检索结果-内蒙古大学图书馆

Anti-Adversarially Manipulated Attributions for Weakly Supervised Semantic Segmentation and Object Localization

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Lee, Jungbeom Kim, Eunji Mok, Jisoo Yoon, Sungroh Department of Electrical and Computer Engineering Seoul National University Seoul08826 Korea Republic of Interdisciplinary Program in AI AIIS ASRI INMC and ISRC Seoul National University Seoul08826 Korea Republic of

Obtaining accurate pixel-level localization from class labels is a crucial process in weakly supervised semantic segmentation and object localization. Attribution maps from a trained classifier are widely used to provide pixel-level localization, but their focus tends to be restricted to a small discriminative region of the target object. An AdvCAM is an attribution map of an image that is manipulated to increase the classification score produced by a classifier before the final softmax or sigmoid layer. This manipulation is realized in an anti-adversarial manner, so that the original image is perturbed along pixel gradients in directions opposite to those used in an adversarial attack. This process enhances non-discriminative yet class-relevant features, which make an insufficient contribution to previous attribution maps, so that the resulting AdvCAM identifies more regions of the target object. In addition, we introduce a new regularization procedure that inhibits the incorrect attribution of regions unrelated to the target object and the excessive concentration of attributions on a small region of the target object. Our method achieves a new state-of-the-art performance in weakly and semi-supervised semantic segmentation, on both the PASCAL VOC 2012 and MS COCO 2014 datasets. In weakly supervised object localization, it achieves a new state-of-the-art performance on the CUB-200-2011 and ImageNet-1K datasets. © 2022, CC BY-NC-ND.

关键词： Semantic Segmentation

Out of Sight, Out of Mind: A Source-View-Wise Feature Aggregation for Multi-View Image-Based Rendering

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Cha, Geonho Shin, Chaehun Yoon, Sungroh Wee, Dongyoon Clova AI NAVER Corp. Korea Republic of Department of Electrical and Computer Engineering Seoul National University Korea Republic of Interdisciplinary Program in Artificial Intelligence Seoul National University Korea Republic of

To estimate the volume density and color of a 3D point in the multi-view image-based rendering, a common approach is to inspect the consensus existence among the given source image features, which is one of the informative cues for the estimation procedure. To this end, most of the previous methods utilize equally-weighted aggregation features. However, this could make it hard to check the consensus existence when some outliers, which frequently occur by occlusions, are included in the source image feature set. In this paper, we propose a novel source-view-wise feature aggregation method, which facilitates us to find out the consensus in a robust way by leveraging local structures in the feature set. We first calculate the source-view-wise distance distribution for each source feature for the proposed aggregation. After that, the distance distribution is converted to several similarity distributions with the proposed learnable similarity mapping functions. Finally, for each element in the feature set, the aggregation features are extracted by calculating the weighted means and variances, where the weights are derived from the similarity distributions. In experiments, we validate the proposed method on various benchmark datasets, including synthetic and real image scenes. The experimental results demonstrate that incorporating the proposed features improves the performance by a large margin, resulting in the state-of-the-art performance. © 2022, CC BY-NC-ND.

关键词： Three dimensional computer graphics

Performance of Building-Integrated Photovoltaic and Addressing the Challenges of Color, Safety, and Light-Emitting Diode Integration

学校读者我要写书评

暂无评论

Physica Status Solidi (A) Applications and Materials Science 2025年第10期222卷

作者： Mohammed, Shurouq Abdulqadir Yousuf, Hasnain Jony, Jaljalalul Abedin Khokhar, Muhammad Quddamah Rahman, Rafi Ur aida, Maha Nur Alamgeer Park, Sangheon Yi, Junsin Department of Electrical and Computer Engineering Sungkyunkwan University Gyeonggi-Do Suwon16419 Korea Republic of Interdisciplinary Program in Photovoltaic System Engineering Sungkyunkwan University Gyeonggi-Do Suwon16419 Korea Republic of Department of Future Energy Engineering Sungkyunkwan University Gyeonggi-Do Suwon16419 Korea Republic of College of Information and Communication Engineering Sungkyunkwan University Gyeonggi-Do Suwon16419 Korea Republic of

Building-integrated photovoltaics (BIPVs) provide a dual-purpose solution, functioning as both energy-generating systems and integral building materials. BIPV adoption has been limited due to a lack of color integration, preventing the panels from being used in building designs. However, efforts to incorporate color into the BIPV layers have been made, but they often reduce panel efficiency. BIPV implementation faces several challenges that hinder widespread adoption, such as fire safety concerns and the integration of color aesthetics and light-emitting diode technology. This article explores these challenges and recent advancements that enhance the aesthetic versatility and efficiency of BIPV systems, including the use of quantum dots, Bragg reflectors, and silver-alloyed copper indium gallium selenide cells. The study also highlights critical considerations for optimizing BIPV systems, such as energy performance and fire safety. PVsyst 7.2 simulation results for a 10 kW BIPV system installed in South Korea demonstrate effective strategies for mitigating efficiency losses through investigating the orientation of angles. Results showcase a high-performance ratio of 0.891, a high production of 14 147 kWh year−1 at a tilt angle of 30° while facing south, and a peak energy of almost 1450 kWh injected into the grid under varying conditions. © 2025 Wiley-VCH GmbH.

关键词： building-integrated photovoltaics colors efficiencies light-emitting diodes technologies

PUCA: Patch-Unshuffle and Channel Attention for Enhanced Self-Supervised Image Denoising

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Jang, Hyemi Park, Junsung Jung, Dahuin Lew, Jaihyun Bae, Ho Yoon, Sungroh Department of Electrical and Computer Engineering Seoul National University Korea Republic of Interdisciplinary Program in Artificial Intelligence Seoul National University Korea Republic of Department of Cyber Security Ewha Womans University Korea Republic of

Although supervised image denoising networks have shown remarkable performance on synthesized noisy images, they often fail in practice due to the difference between real and synthesized noise. Since clean-noisy image pairs from the real world are extremely costly to gather, self-supervised learning, which utilizes noisy input itself as a target, has been studied. To prevent a self-supervised denoising model from learning identical mapping, each output pixel should not be influenced by its corresponding input pixel;This requirement is known as J -invariance. Blind-spot networks (BSNs) have been a prevalent choice to ensure J -invariance in self-supervised image denoising. However, constructing variations of BSNs by injecting additional operations such as downsampling can expose blinded information, thereby violating J -invariance. Consequently, convolutions designed specifically for BSNs have been allowed only, limiting architectural flexibility. To overcome this limitation, we propose PUCA, a novel J -invariant U-Net architecture, for self-supervised denoising. PUCA leverages patch-unshuffle/shuffle to dramatically expand receptive fields while maintaining J -invariance and dilated attention blocks (DABs) for global context incorporation. Experimental results demonstrate that PUCA achieves state-of-the-art performance, outperforming existing methods in self-supervised image denoising. Copyright © 2023, The Authors. All rights reserved.

关键词： Image denoising

How Media Competition Fuels the Spread of Misinformation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Amini, Arash Bayiz, Yigit Ege Lee, Eun-Ju Somer-Topcu, Zeynep Marculescu, Radu Topcu, Ufuk The University of Texas at Austin Oden Institute for Computational Engineering and Sciences United States The University of Texas at Austin Department of Electrical and Computer Engineering United States Seoul National University Department of Communication & Interdisciplinary Program in Artificial Intelligence Korea Republic of The University of Texas at Austin Department of Government United States

Competition among news sources may encourage some sources to share fake news and misinformation to influence the public. While sharing misinformation may lead to a short-term gain in audience engagement, it may damage the reputation of these sources, resulting in a loss of audience. To understand the rationale behind sharing misinformation, we model the competition as a zero-sum sequential game, where each news source influences individuals based on its credibility—how trustworthy the public perceives it—and the individual’s opinion and susceptibility. In this game, news sources can decide whether to share factual information to enhance their credibility or disseminate misinformation for greater immediate attention at the cost of losing credibility. We employ the quantal response equilibrium concept, which accounts for the bounded rationality of human decision-making, allowing for imperfect or probabilistic choices. Our analysis shows that the resulting equilibria for this game reproduce the credibility-bias distribution observed in real-world news sources, with hyper-partisan sources more likely to spread misinformation than centrist ones. It further illustrates that disseminating misinformation can polarize the public. Notably, our model reveals that when one player increases misinformation dissemination, the other player is likely to follow, exacerbating the spread of misinformation. We conclude by discussing potential strategies to mitigate the spread of fake news and promote a more factual and reliable information landscape. Copyright © 2024, The Authors. All rights reserved.

关键词：

Fusion of Mixture of Experts and Generative Artificial Intelligence in Mobile Edge Metaverse

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Liu, Guangyuan Du, Hongyang Niyato, Dusit Kang, Jiawen Xiong, Zehui Jamalipour, Abbas Mao, Shiwen Kim, Dong In The College of Computing and Data Science The Energy Research Institute @ NTU Interdisciplinary Graduate Program Nanyang Technological University Singapore The College of Computing and Data Science Nanyang Technological University Singapore The School of Automation Guangdong University of Technology China The Pillar of Information Systems Technology and Design Singapore University of Technology and Design Singapore The Department of Electrical and Computer Engineering Auburn University Auburn United States The School of Electrical and Computer Engineering University of Sydney Australia The Department of Electrical and Computer Engineering Sungkyunkwan University Korea Republic of

In the digital transformation era, Metaverse offers a fusion of virtual reality (VR), augmented reality (AR), and web technologies to create immersive digital experiences. However, the evolution of the Metaverse is slowed down by the challenges of content creation, scalability, and dynamic user interaction. Our study investigates an integration of Mixture of Experts (MoE) models with Generative Artificial Intelligence (Gai) for mobile edge computing to revolutionize content creation and interaction in the Metaverse. Specifically, we harness an MoE model's ability to efficiently manage complex data and complex tasks by dynamically selecting the most relevant experts running various sub-models to enhance the capabilities of Gai. We then present a novel framework that improves video content generation quality and consistency, and demonstrate its application through case studies. Our findings underscore the efficacy of MoE and Gai integration to redefine virtual experiences by offering a scalable, efficient pathway to harvest the Metaverse's full potential. © 2024, CC BY.

关键词： Augmented reality

Subwavelength-engineered Antislot Photonic Crystals in a Silicon Photonics Foundry for On-chip Communications

学校读者我要写书评

暂无评论

Subwavelength-engineered Antislot Photonic Crystals in a Sil...

CLEO: Science and Innovations in CLEO 2024, CLEO: S and I 2024 - Part of Conference on Lasers and Electro-Optics

作者： Arnold, Kellen P. Allen, Joshua A. Halimi, Sami I. Ryder, Landen D. Afzal, Francis O. Bian, Yusheng Aboketaf, Abdelsalam Hirokawa, Takako Dezfulian, Kevin Rakowski, Michal Augur, Rod Nummy, Karen Weiss, Sharon M. Interdisciplinary Materials Science Program Vanderbilt University 2201 West End Ave NashvilleTN37235 United States Department of Electrical and Computer Engineering Vanderbilt University 2201 West End Ave NashvilleTN37235 United States NASA Goddard Space Flight Center Code 561 GreenbeltMD20771 United States GlobalFoundries 400 Stone Break Rd. Ext. MaltaNY12020 United States

We report scalable foundry fabrication and characterization of photonic crystal nanobeam waveguides incorporating subwavelength-scale dielectric antislot unit cells. This work enables enhanced light-matter interaction and three-fold improvement in VπL when incorporated in Mach-Zehnder modulators. © Optica Publishing Group 2024

关键词： Silicon photonics

Subwavelength-engineered Antislot Photonic Crystals in a Silicon Photonics Foundry for On-chip Communications

学校读者我要写书评

暂无评论

Subwavelength-engineered Antislot Photonic Crystals in a Sil...

2024 Conference on Lasers and Electro-Optics, CLEO 2024

ISBN: (纸本)9781957171395

关键词： Mach-Zehnder interferometers

D-SRGAN: DEM Super-Resolution with Generative Adversarial Networks

学校读者我要写书评

暂无评论

SN computer Science 2021年第1期2卷 48页

作者： Demiray, Bekir Z. Sit, Muhammed Demir, Ibrahim Department of Computer Science University of Iowa Iowa City United States Interdisciplinary Graduate Program in Informatics University of Iowa Iowa City United States Department of Civil and Environmental Engineering University of Iowa Iowa City United States Department of Electrical and Computer Engineering University of Iowa Iowa City United States

Digital elevation model (DEM) is a critical data source for variety of applications such as road extraction, hydrological modeling, flood mapping, and many geospatial studies. The usage of high-resolution DEMs as inputs in many application areas improves the overall reliability and accuracy of the raw dataset. The goal of this study is to develop a machine learning model that increases the spatial resolution of DEM without additional information. In this paper, a GAN based model (D-SRGAN), inspired by single image super-resolution methods, is developed and evaluated to increase the resolution of DEMs. The experiment results show that D-SRGAN produces promising results while constructing 3 feet high-resolution DEMs from 50 feet low-resolution DEMs. It outperforms common statistical interpolation methods and neural network *** study shows that it is possible to use the power of artificial neural networks to increase the resolution of the DEMs. The study also demonstrates that approaches from single image super-resolution can be applied for DEM super-resolution. © 2021, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. part of Springer Nature.

关键词： Deep learning DEM DEM reconstruction Generative adversarial networks (GANs) Super-resolution