检索结果-内蒙古大学图书馆

arXiv 2023年

作者： Yang, Zuopeng Chu, Tianshu Lin, Xin Gao, Erdun Liu, Daqing Yang, Jie Wang, Chaoyue Institute of Image Processing and Pattern Recognition Department of Automation Shanghai Jiao Tong University Shanghai200240 China the Institute of Artificial Intelligence and Blockchain Guangzhou University Guangzhou510006 China the School of Mathematics and Statistics The University of Melbourne MelbourneVIC3010 Australia the JD Explore Academy Beijing100176 China

The recent success of text-to-image generation diffusion models has also revolutionized semantic image editing, enabling the manipulation of images based on query/target texts. Despite these advancements, a significant challenge lies in the potential introduction of contextual prior bias in pre-trained models during image editing, e.g., making unexpected modifications to inappropriate regions. To address this issue, we present a novel approach called Dual-Cycle Diffusion, which generates an unbiased mask to guide image editing. The proposed model incorporates a Bias Elimination Cycle that consists of both a forward path and an inverted path, each featuring a Structural Consistency Cycle to ensure the preservation of image content during the editing process. The forward path utilizes the pre-trained model to produce the edited image, while the inverted path converts the result back to the source image. The unbiased mask is generated by comparing differences between the processed source image and the edited image to ensure that both conform to the same distribution. Our experiments demonstrate the effectiveness of the proposed method, as it significantly improves the D-CLIP score from 0.272 to 0.283. The code will be available at https://***/JohnDreamer/DualCycleDiffsion. © 2023, CC BY-NC-SA.

关键词： Diffusion

来源：评论

学校读者我要写书评

暂无评论

Rephrasing the Reference for Non-Autoregressive Machine Translation

arXiv

引用

arXiv 2022年

作者： Shao, Chenze Zhang, Jinchao Zhou, Jie Feng, Yang Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences China University of Chinese Academy of Sciences China Pattern Recognition Center WeChat AI Tencent Inc China

Non-autoregressive neural machine translation (NAT) models suffer from the multi-modality problem that there may exist multiple possible translations of a source sentence, so the reference sentence may be inappropriate for the training when the NAT output is closer to other translations. In response to this problem, we introduce a rephraser to provide a better training target for NAT by rephrasing the reference sentence according to the NAT output. As we train NAT based on the rephraser output rather than the reference sentence, the rephraser output should fit well with the NAT output and not deviate too far from the reference, which can be quantified as reward functions and optimized by reinforcement learning. Experiments on major WMT benchmarks and NAT baselines show that our approach consistently improves the translation quality of NAT. Specifically, our best variant achieves comparable performance to the autoregressive Transformer, while being 14.7 times more efficient in inference. Copyright © 2022, The Authors. All rights reserved.

关键词： Neural machine translation

来源：评论

学校读者我要写书评

暂无评论

Unsupervised Learning of Depth and Pose Based on Monocular Camera and Inertial Measurement Unit (IMU)

Unsupervised Learning of Depth and Pose Based on Monocular C...

引用

IEEE International Conference on Robotics and Automation (ICRA)

作者： Yanbo Wang Hanwen Yang Jianwei Cai Guangming Wang Jingchuan Wang Yi Huang University of Michigan-Shanghai Jiao Tong University Joint Institute Shanghai Jiao Tong University Shanghai China School of Electronic Information and Electrical Engineering Shanghai Jiao Tong University Shanghai China Institute of Image Processing and Pattern Recognition Shanghai Jiao Tong University Shanghai China Department of Automation Shanghai Jiao Tong University Shanghai China Key Laboratory of System Control and Information Processing Ministry of Education of China Shanghai Weitong Vision Technology Co. Ltd.

The main content of the research in this paper is the estimation of depth and pose based on monocular vision and Inertial Measurement Unit (IMU). The usual depth estimation network and pose estimation network require depth ground truth or pose ground truth as a supervised signal for training, while the depth ground truth and pose ground truth are hard to obtain, and monocular vision based depth estimation cannot predict absolute depth. In this paper, with the help of IMU, which is inexpensive and widely used, we can obtain angular velocity and acceleration information. Two new supervision signals are proposed and the calculation expressions are given. Among them, the model trained with acceleration constraint shows a good ability to estimate the absolute depth during the test. It can be considered that the model can estimate the absolute depth. We also derive the method of estimating the scale factor during the test from the acceleration constraint, and also achieve good results as the acceleration constraint does. In addition, this paper also studies the method of using IMU information as pose network input and as selecting conditions. Moreover, it analyzes and discusses the experimental results. At the same time, we also evaluate the effect of the pose estimation of the relevant models. This article starts by reviewing the achievements and deficiencies of the work in this field, combines the use of IMU, puts forward three new methods such as a new loss function, and conducts a test analysis and discussion of relevant indicators on the KITTI data set.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Universal Adversarial Perturbation Generated by Attacking Layer-wise Relevance Propagation 10

Universal Adversarial Perturbation Generated by Attacking La...

引用

10th IEEE International Conference on Intelligent Systems, IS 2020

作者： Wang, Zifei Huang, Xiaolin Yang, Jie Kasabov, Nikola Shanghai Jiao Tong University Institute of Image Processing and Pattern Recognition Shanghai China Auckland University of Technology Knowledge Engineering and Discovery Research Institute Auckland New Zealand

ISBN: (纸本)9781728154565

The vulnerability of Deep Neural Networks (DNNs) to adversarial attacks has become an important research area of machine learning. It has been known that many state-of-the-art DNNs suffer the risk of universal adversarial perturbations, which are image-agnostic and able to lead misclassifications with high probability. In this paper, we propose a novel method to create such universal adversarial perturbations. Our approach is the first to generate universal perturbations by attacking the attention heat maps with the interpretation method, Layer-wise Relevance Propagation. It is demonstrated that our method achieves high fooling ratios on image classification DNNs pre-trained by imageNet dataset. Moreover, our attack shows good transferability across different DNNs. © 2020 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Synchronization of Nonlinear Neural Networks with Hybrid Couplings and Uncertain Time-Varying Perturbations: A Novel Distributed-Delay Impulsive Comparison Principle

SSRN

引用

SSRN 2024年

作者： Fan, Hongguang Shi, Kaibo Zhao, Yi College of Computer Chengdu University Chengdu610106 China Key Laboratory of Pattern Recognition and Intelligent Information Processing Institutions of Higher Education of Sichuan Province Chengdu University Chengdu610106 China School of Electronic Information and Electrical Engineering Chengdu University Chengdu610106 China School of Mathematical Sciences Shenzhen University Shenzhen518060 China

In this paper, the synchronization of nonlinear drive-response neural networks with uncertain time-varying perturbations, non-delayed coupling, and distributed delay coupling is studied. To address the impact of distributed delay and discrete delay on the system, a novel impulsive comparison principle is established, which can be viewed as an effective continuation of the Halanay inequality. Besides, with the help of Lyapunov stability theory, sufficient conditions guaranteeing the exponential synchronization of concerned neural networks are derived by using a delayed impulsive controller with historical status information, which releases the conventional limitation that impulsive delays should be smaller than impulsive intervals, and generalize the existing synchronization results regarding distributed delay networks. Besides, numerical simulations for chaotic neural networks are presented to demonstrate the validity of theoretical results and the sensitivity of the control gain matrix. © 2024, The Authors. All rights reserved.

关键词： Complex networks

来源：评论

学校读者我要写书评

暂无评论

image recognition Based on Combined Filters with Pseudoinverse Learning Algorithm 15th

Image Recognition Based on Combined Filters with Pseudoinver...

引用

15th IFIP WG 12.5 International Conference on Artificial Intelligence Applications and Innovations, AIAI 2019

作者： Deng, Xiaodan Sun, Xiaoxuan Guo, Ping Yin, Qian Image Processing and Pattern Recognition Laboratory School of Systems Science Beijing Normal University Beijing100875 China Image Processing and Pattern Recognition Laboratory College of Information Science and Technology Beijing Normal University Beijing China

ISBN: (纸本)9783030198220

Deep convolution neural network (CNN) is one of the most popular Deep neural networks (DNN). It has won state-of-the-art performance in many computer vision tasks. The most used method to train DNN is Gradient descent-based algorithm such as Backpropagation. However, backpropagation algorithm usually has the problem of gradient vanishing or gradient explosion, and it relies on repeated iteration to get the optimal result. Moreover, with the need to learn many convolutional kernels, the traditional convolutional layer is the main computational bottleneck of deep CNNs. Consequently, the current deep CNN is inefficient on computing resource and computing time. To solve these problems, we proposed a method which combines Gabor kernel, random kernel and pseudoinverse kernel, incorporating with pseudoinverse learning (PIL) algorithm to speed up DNN training processing. With the multiple fixed convolution kernels and pseudoinverse learning algorithm, it is simple and efficient to use the proposed method. The performance of the proposed model is tested on MNIST and CIFAR-10 datasets without using GPU. Experimental results show that our model is better than existing benchmark methods in speed, at the same time it has the comparative recognition accuracy. © 2019, IFIP International Federation for Information processing.

关键词： image recognition

来源：评论

学校读者我要写书评

暂无评论

LRTHR-Net: A Low-Resolution-to-High-Resolution Framework to Iteratively Refine the Segmentation of Thyroid Nodule in Ultrasound images 23rd

LRTHR-Net: A Low-Resolution-to-High-Resolution Framework to ...

引用

Anatomical Brain Barriers to Cancer Spread: Segmentation from CT and MR images Challenge, ABCs 2020, Learn2Reg Challenge, L2R 2020 and Thyroid Nodule Segmentation and Classification in Ultrasound images Challenge, TN-SCUI 2020 held in conjunction with 23rd International Conference on Medical image Computing and Computer-Assisted Intervention, MICCAI 2020

作者： Chen, Huai Song, Shaoli Wang, Xiuying Wang, Renzhen Meng, Deyu Wang, Lisheng Institute of Image Processing and Pattern Recognition Department of Automation Shanghai Jiao Tong University Shanghai200240 China Department of Nuclear Medicine Fudan University Shanghai Cancer Center Shanghai China School of Computer Science The University of Sydney Sydney Australia School of Mathematics and Statistics Xi’an Jiaotong University Xi’an China

ISBN: (纸本)9783030718268

The thyroid nodule is quickly increasing worldwide and the thyroid ultrasound is the key tool for the diagnosis of it. For the subtle difference between malignant and benign nodules, segmenting lesions is the crucial preliminary step for diagnosis. In this paper, we propose a low-resolution-to-high-resolution segmentation framework for TN-SCUI2020 challenge to alleviate the workload of clinicians and improve the efficiency of diagnosis. Specifically speaking, in order to integrate multi-scale information, several low-resolution segmenting results are obtained firstly and combined with a high-resolution image to refine them and obtain high-resolution results. Secondly, iterative-transfer is proposed to effectively initialize network based on previous trained one on small-scale images. Finally, ensemble refinement is introduced to utilize multiple models to refine the segmentation again. Experimental results showed the effectiveness of the proposed framework. And we won the 2nd place in the segmentation task of TN-SCUI2020. © 2021, Springer Nature Switzerland AG.

关键词： Ultrasonics

来源：评论

学校读者我要写书评

暂无评论

Abundant Semantic Information Module for Object Detection

Abundant Semantic Information Module for Object Detection

引用

Algorithms, High Performance Computing and Artificial Intelligence (AHPCAI), International Conference on

作者： Taida Wu Qishen Li Qiufeng Li Jun Gao School of Information Engineering Nanchang Hangkong University Nanchang Jiangxi China Key Laboratory of Jiangxi Province for Image Processing and Pattern Recognition Nanchang Jiangxi China School of Software Nanchang Hangkong University Nanchang Jiangxi China Key Laboratory of Nondestructive Testing (Ministry of Education) Nanchang Hangkong University Nanchang Jiangxi China

Nowadays, with the high-speed iteration of convolution neural network, the efficient object detector emerges one after another. As an important branch of computer vision, object detection aims to detect where and what the object is. However, nowadays, many detector cannot extract abundant semantic information to discriminate the location and size of the objects, resulting in poor performance of the network. In this paper, a new module is proposed, named Abundant Semantic Information Module (ASIM), to enrich and expand the semantic information of the object with more and larger receptive fields. In ASIM, we blend the extracted feature maps to different degrees with different blending factors and fuse them so that all object information is given full attention. Compared to the baseline method, a wealth of experiments show that our module has achieved a significant performance improvement.

关键词： Location awareness Fuses High performance computing Semantics Neural networks Object detection Detectors

来源：评论

学校读者我要写书评

暂无评论

Low-light image enhancement model based on self-calibration illumination and color adjustment

Low-light image enhancement model based on self-calibration ...

引用

Algorithms, High Performance Computing and Artificial Intelligence (AHPCAI), International Conference on

作者： Guiwen Zhang Qishen Li Qiufeng Li Taida Wu School of Information Engineering Nanchang Hangkong University Nanchang Jiangxi China Key Laboratory of Jiangxi Province for Image Processing and Pattern Recognition Nanchang Jiangxi China School of Software Nanchang Hangkong University Nanchang Jiangxi China Key Laboratory of Nondestructive Testing (Ministry of Education) Nanchang Hangkong University Nanchang Jiangxi China

to improve the clarity of objects in a dark-light environment, and to facilitate the identification and detection of targets behind. People perceive the color and brightness of a point not only depending on the pixel value of the point but also the absolute light entering the human eye is related to the color and brightness around the point. The self-calibration model we used considers the surrounding information, which avoids the information pollution, such as large regions of dark background information. In this model, we introduce a heterogeneous convolution filter, which makes reasonable use of different parts of the filter. Through this operation, information from multiple different scale spaces can be fused, and the field of view when applying the convolution layer is greatly increased without increasing the hyper-parameters, thus producing a more distinctive feature representation. Then the self-calibration model is combined with the backbone reinforcement network, which can not only retain the information of the original scale space but also efficiently collect the latent space information to guide the feature transformation in the original space. After the different channels of the image are processed separately, the dependency between the channels can be established by using the heterogeneous convolution filter. Finally, the test on the ExDark data set proves that our dark target enhancement effect has been significantly improved.

关键词： Training Pollution image color analysis Convolution Brightness Lighting Information filters

来源：评论

学校读者我要写书评

暂无评论

Anomaly Handwritten Text Detection for Automatic Descriptive Answer Evaluation 22

Anomaly Handwritten Text Detection for Automatic Descriptive...

引用

Proceedings of the 2022 11th International Conference on Computing and pattern recognition

作者： Nilanjana Chatterjee Palaiahnaakote Shivakumara Umapada Pal Tong Lu Yue Lu Computer Vision and Pattern Recognition Unit Indian Statistical Institute India Faculty of Computer Science and Information Technology University of Malaya Malaysia National Key Lab for Novel Software Technology Nanjing University China Shanghai Key Laboratory of Multidimensional Information Processing East China Normal University China

ISBN: (纸本)9781450397056

Although there are advanced technologies for character recognition, automatic descriptive answer evaluation is an open challenge for the document image analysis community due to large diversified handwritten text and answers to the question. This paper presents a novel method for detecting anomaly handwritten text in the responses written by the students to the questions. The method is proposed based on the fact that when the students are confident in answering questions, the students usually write answers legibly and neatly while they are not confident, they write sloppy writing which may not be easy for the reader to understand. To detect such anomaly handwritten text, we explore a new combination of Fourier transform and deep learning model for detecting edges. This result preserves the structure of handwritten text. For extracting features for classification of anomaly text and normal text, the proposed method studies the behavior of writing style, especially the variation at ascenders and descenders. Therefore, the proposed work draws principal axis which is invariant to rotation, scaling and some extent to distortion for the edge images. With respect to principal axis, the proposed method draws medial axis using uppermost and lowermost points. The distance between the medial axis and principal axis points are considered as feature vector. Further, the feature vector is passed to Artificial Neural Network for classification of anomaly text. The proposed method is evaluated by testing on our own dataset, standard dataset of gender identification (IAM) and handwritten forgery detection dataset (ACPR 2019). The results on different datasets show that the proposed work outperforms the existing methods.

关键词：

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：