Filters








1,145 Hits in 4.8 sec

Category Anchor-Guided Unsupervised Domain Adaptation for Semantic Segmentation [article]

Qiming Zhang, Jing Zhang, Wei Liu, Dacheng Tao
2019 arXiv   pre-print
In this paper, we propose a novel category anchor-guided (CAG) UDA model for semantic segmentation, which explicitly enforces category-aware feature alignment to learn shared discriminative features and  ...  Unsupervised domain adaptation (UDA) aims to enhance the generalization capability of a certain model from a source domain to a target domain.  ...  Specifically, we propose a novel category anchor-guided unsupervised domain adaptation model (CAG-UDA) for semantic segmentation.  ... 
arXiv:1910.13049v2 fatcat:3qior2olwvgoxoemdxtsnpuxxq

Spatial Attention Pyramid Network for Unsupervised Domain Adaptation [article]

Congcong Li, Dawei Du, Libo Zhang, Longyin Wen, Tiejian Luo, Yanjun Wu, Pengfei Zhu
2020 arXiv   pre-print
We conduct extensive experiments on various challenging datasets for unsupervised domain adaptation on object detection, instance segmentation, and semantic segmentation, which demonstrates that our method  ...  Unsupervised domain adaptation is critical in various computer vision tasks, such as object detection, instance segmentation, and semantic segmentation, which aims to alleviate performance degradation  ...  For semantic segmentation, the guided information is the output map with the size of C sem × H × W from the segmentation head, where C sem is the number of semantic categories.  ... 
arXiv:2003.12979v3 fatcat:fxyeg5aksjer7eraw3qhic4kmy

Task-Assisted Domain Adaptation with Anchor Tasks [article]

Zhizhong Li, Linjie Luo, Sergey Tulyakov, Qieyun Dai, Derek Hoiem
2020 arXiv   pre-print
We evaluate our methods on surface normal estimation on two pairs of datasets (indoor scenes and faces) with two kinds of anchor tasks (semantic segmentation and facial landmarks).  ...  We show that blindly applying domain adaptation or training the auxiliary task on only one domain may hurt performance, while using anchor tasks on both domains is better behaved.  ...  How can we better adapt from synthetic to real data? In this paper, we propose to use anchor tasks as a guide for improving pixel-level domain transfer of the main task.  ... 
arXiv:1908.06079v3 fatcat:maoxeyvhqbbfxab7txsqeff6oy

Margin Preserving Self-paced Contrastive Learning Towards Domain Adaptation for Medical Image Segmentation [article]

Zhizhe Liu, Zhenfeng Zhu, Shuai Zheng, Yang Liu, Jiayu Zhou, Yao Zhao
2021 arXiv   pre-print
To bridge the gap between the source and target domains in unsupervised domain adaptation (UDA), the most common strategy puts focus on matching the marginal distributions in the feature space through  ...  To enhance the supervision for contrastive learning, more informative pseudo-labels are generated in target domain in a self-paced way, thus benefiting the category-aware distribution alignment for UDA  ...  As an anchor-guided UDA model for semantic segmentation, both category-wise domain alignment and self-training were facilitated in an explicit way [14] .  ... 
arXiv:2103.08454v1 fatcat:i7dahzzs6zcabeco4v5zjtqyay

Unsupervised Domain Adaptation in Semantic Segmentation: A Review

Marco Toldo, Andrea Maracani, Umberto Michieli, Pietro Zanuttigh
2020 Technologies  
The aim of this paper is to give an overview of the recent advancements in the Unsupervised Domain Adaptation (UDA) of deep networks for semantic segmentation.  ...  This task is attracting a wide interest since semantic segmentation models require a huge amount of labeled data and the lack of data fitting specific requirements is the main limitation in the deployment  ...  Review of Unsupervised Domain Adaptation Strategies This section reviews the most relevant approaches for Unsupervised Domain Adaptation in semantic segmentation.  ... 
doi:10.3390/technologies8020035 fatcat:qzgjjiw5p5bldk76mh3s3pwlfq

Unsupervised Domain Adaptation in Semantic Segmentation: a Review [article]

Marco Toldo, Andrea Maracani, Umberto Michieli, Pietro Zanuttigh
2020 arXiv   pre-print
The aim of this paper is to give an overview of the recent advancements in the Unsupervised Domain Adaptation (UDA) of deep networks for semantic segmentation.  ...  This task is attracting a wide interest, since semantic segmentation models require a huge amount of labeled data and the lack of data fitting specific requirements is the main limitation in the deployment  ...  Review of Unsupervised Domain Adaptation strategies This section reviews the most relevant approaches for Unsupervised Domain Adaptation in semantic segmentation.  ... 
arXiv:2005.10876v1 fatcat:7t5v6qibxnfcxhwtohqqunhd2u

Style Mixing and Patchwise Prototypical Matching for One-Shot Unsupervised Domain Adaptive Semantic Segmentation [article]

Xinyi Wu and Zhenyao Wu and Yuhang Lu and Lili Ju and Song Wang
2021 arXiv   pre-print
In this paper, we tackle the problem of one-shot unsupervised domain adaptation (OSUDA) for semantic segmentation where the segmentors only see one unlabeled target image during training.  ...  Experimental results show that our method achieves new state-of-the-art performance on two commonly used benchmarks for domain adaptive semantic segmentation under the one-shot setting and is more efficient  ...  Category Anchor-Guided Unsupervised Domain Adaptation for Se- mantic Segmentation. In Advances in Neural Information Processing Systems, 433–443. Zhang, Y.; David, P.; and Gong, B. 2017.  ... 
arXiv:2112.04665v1 fatcat:tht2zbamqjf2tciatpo2if4ele

Coarse-to-Fine Domain Adaptive Semantic Segmentation with Photometric Alignment and Category-Center Regularization [article]

Haoyu Ma, Xiangru Lin, Zifeng Wu, Yizhou Yu
2021 arXiv   pre-print
Unsupervised domain adaptation (UDA) in semantic segmentation is a fundamental yet promising task relieving the need for laborious annotation works.  ...  However, the domain shifts/discrepancies problem in this task compromise the final segmentation performance.  ...  Such an iterative approach is frequently called self-supervised training in the area of unsupervised domain adaptation for semantic segmentation [14, 34, 33, 30] .  ... 
arXiv:2103.13041v1 fatcat:pzy57sziizerjffeuzugllsl2e

Unsupervised Domain Adaptation for Semantic Image Segmentation: a Comprehensive Survey [article]

Gabriela Csurka, Riccardo Volpi, Boris Chidlovskii
2021 arXiv   pre-print
We present the most important semantic segmentation methods; we provide a comprehensive survey on domain adaptation techniques for semantic segmentation; we unveil newer trends such as multi-domain learning  ...  Since unlabelled data is instead significantly cheaper to obtain, it is not surprising that Unsupervised Domain Adaptation reached a broad success within the semantic segmentation community.  ...  In Category Anchor-Guided Unsupervised Domain Adapta- Joint European Conference on Machine Learning and tion for Semantic Segmentation.  ... 
arXiv:2112.03241v1 fatcat:uzlehddvuvfwzf4dfbjimja45e

Unsupervised Adaptation of Semantic Segmentation Models without Source Data [article]

Sujoy Paul, Ansh Khurana, Gaurav Aggarwal
2021 arXiv   pre-print
We consider the novel problem of unsupervised domain adaptation of source models, without access to the source data for semantic segmentation.  ...  Unsupervised domain adaptation aims to adapt a model learned on the labeled source data, to a new unlabeled target dataset.  ...  Cate- invariant examples for domain adaptation in semantic seg- gory anchor-guided unsupervised domain adaptation for se-  ... 
arXiv:2112.02359v1 fatcat:ryryj4kuorb3bob6e6mokulyeu

Phase Consistent Ecological Domain Adaptation [article]

Yanchao Yang, Dong Lao, Ganesh Sundaramoorthi, Stefano Soatto
2020 arXiv   pre-print
Incorporating these two priors in a standard domain adaptation framework improves performance across the board in the most common unsupervised domain adaptation benchmarks for semantic segmentation.  ...  known as unsupervised domain adaptation.  ...  Method We first describe general image translation for unsupervised domain adaptation (UDA) and how it is used in semantic segmentation.  ... 
arXiv:2004.04923v1 fatcat:kt5qeiilivavpbfavsxjem56by

Integrating Categorical Semantics into Unsupervised Domain Translation [article]

Samuel Lavoie, Faruk Ahmed, Aaron Courville
2021 arXiv   pre-print
In particular, we demonstrate that categorical semantics improves the translation between perceptually different domains sharing multiple object categories.  ...  We propose a method to learn, in an unsupervised manner, categorical semantic features (such as object labels) that are invariant of the source and target domains.  ...  We also acknowledge Compute-Canada and Mila for providing the computing ressources used for this work.  ... 
arXiv:2010.01262v2 fatcat:4762qyrt5jhtrlcd2m3etgtghe

Contrast and Mix: Temporal Contrastive Video Domain Adaptation with Background Mixing [article]

Aadarsh Sahoo, Rutav Shah, Rameswar Panda, Kate Saenko, Abir Das
2021 arXiv   pre-print
While many domain adaptation techniques have been proposed for images, the problem of unsupervised domain adaptation in videos remains largely underexplored.  ...  Unsupervised domain adaptation which aims to adapt models trained on a labeled source domain to a completely unlabeled target domain has attracted much attention in recent years.  ...  space for unsupervised video domain adaptation.  ... 
arXiv:2110.15128v1 fatcat:zdlx5taqxraffo33he33w44zpi

ADVENT: Adversarial Entropy Minimization for Domain Adaptation in Semantic Segmentation [article]

Tuan-Hung Vu, Himalaya Jain, Maxime Bucher, Matthieu Cord, Patrick Pérez
2019 arXiv   pre-print
In this work, we address the task of unsupervised domain adaptation in semantic segmentation with losses based on the entropy of the pixel-wise predictions.  ...  Semantic segmentation is a key problem for many computer vision tasks.  ...  Figure 1 : 1 Proposed entropy-based unsupervised domain adaptation for semantic segmentation. The top two rows show results on source and target domain scenes of the model trained without adaptation.  ... 
arXiv:1811.12833v2 fatcat:e7ox63x7svbzvlvnalfbleotp4

Manual-Label Free 3D Detection via An Open-Source Simulator [article]

Zhen Yang and Chi Zhang and Huiming Guo and Zhaoxiang Zhang
2020 arXiv   pre-print
Then a DA-VoxelNet that integrates both a sample-level DA module and an anchor-level DA module is proposed to enable the detector trained by the synthetic data to adapt to real scenario.  ...  In this paper, we propose a manual-label free 3D detection algorithm that leverages the CARLA simulator to generate a large amount of self-labeled training samples and introduces a novel Domain Adaptive  ...  Domain Adaptive VoxelNet We propose a novel unsupervised domain adaptation algorithm to further align the domain shift.  ... 
arXiv:2011.07784v1 fatcat:c4sjmrjyajamrkk73ami7dih3q
« Previous Showing results 1 — 15 out of 1,145 results