Filters








33,999 Hits in 5.9 sec

Learning to adapt class-specific features across domains for semantic segmentation [article]

Mikel Menta, Adriana Romero, Joost van de Weijer
2020 arXiv   pre-print
In this thesis, we present a novel architecture, which learns to adapt features across domains by taking into account per class information.  ...  Additionally, the feature adaptation step often happens globally, at a coarse level, hindering its applicability to tasks such as semantic segmentation, where details are of crucial importance to provide  ...  CONCLUSIONS In this thesis, we have tackled the unsupervised domain adaptation problem for semantic segmentation.  ... 
arXiv:2001.08311v1 fatcat:xeqowbv2bzb5vaitwfuqszlj4a

Meta-Learned Feature Critics for Domain Generalized Semantic Segmentation [article]

Zu-Yun Shiau, Wei-Wei Lin, Ci-Siang Lin, Yu-Chiang Frank Wang
2021 arXiv   pre-print
We propose a novel meta-learning scheme with feature disentanglement ability, which derives domain-invariant features for semantic segmentation with domain generalization guarantees.  ...  How to handle domain shifts when recognizing or segmenting visual data across domains has been studied by learning and vision communities.  ...  By advancing meta-learning strategies for [13] , we apply feature disentanglement models to learn domain-invariant content features across multiple source domains.  ... 
arXiv:2112.13538v1 fatcat:qsmcpc5d7vc43kv3fzevyh7abu

Learning Transferrable Knowledge for Semantic Segmentation with Deep Convolutional Neural Network [article]

Seunghoon Hong, Junhyuk Oh, Bohyung Han, Honglak Lee
2015 arXiv   pre-print
Contrary to existing weakly-supervised approaches, our algorithm exploits auxiliary segmentation annotations available for different categories to guide segmentations on images with only image-level class  ...  To make the segmentation knowledge transferrable across categories, we design a decoupled encoder-decoder architecture with attention model.  ...  The attention model provides not only predictions for localization but also category-specific information that enables us to adapt the decoder trained in source domain to target domain.  ... 
arXiv:1512.07928v1 fatcat:p6kdgj7gbvdrrgxzrdhss275k4

Learning Transferrable Knowledge for Semantic Segmentation with Deep Convolutional Neural Network

Seunghoon Hong, Junhyuk Oh, Honglak Lee, Bohyung Han
2016 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
Contrary to existing weakly-supervised approaches, our algorithm exploits auxiliary segmentation annotations available for different categories to guide segmentations on images with only image-level class  ...  To make segmentation knowledge transferrable across categories, we design a decoupled encoder-decoder architecture with attention model.  ...  The attention model provides not only predictions for localization but also category-specific information that enables us to adapt the decoder trained in source domain to target domain.  ... 
doi:10.1109/cvpr.2016.349 dblp:conf/cvpr/HongOLH16 fatcat:obemojsyvvfflodeaxd3ptfyse

Distribution regularized self-supervised learning for domain adaptation of semantic segmentation

Javed Iqbal, Hamza Rawal, Rehan Hafiz, Yu-Tseh Chi, Mohsen Ali
2022 Image and Vision Computing  
This paper proposes a novel pixel-level distribution regularization scheme (DRSL) for self-supervised domain adaptation of semantic segmentation.  ...  In a typical setting, the classification loss forces the semantic segmentation model to greedily learn the representations that capture inter-class variations in order to determine the decision (class)  ...  Our work is related to semantic segmentation, domain adaptation, and self-supervised learning.  ... 
doi:10.1016/j.imavis.2022.104504 fatcat:dtn6qgbr4jayfecrvm6mzko3sa

FCNs in the Wild: Pixel-level Adversarial and Constraint-based Adaptation [article]

Judy Hoffman, Dequan Wang, Fisher Yu, Trevor Darrell
2016 arXiv   pre-print
Global domain alignment is performed using a novel semantic segmentation network with fully convolutional domain adversarial learning.  ...  In this paper, we introduce the first domain adaptive semantic segmentation method, proposing an unsupervised adversarial approach to pixel prediction problems.  ...  In this work, we propose the first unsupervised domain adaptation method for transferring semantic segmentation FCNs across image domains.  ... 
arXiv:1612.02649v1 fatcat:3ku35eodhvcwxpboh4hrggnq3y

Contextual-Relation Consistent Domain Adaptation for Semantic Segmentation [article]

Jiaxing Huang, Shijian Lu, Dayan Guan, Xiaobing Zhang
2020 arXiv   pre-print
Recent advances in unsupervised domain adaptation for semantic segmentation have shown great potentials to relieve the demand of expensive per-pixel annotations.  ...  Specifically, CrCDA learns and enforces the prototypical local contextual-relations explicitly in the feature space of a labelled source domain while transferring them to an unlabelled target domain via  ...  [19] first applies adversarial learning for UDA based semantic segmentation by aligning feature space at global scale.  ... 
arXiv:2007.02424v2 fatcat:kby6leqi2zdkrniapzwrtnlxny

Cluster-to-adapt: Few Shot Domain Adaptation for Semantic Segmentation across Disjoint Labels [article]

Tarun Kalluri, Manmohan Chandraker
2022 arXiv   pre-print
Domain adaptation for semantic segmentation across datasets consisting of the same categories has seen several recent successes.  ...  In this work, we present Cluster-to-Adapt (C2A), a computationally efficient clustering-based approach for domain adaptation across segmentation datasets with completely different, but possibly related  ...  Similar to any semantic segmentation task, the label distribution is not uniform across all the classes in the images.  ... 
arXiv:2208.02804v1 fatcat:uvz7a5ascjhdjktwyhzp7bx5a4

No More Discrimination: Cross City Adaptation of Road Scene Segmenters [article]

Yi-Hsin Chen, Wei-Yu Chen, Yu-Ting Chen, Bo-Cheng Tsai, Yu-Chiang Frank Wang, Min Sun
2017 arXiv   pre-print
By advancing a joint global and class-specific domain adversarial learning framework, adaptation of pre-trained segmenters to that city can be achieved without the need of any user annotation or interaction  ...  Instead of collecting a large number of annotated images of each city of interest to train or refine the segmenter, we propose an unsupervised learning approach to adapt road scene segmenters across different  ...  Thus, how to extend the idea of domain adversarial learning for adapting segmenters across image domains would be our focus.  ... 
arXiv:1704.08509v1 fatcat:7ldqict6yzba5cm6j2aqsdqcu4

Semantic Domain Adversarial Networks for Unsupervised Domain Adaptation [article]

Dapeng Hu, Jian Liang, Qibin Hou, Hanshu Yan, Yunpeng Chen, Shuicheng Yan, Jiashi Feng
2021 arXiv   pre-print
Experiments on both object recognition and semantic segmentation show that SDAN effectively aligns the multi-modal structures across domains and even outperforms state-of-the-art domain adversarial training  ...  To successfully align the multi-modal data structures across domains, the following works exploit discriminative information in the adversarial training process, e.g., using multiple class-wise discriminators  ...  Different from class-agnostic adversarial learning that pursues the marginal distribution alignment but ignores the semantic consistency across domains, (a) previous conditional adversarial learning methods  ... 
arXiv:2003.13274v3 fatcat:yuwcf73ksbc3tfkv4dmnmaqk44

Affinity Space Adaptation for Semantic Segmentation Across Domains

Wei Zhou, Yukang Wang, Jiajia Chu, Jiehua Yang, Xiang Bai, Yongchao Xu
2020 IEEE Transactions on Image Processing  
Semantic segmentation with dense pixel-wise annotation has achieved excellent performance thanks to deep learning.  ...  Extensive experiments demonstrate that the proposed method achieves superior performance against some state-of-theart methods on several challenging benchmarks for semantic segmentation across domains.  ...  Similar to adaptation on classification, most modern adaptation models for semantic segmentation also rely on adversarial learning to adapt domains either in intermediate feature level [11] , [39] ,  ... 
doi:10.1109/tip.2020.3018221 pmid:32870790 fatcat:vb55cxi645fubnnql2i24cqsji

Context-Aware Domain Adaptation in Semantic Segmentation [article]

Jinyu Yang, Weizhi An, Chaochao Yan, Peilin Zhao, Junzhou Huang
2020 arXiv   pre-print
., what and how to transfer domain knowledge across two domains. Existing methods mainly focus on adapting domain-invariant features (what to transfer) through adversarial learning (how to transfer).  ...  In this paper, we consider the problem of unsupervised domain adaptation in the semantic segmentation.  ...  [13] propose the first domain adaptation model for semantic segmentation by learning domain-invariant features through adversarial training.  ... 
arXiv:2003.04010v1 fatcat:h3eyymo5abd6vahm3xsf5io2ui

SPCL: A New Framework for Domain Adaptive Semantic Segmentation via Semantic Prototype-based Contrastive Learning [article]

Binhui Xie, Mingjia Li, Shuang Li
2022 arXiv   pre-print
Specifically, the semantic prototypes provide supervisory signals for per-pixel discriminative representation learning and each pixel of source and target domains in the feature space is required to reflect  ...  To solve this issue, we propose a novel semantic prototype-based contrastive learning framework for fine-grained class alignment.  ...  Specifically, to design an effective contrastive loss for domain adaptive semantic segmentation, the key is to construct the appropriate query-key pairs.  ... 
arXiv:2111.12358v2 fatcat:tiataqjujfdifhpe6y274giv3m

TUNA-Net: Task-oriented UNsupervised Adversarial Network for Disease Recognition in Cross-Domain Chest X-rays [article]

Yuxing Tang, Youbao Tang, Veit Sandfort, Jing Xiao, Ronald M. Summers
2019 arXiv   pre-print
In this work, we exploit the unsupervised domain adaptation problem for radiology image interpretation across domains.  ...  To address the shortcoming of cross-domain, unpaired image-to-image translation methods which typically ignore class-specific semantics, we propose a task-driven, discriminatively trained, cycle-consistent  ...  The authors thank NVIDIA for GPU donations.  ... 
arXiv:1908.07926v1 fatcat:h2frvrescffr7m772dxjnhckom

Universal Semi-Supervised Semantic Segmentation [article]

Tarun Kalluri, Girish Varma, Manmohan Chandraker, C V Jawahar
2019 arXiv   pre-print
In recent years, the need for semantic segmentation has arisen across several different applications and environments.  ...  In contrast to counterpoints such as fine tuning, joint training or unsupervised domain adaptation, universal semi-supervised segmentation ensures that across all domains: (i) a single model is deployed  ...  class prototypes [59] for performing semantic transfer across domains.  ... 
arXiv:1811.10323v3 fatcat:gzblvf2f4ndndamvpoxda57asu
« Previous Showing results 1 — 15 out of 33,999 results