检索结果-内蒙古大学图书馆

Image Inpainting via Conditional Texture and Structure Dual Generation

学校读者我要写书评

暂无评论

Image Inpainting via Conditional Texture and Structure Dual ...

International Conference on computer Vision (ICCV)

作者： Xiefan Guo Hongyu Yang Di Huang State Key Laboratory of Software Development Environment Beihang University Beijing China School of Computer Science and Engineering Beihang University Beijing China

ISBN: (纸本)9781665428132

Deep generative approaches have recently made considerable progress in image inpainting by introducing structure priors. Due to the lack of proper interaction with image texture during structure reconstruction, however, current solutions are incompetent in handling the cases with large corruptions, and they generally suffer from distorted results. In this paper, we propose a novel two-stream network for image inpainting, which models the structure-constrained texture synthesis and texture-guided structure reconstruction in a coupled manner so that they better leverage each other for more plausible generation. Furthermore, to enhance the global consistency, a Bi-directional Gated Feature Fusion (Bi-GFF) module is designed to exchange and combine the structure and texture information and a Contextual Feature Aggregation (CFA) module is developed to refine the generated contents by region affinity learning and multi-scale feature aggregation. Qualitative and quantitative experiments on the CelebA, Paris StreetView and Places2 datasets demonstrate the superiority of the proposed method. Our code is available at https://***/Xiefan-Guo/CTSDG.

关键词： Image texture computer vision Codes Computational modeling Bidirectional control Logic gates Image reconstruction

A Personalized Preservation Mechanism Satisfying Local Differential Privacy in Location-Based Services 1st

学校读者我要写书评

暂无评论

A Personalized Preservation Mechanism Satisfying Local Diffe...

1st International Conference on Security and Privacy in Digital Economy, SPDE 2020

作者： Wu, Datong Wu, Xiaotong Gao, Jiaquan Ji, Genlin Xu, Xiaolong Qi, Lianyong Dou, Wanchun Changzhou University Changzhou China School of Computer Science and Technology Nanjing Normal University Nanjing China School of Computer and Software Nanjing University of Information Science and Technology Nanjing China School of Information Science and Engineering Qufu Normal University Rizhao China The State Key Lab for Novel Software Technology Nanjing University Nanjing China

ISBN: (纸本)9789811591280

With the wide application of location-based services, there is a huge amount of users’ spatial data generated by mobile devices every day. However, the data is left from mobile users and faced with leakage risk from adversaries or untrusted data receivers. Therefore, spatial data should be perturbed to satisfy local differential privacy (LDP), which is a strong privacy metric in the local setting. In this paper, we study the problem of designing a personalized mechanism satisfying LDP for spatial data. We first construct attack and defense for privacy of spatial data and give a novel privacy definition with LDP and users’ personalized requirements. We propose a personalized location privacy preservation mechanism for spatial data satisfying LDP. We demonstrate the optimal utility and privacy guarantee of our mechanism. We analyze the impact of the key parameters on data utility via the experiments over the real dataset. © 2020, Springer Nature Singapore Pte Ltd.

关键词： Location based services

Improving Deep Assertion Generation via Fine-Tuning Retrieval-Augmented Pre-trained Language Models

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Zhang, Quanjun Fang, Chunrong Zheng, Yi Zhang, Yaxin Zhao, Yuan Huang, Rubing Zhou, Jianyi Yang, Yun Zheng, Tao Chen, Zhenyu Department of Computing Technologies Swinburne University of Technology Melbourne Australia State Key Laboratory for Novel Software Technology Nanjing University Jiangsu Nanjing China School of Computer Science and Engineering Macau University of Science and Technology Jiangsu Nanjing China Huawei Cloud Computing Technologies Co. Ltd. Beijing China Shenzhen Research Institute of Nanjing University Guangdong Shenzhen China

Unit testing validates the correctness of the units of the software system under test and serves as the cornerstone in improving software quality and reliability. To reduce manual efforts in writing unit tests, some techniques have been proposed to generate test assertions automatically, including deep learning (DL)-based, retrieval-based, and integration-based ones. Among them, recent integration-based approaches inherit from both DL-based and retrieval-based approaches and are considered state-of-the-art. Despite being promising, such integration-based approaches suffer from inherent limitations, such as retrieving assertions with lexical matching while ignoring meaningful code semantics, and generating assertions with a limited training corpus. In this paper, we propose a novel Retrieval-Augmented Deep Assertion Generation approach, namely RetriGen, based on a hybrid assertion retriever and a pre-trained language model (PLM)-based assertion generator. Given a focal-test, RetriGen first builds a hybrid assertion retriever to search for the most relevant test-assert pair from external codebases. The retrieval process takes both lexical similarity and semantical similarity into account via a token-based and an embedding-based retriever, respectively. RetriGen then treats assertion generation as a sequence-to-sequence task and designs a PLM-based assertion generator to predict a correct assertion with historical test-assert pairs and the retrieved external assertion. Although our concept is general and can be adapted to various off-the-shelf encoder-decoder PLMs, we implement RetriGen to facilitate assertion generation based on the recent CodeT5 model. We conduct extensive experiments to evaluate RetriGen against six state-of-the-art approaches across two large-scale datasets and two metrics. The experimental results demonstrate that RetriGen achieves 57.66% and 73.24% in terms of accuracy and CodeBLEU, outperforming all baselines with an average improvement of 50.66

关键词： Semantics

Evaluating effects of background stories on graph perception

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Zhao, Ying Shi, Jingcheng Liu, Jiawei Zhao, Jian Zhou, Fangfang Zhang, Wenzhi Chen, Kangyi Zhao, Xin Zhu, Chunyao Chen, Wei The School of Computer Science and Engineering Central South University The School of Computer Science University of Waterloo The State Key Lab of CAD & CG Zhejiang University

A graph is an abstract model that represents relations among entities, for example, the interactions between characters in a novel. A background story endows entities and relations with real-world meanings and describes the semantics and context of the abstract model, for example, the actual story that the novel presents. Considering practical experience and prior research, human viewers who are familiar with the background story of a graph and those who do not know the background story may perceive the same graph differently. However, no previous research has adequately addressed this problem. This research paper thus presents an evaluation that investigated the effects of background stories on graph perception. Three hypotheses that focused on the role of visual focus areas, graph structure identification, and mental model formation on graph perception were formulated and guided three controlled experiments that evaluated the hypotheses using real-world graphs with background stories. An analysis of the resulting experimental data, which compared the performance of participants who read and did not read the background stories, obtained a set of instructive findings. First, having knowledge about a graph's background story influences participants' focus areas during interactive graph explorations. Second, such knowledge significantly affects one's ability to identify community structures but not high degree and bridge structures. Third, this knowledge influences graph recognition under blurred visual conditions. These findings can bring new considerations to the design of storytelling visualizations and interactive graph explorations. Copyright © 2021, The Authors. All rights reserved.

关键词： Semantics

Distilling Virtual Examples for Long-tailed Recognition

学校读者我要写书评

暂无评论

Distilling Virtual Examples for Long-tailed Recognition

International Conference on computer Vision (ICCV)

作者： Yin-Yin He Jianxin Wu Xiu-Shen Wei State Key Laboratory for Novel Software Technology Nanjing University China School of Computer Science and Engineering Nanjing University of Science and Technology China

ISBN: (纸本)9781665428132

We tackle the long-tailed visual recognition problem from the knowledge distillation perspective by proposing a Distill the Virtual Examples (DiVE) method. Specifically, by treating the predictions of a teacher model as virtual examples, we prove that distilling from these virtual examples is equivalent to label distribution learning under certain constraints. We show that when the virtual example distribution becomes flatter than the original input distribution, the under-represented tail classes will receive significant improvements, which is crucial in long-tailed recognition. The proposed DiVE method can explicitly tune the virtual example distribution to become flat. Extensive experiments on three benchmark datasets, including the large-scale iNaturalist ones, justify that the proposed DiVE method can significantly outperform state-of-the-art methods. Further-more, additional analyses and experiments verify the virtual example interpretation, and demonstrate the effectiveness of tailored designs in DiVE for long-tailed problems.

关键词： Visualization computer vision Predictive models Benchmark testing

Deeply Unsupervised Patch Re-Identification for Pre-training Object Detectors

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Ding, Jian Xie, Enze Xu, Hang Jiang, Chenhan Li, Zhenguo Luo, Ping Xia, Gui-Song The State Key Lab LIESMARS School of Computer Scienc Wuhan University Wuhan430079 China The Huawei Noah's Ark Lab The University of Hong Kong Hong Kong The National Engineering Research Center for Multimedia Software School of Computer Science Institute of Artificial Intelligence The State Key Lab LIESMARS Wuhan University Wuhan430072 China

Unsupervised pre-training aims at learning transferable features that are beneficial for downstream tasks. However, most state-of-the-art unsupervised methods concentrate on learning global representations for image-level classification tasks instead of discriminative local region representations, which limits their transferability to region-level downstream tasks, such as object detection. To improve the transferability of pre-trained features to object detection, we present Deeply Unsupervised Patch Re-ID (DUPR), a simple yet effective method for unsupervised visual representation learning. The patch Re-ID task treats individual patch as a pseudo-identity and contrastively learns its correspondence in two views, enabling us to obtain discriminative local features for object detection. Then the proposed patch Re-ID is performed in a deeply unsupervised manner, appealing to object detection, which usually requires multilevel feature maps. Extensive experiments demonstrate that DUPR outperforms state-of-the-art unsupervised pre-trainings and even the ImageNet supervised pre-training on various downstream tasks related to object detection. Copyright © 2021, The Authors. All rights reserved.

关键词： Object recognition

3D U-Net Brain Tumor Segmentation Using VAE Skip Connection 5

学校读者我要写书评

暂无评论

3D U-Net Brain Tumor Segmentation Using VAE Skip Connection

5th IEEE International Conference on Image, Vision and Computing, ICIVC 2020

作者： Li, Ke Kong, Lingwei Zhang, Yifeng School of Information Science and Engineering Southeast University Nanjing China State Key Lab. for Novel Software Technology Nanjing University Nanjing China

ISBN: (纸本)9781728166612

In clinical practice, the determination of the location, shape, and size of brain tumor can greatly assist the diagnosis, monitoring, and treatment of brain tumor. Therefore, accurate and reliable automatic brain tumor segmentation algorithm is of great significance for clinical diagnosis and treatment. With the rapid development of deep learning technology, more and more efficient image segmentation algorithms have also been applied in this field. It has been proven that U-Net model combined with variational auto-encoder can help to effectively regularize the shared encoder and thereby improve the performance. Based on the VAE-U-Net model, this paper proposes a structure called VAE skip connection. By fusing the position information contained in VAE branch into U-Net decoding stage, the network can retain more high-resolution detail information. In addition, we integrate ShakeDrop regularization into the networks to further alleviate the overfitting problem. The experimental results show that the networks after adding VAE skip connection and ShakeDrop can achieve competitive results on the BraTS 2018 dataset. © 2020 IEEE.

关键词： Magnetic resonance imaging

PR-GCN: A Deep Graph Convolutional Network with Point Refinement for 6D Pose Estimation

学校读者我要写书评

暂无评论

arXiv 2021年

作者： Zhou, Guangyuan Wang, Huiqun Chen, Jiaxin Huang, Di State Key Laboratory of Software Development Environment Beihang University Beijing China School of Computer Science and Engineering Beihang University Beijing China

RGB-D based 6D pose estimation has recently achieved remarkable progress, but still suffers from two major limitations: (1) ineffective representation of depth data and (2) insufficient integration of different modalities. This paper proposes a novel deep learning approach, namely Graph Convolutional Network with Point Refinement (PR-GCN), to simultaneously address the issues above in a unified way. It first introduces the Point Refinement Network (PRN) to polish 3D point clouds, recovering missing parts with noise removed. Subsequently, the Multi-Modal Fusion Graph Convolutional Network (MMF-GCN) is presented to strengthen RGB-D combination, which captures geometry-aware inter-modality correlation through local information propagation in the graph convolutional network. Extensive experiments are conducted on three widely used benchmarks, and state-of-the-art performance is reached. Besides, it is also shown that the proposed PRN and MMF-GCN modules are well generalized to other frameworks. © 2021, CC BY-NC-SA.

关键词： Convolution

An Ontological Analysis of Safety-Critical software and Its Anomalies

学校读者我要写书评

暂无评论

An Ontological Analysis of Safety-Critical Software and Its ...

IEEE International Conference on software Quality, Reliability and Security (QRS)

作者： Hezhen Liu Zhi Jin Zheng Zheng Chengqiang Huang Xun Zhang Reliability Technology Lab Huawei Technologies Co. Ltd Shenzhen China Ministry of Education Key Laboratory of High Confidence Software Technologies (Peking University) Beijing China Institute of Software School of Computer Science Peking University Beijing China School of Automation Science and Electrical Engineering Beihang University Beijing China

The progressively dominant role of software in safety-critical systems raise concerns about the software dependability. There are limited mature practices and guides for assessing software dependability and analyzing system-level hazards triggered by software anomalies. A problem is that faults, errors, and failures that represent software anomalies, albeit with different natures, are usually used indistinctly to predict software dependability, leading to unsolid results. The lack of such consensual conceptualization also leads to poor interoperability between supporting tools, and, consequently, difficulties in anomaly management and software maintenance. Anomaly analysis and management is more tough for safety-critical software due to its higher complexity and the safety-critical nature. The complex context of safety-critical software causes difficulties in determining the evolution/propagation path of software anomalies and the impact on system safety. To capture the nature of safety-critical software and support an understanding of mechanisms of software anomalies and associated hazards, we propose three reference ontologies: Safety-critical software Ontology, software Fault Ontology and software-failure-induced Hazard Ontology, which are built based on international standards, guides, and relevant conceptual models. We also discuss the relationships among them. That will facilitate a better understanding of the software anomaly mechanisms and the design of intervening/mitigation solutions. We demonstrate how these ontologies can help analyze software problems of real-world safety-critical systems.

关键词： software maintenance software quality Ontologies software Hazards software reliability Complexity theory