7,217 Hits in 4.6 sec

Progressive Stage-wise Learning for Unsupervised Feature Representation Enhancement [article]

Zefan Li, Chenxi Liu, Alan Yuille, Bingbing Ni, Wenjun Zhang, Wen Gao
2021 arXiv   pre-print
We discover that by progressive stage-wise learning, unsupervised feature representation can be effectively enhanced.  ...  In this work, we explore new dimensions of unsupervised learning by proposing the Progressive Stage-wise Learning (PSL) framework.  ...  In this paper, we propose PSL, a progressive stage-wise learning framework for unsupervised visual representation learning.  ... 
arXiv:2106.05554v2 fatcat:pdcu7fedizdwnchnf4u3vgwlim

Feature Representation Learning for Robust Retinal Disease Detection from Optical Coherence Tomography Images [article]

Sharif Amit Kamran, Khondker Fariha Hossain, Alireza Tavakkoli, Stewart Lee Zuckerbrod, Salah A. Baker
2022 arXiv   pre-print
of disease-specific spatial information, and iii) A novel representation learning module for learning the similarity between encoder-decoder feature and enhancing the accuracy of the model.  ...  To improve the robustness and transferability of knowledge, an enhanced feature-learning module is required to extract meaningful spatial representations from the retinal subspace.  ...  , unsupervised, and feature representation learning to make the robust classifiers for out-of-distribution retinal degeneration detection.  ... 
arXiv:2206.12136v2 fatcat:akfu4djhv5eyni6zhsx2pa6ksm

Large-Scale Unsupervised Deep Representation Learning for Brain Structure [article]

Ayush Jaiswal, Dong Guo, Cauligi S. Raghavendra, Paul Thompson
2018 arXiv   pre-print
In this paper, we present a novel large-scale deep unsupervised approach to learn generic feature representations of structural brain MRI scans, which requires no specialized domain knowledge or manual  ...  Recent advances in Deep Representation Learning have shown great promise in extracting highly non-linear and information-rich features from data.  ...  Computation for the work described in this paper was supported by USC's Center for High-Performance Computing.  ... 
arXiv:1805.01049v1 fatcat:bs2h5ifzu5e2jejr6b3t3eusgy

Unsupervised Deep Learning via Affinity Diffusion

Jiabo Huang, Qi Dong, Shaogang Gong, Xiatian Zhu
In this work, we introduce a general-purpose unsupervised deep learning approach to deriving discriminative feature representations.  ...  However, they usually rely on supervised model learning with the need for massive labelled training data, limiting dramatically their usability and deployability in real-world scenarios without any labelling  ...  Algorithm 1 Unsupervised deep learning via progressive affinity diffusion.  ... 
doi:10.1609/aaai.v34i07.6757 fatcat:j65irmndbvc4xo6s5nq4opr2mq

Instance Similarity Learning for Unsupervised Feature Representation [article]

Ziwei Wang, Yunsong Wang, Ziyi Wu, Jiwen Lu, Jie Zhou
2021 arXiv   pre-print
In this paper, we propose an instance similarity learning (ISL) method for unsupervised feature representation.  ...  On the contrary, our method mines the feature manifold in an unsupervised manner, through which the semantic similarity among instances is learned in order to obtain discriminative representations.  ...  In this paper, we present an ISL method to learn the semantic similarity among instances for unsupervised feature representation.  ... 
arXiv:2108.02721v1 fatcat:aviknxwkpvapxg4kdbp273he3q

Progressive Joint Modeling in Unsupervised Single-Channel Overlapped Speech Recognition

Zhehuai Chen, Jasha Droppo, Jinyu Li, Wayne Xiong
2018 IEEE/ACM Transactions on Audio Speech and Language Processing  
The pretraining regimen uses these modules to solve progressively harder tasks. Transfer learning leverages parallel clean speech to improve the training targets for the network.  ...  learning and a discriminative training criterion.  ...  ACKNOWLEDGMENT We thank Chris Basoglu, Frank Seide for their invaluable assistance with CNTK; Mike Seltzer, Takuya Yoshioka, Hakan Erdogan and Andreas Stolcke for many helpful conversations.  ... 
doi:10.1109/taslp.2017.2765834 fatcat:2e7p7bwqsvhk5beu6lei762ivi

Unsupervised Embedding Learning from Uncertainty Momentum Modeling [article]

Jiahuan Zhou, Yansong Tang, Bing Su, Ying Wu
2021 arXiv   pre-print
Existing popular unsupervised embedding learning methods focus on enhancing the instance-level local discrimination of the given unlabeled images by exploring various negative data.  ...  Moreover, the shortage of positive data and disregard for global discrimination consideration also pose critical issues for unsupervised learning but are always ignored by existing methods.  ...  As the learning progresses, instances in I become more and more discriminative and the uncertainty of instance x i becomes smaller, hence fewer feature candidates are enough for distribution representation  ... 
arXiv:2107.08892v1 fatcat:tkib2zoq7rci7ox7yjxiud6vpi

PyRetri: A PyTorch-based Library for Unsupervised Image Retrieval by Deep Convolutional Neural Networks [article]

Benyi Hu, Ren-Jie Song, Xiu-Shen Wei, Yazhou Yao, Xian-Sheng Hua, Yuehu Liu
2020 arXiv   pre-print
To the best of our knowledge, this is the first open-source library for unsupervised image retrieval by deep learning.  ...  In order to fill this gap, we introduce PyRetri, an open source library for deep learning based unsupervised image retrieval.  ...  To the best of our knowledge, this is the first opensource library for unsupervised image retrieval by deep learning.  ... 
arXiv:2005.02154v2 fatcat:boghtyahzzfqhftixrzbrvhdri

Unsupervised Motion Representation Enhanced Network for Action Recognition [article]

Xiaohang Yang, Lingtong Kong, Jie Yang
2021 arXiv   pre-print
Compared with state-of-the-art unsupervised motion representation learning methods, our model achieves better accuracy while maintaining efficiency, which is competitive with some supervised or more complicated  ...  To fill the gap, we propose UF-TSN, a novel end-to-end action recognition approach enhanced with an embedded lightweight unsupervised optical flow estimator.  ...  CONCLUSION In this paper, we propose UF-TSN, a novel unsupervised motion representation learning framework for action recognition.  ... 
arXiv:2103.03465v1 fatcat:udht4fdnpfhrlpuwd6qkqlo76u

Enhancing Self-supervised Video Representation Learning via Multi-level Feature Optimization [article]

Rui Qian, Yuxi Li, Huabin Liu, John See, Shuangrui Ding, Xian Liu, Dian Li, Weiyao Lin
2021 arXiv   pre-print
We also devise a simple temporal modeling module from multi-level features to enhance motion pattern learning.  ...  The crux of self-supervised video representation learning is to build general features from unlabeled videos.  ...  Conclusion In this work, we propose a multi-level feature optimization framework for unsupervised video representation learning.  ... 
arXiv:2108.02183v2 fatcat:ipbgwj6w6rfgxmhdxfgb5tzgdm

Cross-level Contrastive Learning and Consistency Constraint for Semi-supervised Medical Image Segmentation [article]

Xinkai Zhao, Chaowei Fang, De-Jun Fan, Xutao Lin, Feng Gao, Guanbin Li
2022 arXiv   pre-print
Motivated by this, we propose a cross-level contrastive learning scheme to enhance representation capacity for local features in semi-supervised medical image segmentation.  ...  between global and local patch-wise representations.  ...  Main contributions of this paper are summarized as follows. 1) We devise a cross-level contrastive learning algorithm to enhance the representation capacity for local features in semi-supervised semantic  ... 
arXiv:2202.04074v2 fatcat:whutiwwfajafdnbsdn2xdus6ee

Weakly-Supervised Semantic Segmentation via Sub-category Exploration [article]

Yu-Ting Chang, Qiaosong Wang, Wei-Chih Hung, Robinson Piramuthu, Yi-Hsuan Tsai, Ming-Hsuan Yang
2020 arXiv   pre-print
However, such response maps generated by the classification network usually focus on discriminative object parts, due to the fact that the network does not need the entire object for optimizing the objective  ...  By iteratively clustering image features, the training process does not limit itself to the most discriminative object parts, hence improving the quality of the response maps.  ...  This visually validates that our learned feature representations are enhanced via the sub-category objective in an unsupervised manner.  ... 
arXiv:2008.01183v1 fatcat:gvz6csafdzerxg5hmpqvdowy2q

Adapting Off-the-Shelf Source Segmenter for Target Medical Image Segmentation [article]

Xiaofeng Liu, Fangxu Xing, Chao Yang, Georges El Fakhri, Jonghye Woo
2021 arXiv   pre-print
Unsupervised domain adaptation (UDA) aims to transfer knowledge learned from a labeled source domain to an unlabeled and unseen target domain, which is usually trained on data from both domains.  ...  batch-wise normalization statistics adaptation framework.  ...  The performance was further boosted with the unsupervised learning objective via self-entropy minimization.  ... 
arXiv:2106.12497v1 fatcat:veavv5vu6fag7jq3bkqdurj4qm

MAGAN: Unsupervised low-light image enhancement guided by mixed-attention

Renjun Wang, Bin Jiang, Chao Yang, Qiao Li, Bolin Zhang
2022 Big Data Mining and Analytics  
MAGAN for low-light image enhancement in a fully unsupervised fashion.  ...  Most learning-based low-light image enhancement methods typically suffer from two problems.  ...  To address this issue, unsupervised deep learning-based methods, such as Refs. [13, 27, 28], have been proposed, yet none of them can enhance the image and denoise concurrently in a single-stage network  ... 
doi:10.26599/bdma.2021.9020020 fatcat:5rjrvv4mqzdz5en5pja5zd7nsi

Person Re-identification: A Retrospective on Domain Specific Open Challenges and Future Trends [article]

Asmat Zahra, Nazia Perwaiz, Muhammad Shahzad, Muhammad Moazam Fraz
2022 arXiv   pre-print
Owing to its potential in various applications and research significance, a plethora of deep learning based re-Id approaches have been proposed in the recent years.  ...  For the first time a survey of this type have been presented where the person re-Id approaches are reviewed in such solution-oriented perspective.  ...  A novel descriptor was presented in [192] for effective feature representation and metric learning.  ... 
arXiv:2202.13121v1 fatcat:luwwbcwspndqpauj4dosmmojee
« Previous Showing results 1 — 15 out of 7,217 results