检索结果-内蒙古大学图书馆

arXiv 2020年

作者： Gu, Jinjin Cai, Haoming Chen, Haoyu Ye, Xiaoxing Ren, Jimmy S. Dong, Chao School of Electrical and Information Engineering University of Sydney Australia Chinese University of Hong Kong Shenzhen Hong Kong ShenZhen Key Lab of Computer Vision and Pattern Recognition SIAT-SenseTime Joint Lab Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences SenseTime Research Qing Yuan Research Institute Shanghai Jiao Tong University Shanghai China SIAT Branch Shenzhen Institute of Artificial Intelligence and Robotics for Society

Image quality assessment (IQA) is the key factor for the fast development of image restoration (IR) algorithms. The most recent perceptual IR algorithms based on generative adversarial networks (GANs) have brought in significant improvement on visual performance, but also pose great challenges for quantitative evaluation. Notably, we observe an increasing inconsistency between perceptual quality and the evaluation results. We present two questions: (1) Can existing IQA methods objectively evaluate recent IR algorithms? (2) With the focus on beating current benchmarks, are we getting better IR algorithms? To answer the questions and promote the development of IQA methods, we contribute a large-scale IQA dataset, called Perceptual Image Processing ALgorithms (PIPAL) dataset. Especially, this dataset includes the results of GAN-based IR algorithms, which are missing in previous datasets. We collect more than 1.13 million human judgements to assign subjective scores for PIPAL images using the more reliable "Elo system". Based on PIPAL, we present new benchmarks for both IQA and super-resolution methods. Our results indicate that existing IQA methods cannot fairly evaluate GAN-based IR algorithms. While using appropriate evaluation methods is important, IQA methods should also be updated along with the development of IR algorithms. At last, we shed light on how to improve the IQA performance on GAN-based distortion. Inspired by the find that the existing IQA methods have an unsatisfactory performance on the GAN-based distortion partially because of their low tolerance to spatial misalignment, we propose to improve the performance of an IQA network on GAN-based distortion by explicitly considering this misalignment. We propose the Space Warping Difference Network, which includes the novel l2 pooling layers and Space Warping Difference layers. Experiments demonstrate the effectiveness of the proposed method. © 2020, CC BY.

关键词： Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Haptic display for virtual reality: progress and challenges

引用

Virtual Reality & Intelligent Hardware 2019年第2期1卷 136-162页

作者： Dangxiao WANG Yuan GUO Shiyi LIU Yuru ZHANG Weiliang XU Jing XIAO State Key Laboratory of Virtual Reality Technology and Systems Beihang UniversityBeijing 100083China Beijing Advanced Innovation Center for Biomedical Engineering Beihang UniversityBeijing 100083China Peng Cheng Laboratory Shenzhen 518055China Department of Mechanical Engineering University of AucklandAuckland 1142New Zealand School of Computer Science and the Robotics Engineering Program Worcester Polytechnic InstituteWorcesterMA 01609-2280USA

Immersion, interaction, and imagination are three features of virtual reality (VR). Existing VR systems possess fairly realistic visual and auditory feedbacks, and however, are poor with haptic feedback, by means of which human can perceive the physical world via abundant haptic properties. Haptic display is an interface aiming to enable bilateral signal communications between human and computer, and thus to greatly enhance the immersion and interaction of VR systems. This paper surveys the paradigm shift of haptic display occurred in the past 30 years, which is classified into three stages, including desktop haptics, surface haptics, and wearable haptics. The driving forces, key technologies and typical applications in each stage are critically reviewed. Toward the future high-fidelity VR interaction, research challenges are highlighted concerning handheld haptic device, multimodal haptic device, and high fidelity haptic rendering. In the end, the importance of understanding human haptic perception for designing effective haptic devices is addressed.

关键词： Haptic display Virtual reality Desktop haptics Surface haptics Wearable haptics Multimodal haptics

来源：评论

学校读者我要写书评

暂无评论

F2A2: Flexible Fully-decentralized Approximate Actor-critic for Cooperative Multi-agent Reinforcement Learning

arXiv

引用

arXiv 2020年

作者： Li, Wenhao Jin, Bo Wang, Xiangfeng Yan, Junchi Zha, Hongyuan School of Data Science The Chinese University of Hong Kong Shenzhen China Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen518172 China School of Software Engineering Shanghai Research Institute for Intelligent Autonomous Systems Tongji University Shanghai201804 China School of Computer Science and Technology Key Laboratory of Mathematics and Engineering Applications Ministry of Education East China Normal University Shanghai200062 China Department of Computer Science and Engineering Key Laboratory of Artificial Intelligence Ministry of Education Shanghai Jiao Tong University Shanghai200240 China

Traditional centralized multi-agent reinforcement learning (MARL) algorithms are sometimes unpractical in complicated applications due to non-interactivity between agents, the curse of dimensionality, and computation complexity. Hence, several decentralized MARL algorithms are motivated. However, existing decentralized methods only handle the fully cooperative setting where massive information needs to be transmitted in training. The block coordinate gradient descent scheme they used for successive independent actor and critic steps can simplify the calculation, but it causes serious bias. This paper proposes a flexible fully decentralized actor-critic MARL framework, which can combine most of the actor-critic methods and handle large-scale general cooperative multi-agent settings. A primal-dual hybrid gradient descent type algorithm framework is designed to learn individual agents separately for decentralization. From the perspective of each agent, policy improvement and value evaluation are jointly optimized, which can stabilize multi-agent policy learning. Furthermore, the proposed framework can achieve scalability and stability for the large-scale environment. This framework also reduces information transmission by the parameter sharing mechanism and novel modeling-other-agents methods based on theory-of-mind and online supervised learning. Sufficient experiments in cooperative Multiagent Particle Environment and StarCraft II show that the proposed decentralized MARL instantiation algorithms perform competitively against conventional centralized and decentralized methods. Copyright © 2020, The Authors. All rights reserved.

关键词： Gradient methods

来源：评论

学校读者我要写书评

暂无评论

Pareto Continual Learning: Preference-Conditioned Learning and Adaption for Dynamic Stability-Plasticity Trade-off

arXiv

引用

arXiv 2025年

作者： Lai, Song Zhao, Zhe Zhu, Fei Lin, Xi Zhang, Qingfu Meng, Gaofeng Department of Computer Science City University of Hong Kong Hong Kong Centre for Artificial Intelligence and Robotics HK Institute of Science & Innovation Chinese Academy of Sciences China State Key Laboratory of Multimodal Artificial Intelligence Systems Institute of Automation Chinese Academy of Sciences China School of Artificial Intelligence University of Chinese Academy of Sciences China City University of Hong Kong Shenzhen Research Institute Hong Kong University of Science and Technology of China China

Continual learning aims to learn multiple tasks sequentially. A key challenge in continual learning is balancing between two objectives: retaining knowledge from old tasks (stability) and adapting to new tasks (plasticity). Experience replay methods, which store and replay past data alongside new data, have become a widely adopted approach to mitigate catastrophic forgetting. However, these methods neglect the dynamic nature of the stability-plasticity trade-off and aim to find a fixed and unchanging balance, resulting in suboptimal adaptation during training and inference. In this paper, we propose Pareto Continual Learning (ParetoCL), a novel framework that reformulates the stability-plasticity trade-off in continual learning as a multi-objective optimization (MOO) problem. ParetoCL introduces a preference-conditioned model to efficiently learn a set of Pareto optimal solutions representing different trade-offs and enables dynamic adaptation during inference. From a generalization perspective, ParetoCL can be seen as an objective augmentation approach that learns from different objective combinations of stability and plasticity. Extensive experiments across multiple datasets and settings demonstrate that ParetoCL outperforms state-of-the-art methods and adapts to diverse continual learning scenarios. Copyright © 2025, The Authors. All rights reserved.

关键词： Multi-task learning

来源：评论

学校读者我要写书评

暂无评论

A Framework for the Integration of Coarse Sensing Information and Environmental Constraints*

A Framework for the Integration of Coarse Sensing Informatio...

引用

IEEE International Workshop on Robot and Human Communication (ROMAN)

作者： Rui Li Yingbai Hu Yanjun Cao Mengyao Li School of Automation Chongqing University Chongqing China Informatics 6 Technical University of Munich Munich Germany Department of Computer and Software Engineering École Polytechnique de Montréal Boul Édouard-Montpetit Québec Guangdong Provincial Key Laboratory of Robotics and Intelligent System Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Shenzhen China

ISBN: (数字)9781728160757

ISBN: (纸本)9781728160764

A series of previous work has found that the environmental constraint (EC), which is the natural result of the contact of the robot and the interacting objects, is immensely helpful for the realization of high-precision robotic tasks. However, due to the existence of multifarious errors, such as mechanical error, modeling error and sensing error, there would be discrepancy between the actual constraints and the ideal models. In such case, it is hard to realize manipulation with EC-based strategies. Inspired by human, a preliminary framework which aims at the integration of the coarse sensing information and the environmental constraints is proposed for robotic manipulation. By mapping the sensing information into the new space, where the environmental constraint can be formally described, the region that integrates the sensing information and the environmental constraint is constructed and the conditions to achieve high-precision manipulation are derived. Based on the conditions, a motion planning strategy is proposed to achieve the required task. The effectiveness of this strategy is verified by case studies.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Image Restoration for Terahertz Image Based on Complex-Valued Deconvolution 8

Image Restoration for Terahertz Image Based on Complex-Value...

引用

8th Asia-Pacific Conference on Antennas and Propagation, APCAP 2019

作者： Ning, Wei Qi, Feng Wang, Jinkuan Northeastern University School of Computer Science and Engineering Shenyang110169 China Chinese Academy of Sciences Shenyang Institute of Automation Shenyang110016 China Institutes for Robotics and Intelligent Manufacturing Chinese Academy of Sciences Shenyang110016 China Key Laboratory of Opto-Electronic Information Processing Chinese Academy of Sciences Shenyang110016 China Key Lab of Image Understanding and Computer Vision Liaoning Province Shenyang110016 China

ISBN: (纸本)9781665400541

According to the unique characteristics of terahertz (THz) waves, THz imaging has become a hot topic in widely application areas. However, the imaging resolution is constrained by its long wavelength. Generally, the deconvolution is always used to reconstruct the object function. In this paper, a method that combining the classic deconvolution algorithm with the complex-value data processing is investigated. It is posed that the proposed complex-number deconvolution achieves better performance compared to the real-valued deconvolution. © 2019 IEEE.

关键词： Deconvolution

来源：评论

学校读者我要写书评

暂无评论

Multi-View Partial (MVP) Point Cloud Challenge 2021 on Completion and Registration: Methods and Results

arXiv

引用

arXiv 2021年

作者： Pan, Liang Wu, Tong Cai, Zhongang Liu, Ziwei Yu, Xumin Rao, Yongming Lu, Jiwen Zhou, Jie Xu, Mingye Luo, Xiaoyuan Fu, Kexue Gao, Peng Wang, Manning Wang, Yali Qiao, Yu Zhou, Junsheng Wen, Xin Xiang, Peng Liu, Yu-Shen Han, Zhizhong Yan, Yuanjie An, Junyi Zhu, Lifa Lin, Changwei Liu, Dongrui Li, Xin Gómez-Fernández, Francisco Wang, Qinlong Yang, Yang S-Lab Nanyang Technological University Singapore SenseTime-CUHK Joint Lab The Chinese University of Hong Kong Hong Kong Sensetime Research Shanghai AI Laboratory China Department of Automation Tsinghua University China University of Chinese Academy of Sciences China ShenZhen Key Lab of Computer Vision and Pattern Recognition SIAT-SenseTime Joint Lab Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences China Digital Medical Research Center School of Basic Medical Science Fudan University China School of Software BNRist Tsinghua University China *** Wayne State University State Key Laboratory for Novel Software Technology Nanjing University China DeepGlint Shanghai Jiao Tong University China Sichuan University China Xi'an Jiaotong University China

As real-scanned point clouds are mostly partial due to occlusions and viewpoints, reconstructing complete 3D shapes based on incomplete observations becomes a fundamental problem for computer vision. With a single incomplete point cloud, it becomes the partial point cloud completion problem. Given multiple different observations, 3D reconstruction can be addressed by performing partial-to-partial point cloud registration. Recently, a large-scale Multi-View Partial (MVP) point cloud dataset has been released, which consists of over 100,000 high-quality virtual-scanned partial point clouds. Based on the MVP dataset, this paper reports methods and results in the Multi-View Partial Point Cloud Challenge 2021 on Completion and Registration. In total, 128 participants registered for the competition, and 31 teams made valid submissions. The top-ranked solutions will be analyzed, and then we will discuss future research directions. Copyright © 2021, The Authors. All rights reserved.

关键词： Surface measurement

来源：评论

学校读者我要写书评

暂无评论

Geometry sharing network for 3D point cloud classification and segmentation

arXiv

引用

arXiv 2019年

作者： Xu, Mingye Zhou, Zhipeng Qiao, Yu ShenZhen Key Lab of Computer Vision and Pattern Recognition SIAT-SenseTime Joint Lab Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences University of Chinese Academy of Sciences Siat Branch Shenzhen Institute of Artificial Intelligence and Robotics for Society

In spite of the recent progresses on classifying 3D point cloud with deep CNNs, large geometric transformations like rotation and translation remain challenging problem and harm the final classification performance. To address this challenge, we propose Geometry Sharing Network (GS-Net) which effectively learns point descriptors with holistic context to enhance the robustness to geometric transformations. Compared with previous 3D point CNNs which perform convolution on nearby points, GS-Net can aggregate point features in a more global way. Specially, GS-Net consists of Geometry Similarity Connection (GSC) modules which exploit Eigen-Graph to group distant points with similar and relevant geometric information, and aggregate features from nearest neighbors in both Euclidean space and Eigenvalue space. This design allows GS-Net to efficiently capture both local and holistic geometric features such as symmetry, curvature, convexity and connectivity. Theoretically, we show the nearest neighbors of each point in Eigenvalue space are invariant to rotation and translation. We conduct extensive experiments on public datasets, ModelNet40, ShapeNet Part. Experiments demonstrate that GS-Net achieves the state-of-the-art performances on major datasets, 93.3% on ModelNet40, and are more robust to geometric transformations. Code is released on https://***/MingyeXu/GS-Net. Copyright © 2019, The Authors. All rights reserved.

关键词： Geometry

来源：评论

学校读者我要写书评

暂无评论

Miniature 3D Depth Camera for Real-time Reconstruction

Miniature 3D Depth Camera for Real-time Reconstruction

引用

IEEE International Conference on robotics and Biomimetics

作者： Zilong Zhao Feifei Gu Pengju Xie Huazhao Cao Zhan Song College of Computer Science and Electronic Engineering Hunan University Changsha China Shenzhen Key Laboratory of Minimally Invasive Surgical Robotics and System Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Shenzhen China

This paper developed a miniature 3D vision system for dynamic depth recovery. Our system is constituted by a DOE projector and two cameras. A single pattern with pseudo-random speckles was projected onto the surface of the target object and the coupling pattern was imaged by the stereo cameras. An optimal camera calibration algorithm is adopted to guarantee the measurement accuracy. The speckle pattern projected through a diffractive optical element (DOE) of size 8mm×8mm. The baseline between the stereo cameras is as short as 50mm. The system has the advantages of simple structure, small size, and low cost. It can be easily integrated into mobile devices such as tablet computers and mobile phones. In comparison with several other mature products, the proposed 3D depth camera can achieve 3D reconstruction with higher accuracy.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Boundary and entropy-driven adversarial learning for fundus image segmentation

arXiv

引用

arXiv 2019年

作者： Wang, Shujun Yu, Lequan Li, Kang Yang, Xin Fu, Chi-Wing Heng, Pheng-Ann Department of Computer Science and Engineering Chinese University of Hong Kong Hong Kong Guangdong Provincial Key Laboratory of Computer Vision and Virtual Reality Technology Shenzhen Institutes of Advanced Technology Chinese Academy of Sciences Shenzhen China

Accurate segmentation of the optic disc (OD) and cup (OC) in fundus images from different datasets is critical for glaucoma disease screening. The cross-domain discrepancy (domain shift) hinders the generalization of deep neural networks to work on different domain datasets. In this work, we present an unsupervised domain adaptation framework, called Boundary and Entropy-driven Adversarial Learning (BEAL), to improve the OD and OC segmentation performance, especially on the ambiguous boundary regions. In particular, our proposed BEAL frame-work utilizes the adversarial learning to encourage the boundary prediction and mask probability entropy map (uncertainty map) of the target domain to be similar to the source ones, generating more accurate boundaries and suppressing the high uncertainty predictions of OD and OC segmentation. We evaluate the proposed BEAL framework on two public retinal fundus image datasets (Drishti-GS and RIM-ONE-r3), and the experiment results demonstrate that our method outperforms the state-of-the-art unsupervised domain adaptation methods. Our code is available at https://***/EmmaW8/BEAL. Copyright © 2019, The Authors. All rights reserved.

关键词： Entropy

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：