Differential Treatment for Stuff and Things: A Simple Unsupervised Domain Adaptation Method for Semantic Segmentation

Wang, Zhonghao; Yu, Mo; Wei, Yunchao; Feris, Rogério; Xiong, Jinjun; Hwu, Wen-mei W.; Huang, Thomas S.; Shi, Honghui

doi:10.1109/cvpr42600.2020.01265

Cited by 242 publications

(237 citation statements)

References 28 publications

Supporting

Mentioning

209

Contrasting

Order By: Relevance

“…They perform their experiments on HSR-RSIs and multimodal remote sensing datasets, which shows the performance of LUM is improved than other stateof-the-art domain results. In addition, more recently explored methods [143]- [145] have adopted the adversarial training framework, where the feature network generates domaininvariant features to fool the discriminator that works on image-level. Another method of semantic segmentation based on unsupervised domain adaptation is pseudo-label retraining [146], which finetunes the trained model on the source images by taking high-confident predictions as pseudo ground truth for the unlabeled images.…”

Section: Summary Of Dl-based Lum Methodsmentioning

confidence: 99%

Land-Use Mapping for High-Spatial Resolution Remote Sensing Image Via Deep Learning: A Review

Zang

Cao

Wang

et al. 2021

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Land-use mapping (LUM) using high spatial resolution remote sensing images (HSR-RSIs) is a challenging and crucial technology. However, due to the characteristics of HSR-RSIs, such as different image acquisition conditions and massive, detailed information, performing LUM faces unique scientific challenges. With the emergence of new deep learning (DL) algorithms in recent years, methods to LUM with DL have achieved huge breakthroughs, which offers novel opportunities for the development of LUM for HSR-RSIs. This paper aims to provide a thorough review of recent achievements in this field. Existing high spatial resolution datasets in the research of semantic segmentation and single object segmentation are presented firstly. Next, we introduce several basic DL approaches that are frequently adopted for LUM. After reviewing DLbased LUM methods comprehensively, which highlights the contributions of researchers in the field of LUM for HSR-RSIs, we summarize these DL-based approaches based on two LUM criteria. Individually, the first one has supervised learning, semisupervised learning, or unsupervised learning, while another one is pixel-based or object-based. We then briefly review the fundamentals and the developments of the development of semantic segmentation and single object segmentation. At last, quantitative results that experiment on the dataset of ISPRS Vaihingen and ISPRS Potsdam are given for several representative models such as FCN and U-Net, following up with a comparison and discussion of the results.

show abstract

Section: Summary Of Dl-based Lum Methodsmentioning

confidence: 99%

Land-Use Mapping for High-Spatial Resolution Remote Sensing Image Via Deep Learning: A Review

Zang

Cao

Wang

et al. 2021

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

show abstract

“…More recent approaches [33,16,19,57,23,14] employed adversarial training in image level. [2,15,30,42,48,45] leveraged adversarial training to learn domain-invariant representations in feature level. There are some works [38,51,2,49,22] using self-training to mitigate the domain gap via assigning labels to the most confident samples in the target domain.…”

Section: Related Workmentioning

confidence: 99%

Content-Consistent Matching for Domain Adaptive Semantic Segmentation

Kang

Liu

et al. 2020

Lecture Notes in Computer Science

Self Cite

122

View full text Add to dashboard Cite

This paper considers the adaptation of semantic segmentation from the synthetic source domain to the real target domain. Different from most previous explorations that often aim at developing adversarialbased domain alignment solutions, we tackle this challenging task from a new perspective, i.e., content-consistent matching (CCM). The target of CCM is to acquire those synthetic images that share similar distribution with the real ones in the target domain, so that the domain gap can be naturally alleviated by employing the content-consistent synthetic images for training. To be specific, we facilitate the CCM from two aspects, i.e., semantic layout matching and pixel-wise similarity matching. First, we use all the synthetic images from the source domain to train an initial segmentation model, which is then employed to produce coarse pixel-level labels for the unlabeled images in the target domain. With the coarse/accurate label maps for real/synthetic images, we construct their semantic layout matrixes from both horizontal and vertical directions and perform the matrixes matching to find out the synthetic images with similar semantic layout to real images. Second, we choose those predicted labels with high confidence to generate feature embeddings for all classes in the target domain, and further perform the pixel-wise matching on the mined layout-consistent synthetic images to harvest the appearance-consistent pixels. With the proposed CCM, only those content-consistent synthetic images are taken into account for learning the segmentation model, which can effectively alleviate the domain bias caused by those content-irrelevant synthetic images. Extensive experiments are conducted on two popular domain adaptation tasks, i.e., GTA5− →Cityscapes and SYNTHIA− →Cityscapes. Our CCM yields consistent improvements over the baselines and performs favorably against previous state-of-the-arts.

show abstract

“…Adversarial training based methods usually learn domain-invariant feature representations to achieve adaptation, while self-training based methods usually mitigate the domain gap iteratively through various strategies. Notably, many works [33,44,45,47] integrate both of them to achieve better performance.…”

Section: Introductionmentioning

confidence: 99%

Domain Adaptive Semantic Segmentation without Source Data

You

Zhu

et al. 2021

Proceedings of the 29th ACM International Conference on Multimedia

View full text Add to dashboard Cite

Domain adaptive semantic segmentation is recognized as a promising technique to alleviate the domain shift between the labeled source domain and the unlabeled target domain in many real-world applications, such as automatic pilot. However, large amounts of source domain data often introduce significant costs in storage and training, and sometimes the source data is inaccessible due to privacy policies. To address these problems, we investigate domain adaptive semantic segmentation without source data, which assumes that the model is pre-trained on the source domain, and then adapting to the target domain without accessing source data anymore. Since there is no supervision from the source domain data, many self-training methods tend to fall into the "winner-takesall" dilemma, where the majority classes totally dominate the segmentation networks and the networks fail to classify the minority classes. Consequently, we propose an effective framework for this challenging problem with two components: positive learning and negative learning. In positive learning, we select the class-balanced pseudo-labeled pixels with intra-class threshold, while in negative learning, for each pixel, we investigate which category the pixel does not belong to with the proposed heuristic complementary label selection. Notably, our framework can be easily implemented and incorporated with other methods to further enhance the performance. Extensive experiments on two widely-used synthetic-to-real benchmarks demonstrate our claims and the effectiveness of our framework, which outperforms the baseline with a large margin. Code is available at https://github.com/fumyou13/LDBE.

show abstract

Differential Treatment for Stuff and Things: A Simple Unsupervised Domain Adaptation Method for Semantic Segmentation

Cited by 242 publications

References 28 publications

Land-Use Mapping for High-Spatial Resolution Remote Sensing Image Via Deep Learning: A Review

Land-Use Mapping for High-Spatial Resolution Remote Sensing Image Via Deep Learning: A Review

Content-Consistent Matching for Domain Adaptive Semantic Segmentation

Domain Adaptive Semantic Segmentation without Source Data

Contact Info

Product

Resources

About