Filters








205 Hits in 1.1 sec

Rethinking Deep Image Prior for Denoising [article]

Yeonsik Jo, Se Young Chun, Jonghyun Choi
2021 arXiv   pre-print
Deep image prior (DIP) serves as a good inductive bias for diverse inverse problems. Among them, denoising is known to be particularly challenging for the DIP due to noise fitting with the requirement of an early stopping. To address the issue, we first analyze the DIP by the notion of effective degrees of freedom (DF) to monitor the optimization progress and propose a principled stopping criterion before fitting to noise without access of a paired ground truth image for Gaussian noise. We also
more » ... propose the 'stochastic temporal ensemble (STE)' method for incorporating techniques to further improve DIP's performance for denoising. We additionally extend our method to Poisson noise. Our empirical validations show that given a single noisy image, our method denoises the image while preserving rich textual details. Further, our approach outperforms prior arts in LPIPS by large margins with comparable PSNR and SSIM on seven different datasets.
arXiv:2108.12841v1 fatcat:bexeypgirbfwxa3oubd2voga4y

Learning Architectures for Binary Networks [article]

Dahyun Kim, Kunal Pratap Singh, Jonghyun Choi
2020 arXiv   pre-print
Backbone architectures of most binary networks are well-known floating point architectures such as the ResNet family. Questioning that the architectures designed for floating point networks would not be the best for binary networks, we propose to search architectures for binary networks (BNAS) by defining a new search space for binary architectures and a novel search objective. Specifically, based on the cell based search method, we define the new search space of binary layer types, design a
more » ... cell template, and rediscover the utility of and propose to use the Zeroise layer instead of using it as a placeholder. The novel search objective diversifies early search to learn better performing binary architectures. We show that our proposed method searches architectures with stable training curves despite the quantization error inherent in binary networks. Quantitative analyses demonstrate that our searched architectures outperform the architectures used in state-of-the-art binary networks and outperform or perform on par with state-of-the-art binary networks that employ various techniques other than architectural changes.
arXiv:2002.06963v2 fatcat:edcfsdrygbhuvdvoggmphzeffi

Learning Temporal Regularity in Video Sequences [article]

Mahmudul Hasan, Jonghyun Choi, Jan Neumann, Amit K. Roy-Chowdhury, Larry S. Davis
2016 arXiv   pre-print
Perceiving meaningful activities in a long video sequence is a challenging problem due to ambiguous definition of 'meaningfulness' as well as clutters in the scene. We approach this problem by learning a generative model for regular motion patterns, termed as regularity, using multiple sources with very limited supervision. Specifically, we propose two methods that are built upon the autoencoders for their ability to work with little to no supervision. We first leverage the conventional
more » ... ted spatio-temporal local features and learn a fully connected autoencoder on them. Second, we build a fully convolutional feed-forward autoencoder to learn both the local features and the classifiers as an end-to-end learning framework. Our model can capture the regularities from multiple datasets. We evaluate our methods in both qualitative and quantitative ways - showing the learned regularity of videos in various aspects and demonstrating competitive performance on anomaly detection datasets as an application.
arXiv:1604.04574v1 fatcat:2ejzqxcrpzcmbpgrsgv6tre2xu

Confidence Calibration for Incremental Learning

Dongmin Kang, Yeonsik Jo, Yeongwoo Nam, Jonghyun Choi
2020 IEEE Access  
doi:10.1109/access.2020.3007234 fatcat:lcykyhozofffdkdx7fi2ahdxfu

BNAS v2: Learning Architectures for Binary Networks with Empirical Improvements [article]

Dahyun Kim, Kunal Pratap Singh, Jonghyun Choi
2021 arXiv   pre-print
Backbone architectures of most binary networks are well-known floating point (FP) architectures such as the ResNet family. Questioning that the architectures designed for FP networks might not be the best for binary networks, we propose to search architectures for binary networks (BNAS) by defining a new search space for binary architectures and a novel search objective. Specifically, based on the cell based search method, we define the new search space of binary layer types, design a new cell
more » ... emplate, and rediscover the utility of and propose to use the Zeroise layer instead of using it as a placeholder. The novel search objective diversifies early search to learn better performing binary architectures. We show that our method searches architectures with stable training curves despite the quantization error inherent in binary networks. Quantitative analyses demonstrate that our searched architectures outperform the architectures used in state-of-the-art binary networks and outperform or perform on par with state-of-the-art binary networks that employ various techniques other than architectural changes. In addition, we further propose improvements to the training scheme of our searched architectures. With the new training scheme for our searched architectures, we achieve the state-of-the-art performance by binary networks by outperforming all previous methods by non-trivial margins.
arXiv:2110.08562v1 fatcat:6ru32qv7wfd2pkdm3aftn5wy5y

Zero-shot Natural Language Video Localization [article]

Jinwoo Nam and Daechul Ahn and Dongyeop Kang and Seong Jong Ha and Jonghyun Choi
2021 arXiv   pre-print
Understanding videos to localize moments with natural language often requires large expensive annotated video regions paired with language queries. To eliminate the annotation costs, we make a first attempt to train a natural language video localization model in zero-shot manner. Inspired by unsupervised image captioning setup, we merely require random text corpora, unlabeled video collections, and an off-the-shelf object detector to train a model. With the unpaired data, we propose to generate
more » ... pseudo-supervision of candidate temporal regions and corresponding query sentences, and develop a simple NLVL model to train with the pseudo-supervision. Our empirical validations show that the proposed pseudo-supervised method outperforms several baseline approaches and a number of methods using stronger supervision on Charades-STA and ActivityNet-Captions.
arXiv:2110.00428v1 fatcat:ca7uj2vn7za6bixhqhgh2x3ax4

Incremental Learning with Maximum Entropy Regularization: Rethinking Forgetting and Intransigence [article]

Dahyun Kim, Jihwan Bae, Yeonsik Jo, Jonghyun Choi
2019 arXiv   pre-print
Correspondence to: Jonghyun Choi <jhc@gist.ac.kr>. arXiv Preprint. 2018).  ... 
arXiv:1902.00829v1 fatcat:zv4xhihgcvdwlon2kcx4vvfilq

Mining Discriminative Triplets of Patches for Fine-Grained Classification [article]

Yaming Wang, Jonghyun Choi, Vlad I. Morariu, Larry S. Davis
2016 arXiv   pre-print
Fine-grained classification involves distinguishing between similar sub-categories based on subtle differences in highly localized regions; therefore, accurate localization of discriminative regions remains a major challenge. We describe a patch-based framework to address this problem. We introduce triplets of patches with geometric constraints to improve the accuracy of patch localization, and automatically mine discriminative geometrically-constrained triplets for classification. The
more » ... approach only requires object bounding boxes. Its effectiveness is demonstrated using four publicly available fine-grained datasets, on which it outperforms or achieves comparable performance to the state-of-the-art in classification.
arXiv:1605.01130v1 fatcat:gsf5nxmilrfw5go3plonqn7cwa

Implementation of Single Source Based Hospital Information System for the Catholic Medical Center Affiliated Hospitals

Inyoung Choi, Ran Choi, Jonghyun Lee, Byung Gil Choi
2010 Healthcare Informatics Research  
Figure 3 . 3 Information technology coordinators (ITC) organization. doi: 10.4258/hir.2010.16.2.133 www.e-hir.org Inyoung Choi et al doi: 10.4258/hir.2010.16.2.133 www.e-hir.org Inyoung Choi et al manitarian  ...  Mary's hospital opened with cutting-edge facilities including 22 floors, 6 basements, 1,200 patient beds and around Inyoung Choi et al information system should be reduced.  ... 
doi:10.4258/hir.2010.16.2.133 pmid:21818432 pmcid:PMC3089864 fatcat:o2kyjm537rci5d4p7wo7hdhule

Attention-Based Automated Feature Extraction for Malware Analysis

Sunoh Choi, Jangseong Bae, Changki Lee, Youngsoo Kim, Jonghyun Kim
2020 Sensors  
Every day, hundreds of thousands of malicious files are created to exploit zero-day vulnerabilities. Existing pattern-based antivirus solutions face difficulties in coping with such a large number of new malicious files. To solve this problem, artificial intelligence (AI)-based malicious file detection methods have been proposed. However, even if we can detect malicious files with high accuracy using deep learning, it is difficult to identify why files are malicious. In this study, we propose a
more » ... malicious file feature extraction method based on attention mechanism. First, by adapting the attention mechanism, we can identify application program interface (API) system calls that are more important than others for determining whether a file is malicious. Second, we confirm that this approach yields an accuracy that is approximately 12% and 5% higher than a conventional AI-based detection model using convolutional neural networks and skip-connected long short-term memory-based detection model, respectively.
doi:10.3390/s20102893 pmid:32443750 fatcat:ggdjewjdwff2jj4anh5db5pn4e

Predictable Dual-View Hashing

Mohammad Rastegari, Jonghyun Choi, Shobeir Fakhraei, Hal Daumé III, Larry S. Davis
2013 International Conference on Machine Learning  
We propose a Predictable Dual-View Hashing (PDH) algorithm which embeds proximity of data samples in the original spaces. We create a cross-view hamming space with the ability to compare information from previously incomparable domains with a notion of 'predictability'. By performing comparative experimental analysis on two large datasets, PASCAL-Sentence and SUN-Attribute, we demonstrate the superiority of our method to the state-of-the-art dual-view binary code learning algorithms.
dblp:conf/icml/RastegariCFHD13 fatcat:vwdmcihf2nc43fkkv4eqhxllee

Toward Sparse Coding on Cosine Distance

Jonghyun Choi, Hyunjong Cho, Jungsuk Kwac, Larry S. Davis
2014 2014 22nd International Conference on Pattern Recognition  
Sparse coding is a regularized least squares solution using the L1 or L0 constraint, based on the Euclidean distance between original and reconstructed signals with respect to a predefined dictionary. The Euclidean distance, however, is not a good metric for many feature descriptors, especially histogram features, e.g. many visual features including SIFT, HOG, LBP and Bag-of-visual-words. In contrast, cosine distance is a more appropriate metric for such features. To leverage the benefit of the
more » ... cosine distance in sparse coding, we formulate a new sparse coding objective function based on approximate cosine distance by constraining a norm of the reconstructed signal to be close to the norm of the original signal. We evaluate our new formulation on three computer vision datasets (UCF101 Action dataset, AR dataset and Extended YaleB dataset) and show improvements over the Euclidean distance based objective.
doi:10.1109/icpr.2014.757 dblp:conf/icpr/ChoiCKD14 fatcat:qei4su2vuvaeln3kjupij77hs4

Unsupervised Domain Adaptation for 3D Point Clouds by Searched Transformations

Dongmin Kang, Yeongwoo Nam, Daeun Kyung, Jonghyun Choi
2022 IEEE Access  
Input-level domain adaptation reduces the burden of a neural encoder without supervision by reducing the domain gap at the input level. Input-level domain adaptation is widely employed in 2D visual domain, e.g., images and videos, but is not utilized for 3D point clouds. We propose the use of input-level domain adaptation for 3D point clouds, namely, point-level domain adaptation. Specifically, we propose to learn a transformation of 3D point clouds by searching the best combination of
more » ... s on point clouds that transfer data from the source domain to the target domain while maintaining the classification label without supervision of the target label. We decompose the learning objective into two terms, resembling domain shift and preserving label information. On the PointDA-10 benchmark dataset, our method outperforms state-ofthe-art, unsupervised, point cloud domain adaptation methods by large margins (up to + 3.97 % in average). INDEX TERMS Data transformation, domain adaptation, point cloud recognition.
doi:10.1109/access.2022.3176719 fatcat:sejzxm3lzvai7osbs4hqw6ofji

Face Identification Using Large Feature Sets

W. R. Schwartz, Huimin Guo, Jonghyun Choi, L. S. Davis
2012 IEEE Transactions on Image Processing  
Choi is with the Department of Electrical and Computer Engineering, University of Maryland, College Park, MD 20742 USA (e-mail: jhchoi@umd.edu).  ... 
doi:10.1109/tip.2011.2176951 pmid:22128005 fatcat:qcrteqsjebgg7inf2h23pxov3u

Evaluations of AI‐based malicious PowerShell detection with feature optimizations

Jihyeon Song, Jungtae Kim, Sunoh Choi, Jonghyun Kim, Ikkyun Kim
2021 ETRI Journal  
ICT infrastructure protection against intelligent malware threats). Cyberattacks are often difficult to identify with traditional signature-based detection, because attackers continually find ways to bypass the detection methods. Therefore, researchers have introduced artificial intelligence (AI) technology for cybersecurity analysis to detect malicious PowerShell scripts. In this paper, we propose a feature optimization technique for AI-based approaches to enhance the accuracy of malicious
more » ... rShell script detection. We statically analyze the PowerShell script and preprocess it with a method based on the tokens and abstract syntax tree (AST) for feature selection. Here, tokens and AST represent the vocabulary and structure of the PowerShell script, respectively. Performance evaluations with optimized features yield detection rates of 98% in both machine learning (ML) and deep learning (DL) experiments. Among them, the ML model with the 3-gram of selected five tokens and the DL model with experiments based on the AST 3-gram deliver the best performance. K E Y W O R D S Deep learning, feature optimization, fileless malware, machine learning, PowerShell script 550 | SONG et al.
doi:10.4218/etrij.2020-0215 fatcat:zq726jln4rcfjb3o6xh2aauo4a
« Previous Showing results 1 — 15 out of 205 results