检索结果-内蒙古大学图书馆

11th International Conference on Language Resources and Evaluation (LREC)

作者： Wiedemann, Gregor Heyer, Gerhard Univ Leipzig Dept Comp Sci Leipzig Germany

ISBN: (纸本)9791095546009

In recent years, (retro-)digitizing paper-based files became a major undertaking for private and public archives as well as an important task in electronic mailroom applications. As a first step, the workflow involves scanning and Optical Character Recognition (OCR) of documents. Preservation of document contexts of single page scans is a major requirement in this context. To facilitate workflows involving very large amounts of paper scans, page stream segmentation (PSS) is the task to automatically separate a stream of scanned images into multi-page documents. In a digitization project together with a German federal archive, we developed a novel approach based on convolutional neural networks (CNN) combining image and text features to achieve optimal document separation results. Evaluation shows that our PSS architecture achieves an accuracy up to 93 % which can be regarded as a new state-of-the-art for this task.

关键词： page stream segmentation convolutional neural nets document image classification document management text classification

来源：评论

学校读者我要写书评

暂无评论

Bilingual Word Embeddings for Cross-Lingual Personality Recognition Using convolutional neural nets 18

Bilingual Word Embeddings for Cross-Lingual Personality Reco...

引用

18th Annual Conference of the International-Speech-Communication-Association (INTERSPEECH 2017)

作者： Bin Siddique, Farhad Fung, Pascale Hong Kong Univ Sci & Technol Dept Elect & Comp Engn Human Language Technol Ctr Clear Water Bay Hong Kong Peoples R China

ISBN: (纸本)9781510848764

We propose a multilingual personality classifier that uses text data from social media and Youtube Vlog transcriptions, and maps them into Big Five personality traits using a convolutional neural Network (CNN). We first train unsupervised bilingual word embeddings from an English-Chinese parallel corpus, and use these trained word representations as input to our CNN. This enables our model to yield relatively high cross-lingual and multilingual performance on Chinese texts, after training on the English dataset for example. We also train monolingual Chinese embeddings from a large Chinese text corpus and then train our CNN model on a Chinese dataset consisting of conversational dialogue labeled with personality. We achieve an average F-score of 66.1 in our multilingual task compared to 63.3 F-score in cross-lingual. and 63.2 F-score in the monolingual performance.

关键词： Computational Personality Recognition Big Five Multilingual Analysis convolutional neural nets

来源：评论

学校读者我要写书评

暂无评论

A unified multi-class anomaly detection model based on reverse distillation

引用

IET IMAGE PROCESSING 2025年第1期19卷

作者： Fu, Maoli Fu, Zhongliang Chinese Acad Sci ChengDu Inst Comp Applicat Chengdu Sichuan Peoples R China Univ Chinese Acad Sci Chengdu Inst Comp Applicat Beijing Peoples R China Shenzhen CBPM KEXIN Banking Technol CO LTD R&D Dept Banknote Printing Prod Line Shenzhen Peoples R China

The multi-class single-model detection paradigm is a prevalent design for industrial anomaly detection, exhibiting suitability for varying industrial classes and dynamic, flexible production scenarios. This approach not only enhances model adaptability but also minimizes maintenance costs. However, the current popular methods are susceptible to the 'copying shortcut' phenomenon, which constrains their performance on benchmark datasets. To overcome this limitation, this article proposes a multi-class anomaly detection model: UniRD, based on the reverse distillation method. This model creates an image corruptor that expands the dataset and generates 'normal-corrupt' image pairs. During the training process, their correspondence is used to optimize the reverse distillation. This process greatly exploits and releases the potential of the student decoder. Furthermore, a teacher feature adaptation module is devised to enhance the compatibility between the pre-trained model and the anomaly detection task. This has the effect of reducing the discrepancy between teacher and student features while ensuring the consistency of normal sample features. The comprehensive evaluation results in two mainstream datasets, MVTec and VisA, and demonstrates that the proposed method exhibits improvement in all indicators compared to benchmark methods. The proposed method attains the state-of-the-art, substantiating its effectiveness and advancement.

关键词： computer vision convolution convolutional neural nets encoding feature extraction image processing

来源：评论

学校读者我要写书评

暂无评论

Generating Transferable Adversarial Point Clouds via Autoencoders for 3D Object Classification

引用

IET COMPUTER VISION 2025年第1期19卷

作者： Xu, Mengyao Chen, Hai Zhang, Chonghao Zou, Yuanjun Xu, Chenchu Zhang, Yanping Qian, Fulan Anhui Univ Sch Comp Sci & Technol Hefei Peoples R China Anhui Univ Artificial Intelligence Inst Hefei Peoples R China Anhui Univ Informat Mat & Intelligent Sensing Lab Anhui Prov Hefei Peoples R China

Recent studies have shown that deep neural networks are vulnerable to adversarial attacks. In the field of 3D point cloud classification, transfer-based black-box attack strategies have been explored to address the challenge of limited knowledge about the model in practical scenarios. However, existing approaches typically rely excessively on network structure, resulting in poor transferability of the generated adversarial examples. To address the above problem, the authors propose AEattack, an adversarial attack method capable of generating highly transferable adversarial examples. Specifically, AEattack employs an autoencoder (AE) to extract features from the point cloud data and reconstruct the adversarial point cloud based on these features. Notably, the AE does not require pre-training, and its parameters are jointly optimised using a loss function during the process of generating adversarial point clouds. The method makes the generated adversarial point cloud not overly dependent on the network structure, but more concerned with the data distribution. Moreover, this design endows AEattack with a broader potential for application. Extensive experiments on the ModelNet40 dataset show that AEattack is capable of generating highly transferable adversarial point clouds, with up to 61.8% improvement in transferability compared to state-of-the-art adversarial attacks.

关键词： computer vision convolutional neural nets feature extraction object recognition

来源：评论

学校读者我要写书评

暂无评论

Unsupervised image blind super resolution via real degradation feature learning

引用

IET COMPUTER VISION 2024年第4期18卷 485-498页

作者： Yang, Cheng Lu, Guanming Nanjing Univ Posts & Telecommun Coll Telecommun & Informat Engn Nanjing Peoples R China Changzhou Coll Informat Technol Changzhou Peoples R China

In recent years, many methods for image super-resolution (SR) have relied on pairs of low-resolution (LR) and high-resolution (HR) images for training, where the degradation process is predefined by bicubic downsampling. While such approaches perform well in standard benchmark tests, they often fail to accurately replicate the complexity of real-world image degradation. To address this challenge, researchers have proposed the use of unpaired image training to implicitly model the degradation process. However, there is a significant domain gap between the real-world LR and the synthetic LR images from HR, which severely degrades the SR performance. A novel unsupervised image-blind super-resolution method that exploits degradation feature-based learning for real-image super-resolution reconstruction (RDFL) is proposed. Their approach learns the degradation process from HR to LR using a generative adversarial network (GAN) and constrains the data distribution of the synthetic LR with real degraded images. The authors then encode the degraded features into a Transformer-based SR network for image super-resolution reconstruction through degradation representation learning. Extensive experiments on both synthetic and real datasets demonstrate the effectiveness and superiority of the RDFL method, which achieves visually pleasing reconstruction results. Their approach learns the degradation process from HR to LR using a generative adversarial network (GAN) and constrains the data distribution of the synthetic LR with real degraded images. The authors then encode the degraded features into a Transformer-based SR network for image super-resolution reconstruction through degradation representation ***

关键词： computer vision convolutional neural nets feature extraction image processing image reconstruction image restoration neural nets

来源：评论

学校读者我要写书评

暂无评论

From pixels to prognosis: Attention-CNN model for COVID-19 diagnosis using chest CT images

引用

IET IMAGE PROCESSING 2025年第1期19卷

作者： Suseela, Suba Parekh, Nita Int Inst Informat Technol Ctr Computat Nat Sci & Bioinformat Prof CR Rao Rd Hyderabad 500032 Telangana India

Deep learning assisted diagnosis for assessing the severity of various respiratory infections using chest computed tomography (CT) scan images has gained much attention after the COVID-19 pandemic. Major tasks while building such models require an understanding of the characteristic features associated with the disease, patient-to-patient variations and changes associated with disease severity. In this work, an attention-based convolutional neural network (CNN) model with customized bottleneck residual module (Attn-CNN) is proposed for classifying CT images into three classes: COVID-19, normal, and other pneumonia. The efficacy of the model is evaluated by carrying out various experiments, such as effect of class imbalance, impact of attention module, generalizability of the model and providing visualization of model's prediction for the interpretability of results. Comparative performance evaluation with five state-of-the-art deep architectures such as MobileNet, EfficientNet-B7, Inceptionv3, ResNet-50 and VGG-16, and with published models such as COVIDNet-CT, COVNet, COVID-Net CT2, etc. is discussed.

关键词： computerised tomography convolutional neural nets health care image classification learning (artificial intelligence) lung medical image processing neural net architecture

来源：评论

学校读者我要写书评

暂无评论

Zero-shot moving ship segmentation based on segment anything network and optical flow network

引用

ELECTRONICS LETTERS 2025年第1期61卷

作者： Liu, Wenhui Qiao, Yulong Xing, Zhengyi Zhao, Yue Harbin Engn Univ Coll Informat & Commun Engn Harbin Peoples R China

The wake of the ship and the waves on the water surface increased the difficulty of video-based ship segmentation. To address this challenging problem, we propose a zero-shot segmentation method based on the segment anything model (SAM) and optical flow network. Specifically, the dark channel filtering technology is employed as a preprocessing step to partly eliminate the noise from the sea surface. Then, the SAM with the point prompt is utilized to detect and segment ship targets in conjunction with finding the best mask algorithms, in which a supervision mechanism is introduced to determine the ship moving in or out of the surveillance region. Finally, we use the optical flow network to predict the position of the ship target and leverage the segmentation by using the network SAM with bounding boxes as prompts. The experiments on the benchmark dataset IPATCH demonstrate the performance of the proposed zero-shot moving ship segmentation method.

关键词： artificial intelligence computer vision convolutional neural nets image segmentation ships video signal processing

来源：评论

学校读者我要写书评

暂无评论

Noise2Variance: Dual networks with variance constraint for self-supervised real-world image denoising

引用

IET IMAGE PROCESSING 2024年第12期18卷 3251-3261页

作者： Tan, Hanlin Liu, Yu Zhang, Maojun Natl Univ Def Technol Coll Syst Engn Changsha 410073 Peoples R China

Image denoising aims to restore a clean image from a noisy image. Traditional methods utilizing convolutional neural networks (CNN) for denoising are trained using pairs of noisy and clean images to comprehend the transformation from a noisy image to a clean one. However, the acquisition of such image pairs in real-world scenarios presents a challenge. Hence, numerous self-supervised denoising techniques have been developed that do not require clean images for training. This study demonstrates that a straightforward loss design, concentrating on variance, can effectively train a standard CNN denoiser in a self-supervised fashion. A novel theoretical framework is introduced for training a basic CNN denoising model using three constraints: mean, variance, and augmentation. The variance constraint is crucial as it prevents the trained model from converging to trivial solutions such as identity or zero mapping. This theory provides valuable insights for the development of new self-supervised denoising methods. Furthermore, a method that applies this theory to proposed dual networks is developed, which consist of two standard CNN models predicting both the clean image and the noise. This approach enhances model capacity during training while minimizing computational costs during inference. This method exemplifies the implementation of the variance constraint and introduces a data constraint for dual networks. Notably, the proposed method only assumes the presence of additive white noise, irrespective of the noise distribution. This minimal assumption enhances the model's robustness against noise with complex or unknown distributions in real-world distorted images. Experimental results indicate that the proposed Noise2Variance method exhibits commendable performance on peak signal noise ratio and structural similarity metrics compared to existing self-supervised denoising techniques. Visual comparison of results further substantiates the efficacy of the proposed method. A

关键词： convolutional neural nets image denoising unsupervised learning

来源：评论

学校读者我要写书评

暂无评论

Modelling appearance variations in expressive and neutral face image for automatic facial expression recognition

引用

IET IMAGE PROCESSING 2024年第9期18卷 2449-2460页

作者： Kumar, Naveen H. N. Prasad, Guru M. S. Shah, Mohd Asif Mahadevaswamy Jagadeesh, B. Sudheesh, K., V Vidyavardhaka Coll Engn Dept Elect & Commun Engn Mysuru Karnataka India Graph Era Deemed Univ Dept Comp Sci & Engn Dehra Dun India Kebri Dehar Univ Coll Business & Econ Dept Econ POB 250 Kebri Dehar Somali Ethiopia Chitkara Univ Chitkara Univ Inst Engn & Technol Ctr Res Impact & Outcome Rajpura Punjab India Lovely Profess Univ Div Res & Dev Phagwara Punjab India

In automatic facial expression recognition (AFER) systems, modelling the spatio-temporal feature information in a specific manner, coalescing, and its effective utilization is challenging. The state-of-the-art studies have examined integrating multiple features to enhance the recognition rate of AFER systems. However, the feature variations between expressive and neutral face images are not fully explored to identify the expression class. The proposed research presents an innovative approach to AFER by modelling appearance variations in both expressive and neutral face images. The prominent contributions of the work are developing a novel and hybrid feature space by integrating the discriminative feature distribution derived from expressive and neutral face images;preserving the highly discriminative latent feature distribution using autoencoders. Local binary pattern (LBP) and histogram of oriented gradients (HOG) are the feature descriptors employed to derive the discriminative texture and shape information, respectively. The component-based approach is employed, wherein the features are derived from the salient facial regions instead of the whole face. The three-stage stacked deep convolutional autoencoder (SDCA) and multi-class support vector machine (MSVM) are employed to address dimensionality reduction and classification, respectively. The efficacy of the proposed model is substantiated by empirical findings, which establish its superiority in terms of accuracy in AFER tasks on widely recognized benchmark datasets. This paper develops a hybrid feature space by integrating the discriminative power of HOG and LBP feature distributions derived from the salient facial regions. Also, a novel feature space is developed by integrating the hybrid feature space derived from the expressive and neutral face images to further amplify the discriminative power of the feature descriptor. A three-stage stacked deep convolutional autoencoder (SDCA) derives the abstract repres

关键词： convolutional neural nets emotion recognition feature selection image classification

来源：评论

学校读者我要写书评

暂无评论

Determining the proper number of proposals for individual images

引用

IET COMPUTER VISION 2024年第1期18卷 141-149页

作者： He, Zihang Li, Yong Beijing Univ Posts & Telecommun Sch Elect Engn Beijing Peoples R China

The region proposal network is indispensable to two-stage object detection methods. It generates a fixed number of proposals that are to be classified and regressed by detection heads to produce detection boxes. However, the fixed number of proposals may be too large when an image contains only a few objects but too small when it contains much more objects. Considering this, the authors explored determining a proper number of proposals according to the number of objects in an image to reduce the computational cost while improving the detection accuracy. Since the number of ground truth objects is unknown at the inference stage, the authors designed a simple but effective module to predict the number of foreground regions, which will be substituted for the number of objects for determining the proposal number. Experimental results of various two-stage detection methods on different datasets, including MS-COCO, PASCAL VOC, and CrowdHuman showed that equipping the designed module increased the detection accuracy while decreasing the FLOPs of the detection head. For example, experimental results on the PASCAL VOC dataset showed that applying the designed module to Libra R-CNN and Grid R-CNN increased over 1.5 AP(50) while decreasing the FLOPs of detection heads from 28.6 G to nearly 9.0 G.

关键词： computer vision convolutional neural nets object detection

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：