to provide self-supervision for the unlabeled target data. Experiments and extensive ablation studies demonstrate the effectiveness of the proposed approach on two challenging benchmarks, on which we achieve up to 9.95% and 7.53% improvements in accuracy over the state-of-the-art methods, respectively.