检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Zhu, Wentao Jin, Yuan Ma, Gege Chen, Geng Egger, Jan Zhang, Shaoting Metaxas, Dimitris N. Research Center for Healthcare Data Science Zhejiang Lab Hangzhou311121 China School of Computer Science and Engineering Northwestern Polytechnical University Shaanxi Xi’an710072 China Institute of Computer Graphics and Vision Graz University of Technology Graz8010 Austria Shanghai Artificial Intelligence Laboratory Shanghai200120 China Department of Computer Science Rutgers University PiscatawayNJ08854 United States

The accurate diagnosis on pathological subtypes for lung cancer is of significant importance for the follow-up treatments and prognosis managements. In this paper, we propose self-generating hybrid feature network (SGHF-Net) for accurately classifying lung cancer subtypes on computed tomography (CT) images. Inspired by studies stating that cross-scale associations exist in the image patterns between the same case’s CT images and its pathological images, we innovatively developed a pathological feature synthetic module (PFSM), which quantitatively maps cross-modality associations through deep neural networks, to derive the "gold standard" information contained in the corresponding pathological images from CT images. Additionally, we designed a radiological feature extraction module (RFEM) to directly acquire CT image information and integrated it with the pathological priors under an effective feature fusion framework, enabling the entire classification model to generate more indicative and specific pathologically related features and eventually output more accurate predictions. The superiority of the proposed model lies in its ability to self-generate hybrid features that contain multi-modality image information based on a single-modality input. To evaluate the effectiveness, adaptability, and generalization ability of our model, we performed extensive experiments on a large-scale multi-center dataset (i.e., 829 cases from three hospitals) to compare our model and a series of state-of-the-art (SOTA) classification models. The experimental results demonstrated the superiority of our model for lung cancer subtypes classification with significant accuracy improvements in terms of accuracy (ACC), area under the curve (AUC), and F1 score. © 2023, CC0.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Traceable and Authenticable Image Tagging for Fake News Detection

arXiv

引用

arXiv 2022年

作者： Meng, Ruohan Zhou, Zhili Cui, Qi Lam, Kwok-Yan Kot, Alex School of Computer and Software Nanjing University of Information Science and Technology China Institute of Artificial Intelligence and Blockchain Guangzhou University China Computer Science and Engineering Nanyang Technological University Singapore Lab Nanyang Technological University Singapore Centre for Computer Vision and Deep Learning University of Windsor Canada

To prevent fake news images from misleading the public, it is desirable not only to verify the authenticity of news images but also to trace the source of fake news, so as to provide a complete forensic chain for reliable fake news detection. To simultaneously achieve the goals of authenticity verification and source tracing, we propose a traceable and authenticable image tagging approach that is based on a design of Decoupled Invertible Neural Network (DINN). The designed DINN can simultaneously embed the dual-tags, i.e., authenticable tag and traceable tag, into each news image before publishing, and then separately extract them for authenticity verification and source tracing. Moreover, to improve the accuracy of dual-tags extraction, we design a parallel Feature Aware Projection Model (FAPM) to help the DINN preserve essential tag information. In addition, we define a Distance Metric-Guided Module (DMGM) that learns asymmetric one-class representations to enable the dual-tags to achieve different robustness performances under malicious manipulations. Extensive experiments, on diverse datasets and unseen manipulations, demonstrate that the proposed tagging approach achieves excellent performance in the aspects of both authenticity verification and source tracing for reliable fake news detection and outperforms the prior works. Copyright © 2022, The Authors. All rights reserved.

关键词： Image annotation

来源：评论

学校读者我要写书评

暂无评论

On the analysis of HEVC Intra Prediction Mode Decision Variants 3

On the analysis of HEVC Intra Prediction Mode Decision Varia...

引用

3rd International Conference on Computing and Network Communications, CoCoNet 2019

作者： Nair, Preethi S. Nair, Madhu S. Department of Computer Science University of Kerala Kariavattom Thiruvananthapuram Kerala695581 India Artificial Intelligence and Computer Vision Lab Department of Computer Science Cochin University of Science and Technology Kochi Kerala682022 India

This paper presents a comparative analysis of some major High Efficiency Video Coding (HEVC) intra prediction mode decision methods reported in the literature. Intra coding in HEVC is based on spatial sample prediction and can be described as a collection of many elements including quad tree partitioning, 35 intra prediction modes, reference sample smoothing, boundary sample filtering process, intra mode coding and so on. Each of these elements results in an improved coding efficiency in HEVC intra coding. In this review, we provide an extensive and up-to-date discussion of the research proposals regarding a variety of HEVC intra prediction mode decision strategies. Moreover, some most significant research contributions on fast intra mode decision methods in HEVC to improvise the encoder efficiency are reviewed in detail along with a comparative performance analysis of the reviewed proposals in terms of some standard evaluation metrics such as bit rate, PSNR and time savings. In addition, this paper point out some research directions towards further developments in the intra prediction for the betterment of the encoder efficacy. © 2020 The Authors. Published by Elsevier B.V.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

BiLSTM-Autoencoder Architecture for Stance Prediction

BiLSTM-Autoencoder Architecture for Stance Prediction

引用

2020 International Conference on Data Science and Engineering, ICDSE 2020

作者： Padnekar, S Meena Kumar, G Santhosh Deepak, P. Cochin University of Science and Technology Artificial Intelligence and Computer Vision Lab Department of Computer Science Kochi India Queen's University School of Electronics Electrical Engineering and Computer Science Belfast United Kingdom

ISBN: (纸本)9781728189192

The recent surge in the abundance of fake news appearing on social media and news websites poses a potential threat to high-quality journalism. Misinformation hurts people, society, science, and democracy. This reason has led many researchers to develop techniques to identify fake news. In this paper, we discuss a stance prediction technique using the Deep Learning approach, which can be used as a factor to determine the authenticity of news articles. The Fake News Stance Prediction is the process of automatically classifying the stance of a news article towards a target into one of the following classes: Agree, Disagree, Discuss, Unrelated. The stance prediction task's input is the news articles containing a pair: a headline as the target and a body as a claim. This paper proposes a deep learning architecture using Bi-directional Long Short Term Memory and Autoencoder for stance prediction. We illustrate, through empirical studies, that the method is reasonably accurate at predicting stance, achieving a classification accuracy as high as 94%. The proposed stance detection method would be useful for assessing the credibility of news articles. © 2020 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Generalized real-world super-resolution through adversarial robustness

arXiv

引用

arXiv 2021年

作者： Castillo, Angela Escobar, María Pérez, Juan C. Romero, Andrés Timofte, Radu van Gool, Luc Arbeláez, Pablo Center for Research and Formation in Artificial Intelligence Universidad de los Andes Colombia Saudi Arabia Computer Vision Lab ETH Zürich Switzerland

Real-world Super-Resolution (SR) has been traditionally tackled by first learning a specific degradation model that resembles the noise and corruption artifacts in low-resolution imagery. Thus, current methods lack generalization and lose their accuracy when tested on unseen types of corruption. In contrast to the traditional proposal, we present Robust Super-Resolution (RSR), a method that leverages the generalization capability of adversarial attacks to tackle real-world SR. Our novel framework poses a paradigm shift in the development of real-world SR methods. Instead of learning a dataset-specific degradation, we employ adversarial attacks to create difficult examples that target the model's weaknesses. Afterward, we use these adversarial examples during training to improve our model's capacity to process noisy inputs. We perform extensive experimentation on synthetic and real-world images and empirically demonstrate that our RSR method generalizes well across datasets without re-training for specific noise priors. By using a single robust model, we outperform state-of-the-art specialized methods on real-world benchmarks. © 2021, CC BY-NC-SA.

关键词： Optical resolving power

来源：评论

学校读者我要写书评

暂无评论

Activating More Pixels in Image Super-Resolution Transformer

arXiv

引用

arXiv 2022年

作者： Chen, Xiangyu Wang, Xintao Zhou, Jiantao Qiao, Yu Dong, Chao State Key Laboratory of Internet of Things for Smart City University of Macau China Shenzhen Key Lab of Computer Vision and Pattern Recognition Shenzhen Institute of Advanced Technology Chinese Academy of Sciences China Shanghai Artificial Intelligence Laboratory China ARC Lab Tencent PCG China

Transformer-based methods have shown impressive performance in low-level vision tasks, such as image super-resolution. However, we find that these networks can only utilize a limited spatial range of input information through attribution analysis. This implies that the potential of Transformer is still not fully exploited in existing networks. In order to activate more input pixels for better reconstruction, we propose a novel Hybrid Attention Transformer (HAT). It combines both channel attention and window-based self-attention schemes, thus making use of their complementary advantages of being able to utilize global statistics and strong local fitting capability. Moreover, to better aggregate the cross-window information, we introduce an overlapping cross-attention module to enhance the interaction between neighboring window features. In the training stage, we additionally adopt a same-task pre-training strategy to exploit the potential of the model for further improvement. Extensive experiments show the effectiveness of the proposed modules, and we further scale up the model to demonstrate that the performance of this task can be greatly improved. Our overall method significantly outperforms the state-of-the-art methods by more than 1dB. Copyright © 2022, The Authors. All rights reserved.

关键词： Pixels

来源：评论

学校读者我要写书评

暂无评论

UniFormer: Unifying Convolution and Self-attention for Visual Recognition

arXiv

引用

arXiv 2022年

作者： Li, Kunchang Wang, Yali Zhang, Junhao Gao, Peng Song, Guanglu Liu, Yu Li, Hongsheng Qiao, Yu ShenZhen Key Lab of Computer Vision and Pattern Recognition Shenzhen Institute of Advanced Technology Chinese Academy of Sciences Shenzhen518055 China University of Chinese Academy of Sciences Beijing100049 China Shanghai Artificial Intelligence Laboratory Shanghai200232 China National University of Singapore Singapore Shanghai Artificial Intelligence Laboratory China SenseTime Research China The Chinese University of Hong Kong Hong Kong

It is a challenging task to learn discriminative representation from images and videos, due to large local redundancy and complex global dependency in these visual data. Convolution neural networks (CNNs) and vision transformers (ViTs) have been two dominant frameworks in the past few years. Though CNNs can efficiently decrease local redundancy by convolution within a small neighborhood, the limited receptive field makes it hard to capture global dependency. Alternatively, ViTs can effectively capture long-range dependency via self-attention, while blind similarity comparisons among all the tokens lead to high redundancy. To resolve these problems, we propose a novel Unified transFormer (UniFormer), which can seamlessly integrate the merits of convolution and self-attention in a concise transformer format. Different from the typical transformer blocks, the relation aggregators in our UniFormer block are equipped with local and global token affinity respectively in shallow and deep layers, allowing tackling both redundancy and dependency for efficient and effective representation learning. Finally, we flexibly stack our blocks into a new powerful backbone, and adopt it for various vision tasks from image to video domain, from classification to dense prediction. Without any extra training data, our UniFormer achieves 86.3 top-1 accuracy on ImageNet-1K classification task. With only ImageNet-1K pre-training, it can simply achieve state-of-the-art performance in a broad range of downstream tasks. It obtains 82.9/84.8 top-1 accuracy on Kinetics-400/600, 60.9/71.2 top-1 accuracy on Something-Something V1/V2 video classification tasks, 53.8 box AP and 46.4 mask AP on COCO object detection task, 50.8 mIoU on ADE20K semantic segmentation task, and 77.4 AP on COCO pose estimation task. Moreover, we build an efficient UniFormer with a concise hourglass design of token shrinking and recovering, which achieves 2-4× higher throughput than the recent lightweight models. Code is av

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Density-aware and Depth-aware Visual Representation for Zero-Shot Object Counting

Density-aware and Depth-aware Visual Representation for Zero...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Fang Nan Feng Tian Ni Zhang Nian Liu Haonan Miao Guang Dai Mengmeng Wang Faculty of Electronic and Information Engineering Xi’an Jiaotong University Xi’an China Ministry of Education Key Laboratory of Intelligent Networks and Network Security Xi’an Jiaotong University Xi’an China School of Automation Northwestern Polytechnical University Xi’an China Computer Vision Department Mohamed bin Zayed University of Artificial Intelligence Abu Dhabi UAE SGIT AI Lab State Grid Corporation of China Zhejiang University of Technology Hang Zhou China

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Previous methods often utilize CLIP semantic classifiers with class names for zero-shot object counting. However, they ignore crucial density and depth knowledge for counting tasks. Thus, we propose a density-aware and depth-aware prompt counting model, which captures density information via learning density-aware prompts based on density-aware contrastive loss and incorporates depth guidance with predefined depth-aware prompts. To facilitate the training process, we design two strategies for standard counting loss and the contrastive loss, where the former prioritizes larger and sparser objects initially, gradually focusing on smaller and denser objects, and the latter adopts coarse-to-fine density learning. Besides, we construct a dataset named LVIS-372 with more real-world scenarios and balanced instance distribution compared to existing ones. Finally, the experimental results demonstrate the effectiveness of our proposed method.

关键词： Training Visualization Semantics Focusing Signal processing Acoustics Power capacitors Speech processing Standards

来源：评论

学校读者我要写书评

暂无评论

An interactive system to improve cognitive abilities using electromyographic signals 21

An interactive system to improve cognitive abilities using e...

引用

Proceedings of the 5th International Conference on Advances in artificial intelligence

作者： Marco Enrique Benalcázar Palacios Xavier Aguas Ángel Leonardo Valdivieso Caraguay Lorena Isabel Barona López Rubén Nogales Jaime Guilcapi Freddy Benalcázar Artificial intelligence and Computer Vision Research Lab Departamento de Informática y Ciencias de la Computación(DiCC) Escuela Politecnica Nacional Ecuador Artificial intelligence and Computer Vision Research Lab Departamento de Informática y Ciencias de la Computación(DiCC) Escuela Politécnica Nacional Ecuador Facultad de ingeniería en Sistemas Electrónica e industrial Universidad Técnica de Ambato Ecuador Facultad de Ingeniería en Sistemas Electrónica e Industrial Universidad Técnica de Ambato Ecuador

ISBN: (纸本)9781450390699

This article presents a system for improving people’s cognitive abilities using electromyography (EMG) signals. This system was created under the serious game mode with a new way of interaction with the user. The sensor Myo Armband was used to measure the EMG signals, which sends the signals via Bluetooth to a Hand Gesture Recognition Model based on a Support Vector Machine to detect the gesture made by the player. This Hand-Gesture Recognition Model was developed in Matlab 2020a. The correct commands given by the recognition model are sent to another computer through the TCP / IP communication protocol to visualize the effects on the game character. The game was developed with the Unity 3D graphics engine to generate an application that optimizes resources. The objective of the game is to solve mazes with two levels of difficulty which allow improving cognitive skills such as attention, concentration, observation, abstract thinking, psycho-motor skills and problem solving.

关键词： Matlab Serious Game Hand-Gesture Recognition Mode Unity 3D Support Vector Machine

来源：评论

学校读者我要写书评

暂无评论

Low-Resolution Action Recognition for Tiny Actions Challenge

arXiv

引用

arXiv 2022年

作者： Chen, Boyu Qiao, Yu Wang, Yali ShenZhen Key Lab of Computer Vision and Pattern Recognition Shenzhen Institute of Advanced Technology Chinese Academy of Sciences China University of Chinese Academy of Sciences China Shanghai AI Laboratory Shanghai China SIAT Branch Shenzhen Institute of Artificial Intelligence and Robotics for Society China

Tiny Actions Challenge focuses on understanding human activities in real-world surveillance. Basically, there are two main difficulties for activity recognition in this scenario. First, human activities are often recorded at a distance, and appear in a small resolution without much discriminative clue. Second, these activities are naturally distributed in a long-tailed way. It is hard to alleviate data bias for such heavy category imbalance. To tackle these problems, we propose a comprehensive recognition solution in this paper. First, we train video backbones with data balance, in order to alleviate overfitting in the challenge benchmark. Second, we design a dual-resolution distillation framework, which can effectively guide low-resolution action recognition by super-resolution knowledge. Finally, we apply model ensemble with post-processing, which can further boost performance on the long-tailed categories. Our solution ranks Top-1 on the leaderboard. Copyright © 2022, The Authors. All rights reserved.

关键词： Distillation

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：