检索结果-内蒙古大学图书馆

Afnet: Local Aggregation and Context Fusion Network for Plant Point Cloud Part Segmentation

学校读者我要写书评

暂无评论

SSRN

SSRN 2024年

作者： Qi, Yanyu Guo, Ruohao Li, Zhenbo Qu, Liao Niu, Dantong College of Information and Electrical Engineering China Agricultural University Beijing100083 China National Innovation Center for Digital Fishery Beijing100083 China School of Intelligence Science and Technology Peking University Beijing100871 China Key Laboratory of Smart Farming Technologies for Aquatic Animal and Livestock Ministry of Agriculture and Rural Affairs Beijing100083 China Beijing Engineering and Technology Research Center for Internet of Things in Agriculture Beijing100083 China Electrical and Computer Engineering department Carnegie Mellon University PittsburghPA15213 United States BAIR Lab University of California BerkeleyCA94709 United States

Convolution-like models have emerged in recent years to address diverse point cloud processing tasks. However, a notable challenge persists in effectively capturing subtle spatial relationships and contextual semantics essential for achieving a thorough comprehension of point clouds. In this work, we propose AFNet, a local aggregation and context fusion network, to capture complicated shape structures, global semantics and contextual representations. Initially, a novel plug-and-play convolution operator is designed to model local points, leveraging both low-level positional relations and high-level feature correlations. This operator employs element-wise multiplication to aggregate transformed positions and features, followed by a reduction function for information abstraction from neighbors. By utilizing the aforementioned convolution as a foundational operator, a deep residual architecture is constructed, wherein each layer adopts one-shot connection strategy. Consequently, global semantics and complicated shapes can be fully obtained in an elegant manner. Furthermore, this work presents PlantSeg3D, a point cloud dataset of 3,600 3D plant models for plant phenotyping, to alleviate the scarcity of plant datasets in the community. Extensive experiments showcase that AFNet performs well on several public benchmarks across different tasks, surpassing previous state-of-the-art approaches with an additional increase of +4.1 mIoU. The impressive performance on both part segmentation and classification tasks demonstrates noteworthy generalization capability. © 2024, The Authors. All rights reserved.

关键词： Convolution

Unsupervised Domain Adaptation on Person Re-Identification via Dual-level Asymmetric Mutual Learning

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Wu, Qiong Li, Jiahan Dai, Pingyang Ye, Qixiang Cao, Liujuan Wu, Yongjian Ji, Rongrong Institute of Artificial Intelligence The Media Analytics and Computing Laboratory Department of Artificial Intelligence School of Informatics Xiamen University Xiamen361005 China School of Information and Control Engineering China University of Mining and Technology Xuzhou221000 China The Media Analytics and Computing Laboratory Department of Artificial Intelligence School of Informatics Xiamen University Xiamen361005 China The Peng Cheng Laboratory Shenzhen518066 China The School of Electronics Electrical and Communication Engineering University of Chinese Academy of Sciences Beijing100049 China The Media Analytics and Computing Lab Department of Computer Science School of Informatics Xiamen University Xiamen361005 China The Youtu Laboratory Tencent Shanghai200233 China The Fujian Engineering Research Center of Trusted Artificial Intelligence Analysis and Application Institute of Artificial Intelligence Xiamen University Xiamen361005 China

Unsupervised domain adaptation person re-identification (Re-ID) aims to identify pedestrian images within an unlabeled target domain with an auxiliary labeled source-domain dataset. Many existing works attempt to recover reliable identity information by considering multiple homogeneous networks. And take these generated labels to train the model in the target domain. However, these homogeneous networks identify people in approximate subspaces and equally exchange their knowledge with others or their mean net to improve their ability, inevitably limiting the scope of available knowledge and putting them into the same mistake. This paper proposes a Dual-level Asymmetric Mutual Learning method (DAML) to learn discriminative representations from a broader knowledge scope with diverse embedding spaces. Specifically, two heterogeneous networks mutually learn knowledge from asymmetric subspaces through the pseudo label generation in a hard distillation manner. The knowledge transfer between two networks is based on an asymmetric mutual learning manner. The teacher network learns to identify both the target and source domain while adapting to the target domain distribution based on the knowledge of the student. Meanwhile, the student network is trained on the target dataset and employs the ground-truth label through the knowledge of the teacher. Extensive experiments in Market-1501, CUHK-SYSU, and MSMT17 public datasets verified the superiority of DAML over state-of-the-arts. Copyright © 2023, The Authors. All rights reserved.

关键词： Distillation

NTIRE 2023 Image Shadow Removal Challenge Report

学校读者我要写书评

暂无评论

NTIRE 2023 Image Shadow Removal Challenge Report

2023 IEEE/CVF Conference on computer Vision and Pattern Recognition Workshops, CVPRW 2023

作者： Vasluianu, Florin-Alexandru Seizinger, Tim Timofte, Radu Cui, Shuhao Huang, Junshi Tian, Shuman Fan, Mingyuan Zhang, Jiaqi Zhu, Li Wei, Xiaoming Wei, Xiaolin Luo, Ziwei Gustafsson, Fredrik K. Zhao, Zheng Sjölund, Jens Schön, Thomas B. Dong, Xiaoyi Zhang, Xi Sheryl Li, Chenghua Leng, Cong Yeo, Woon-Ha Oh, Wang-Taek Lee, Yeo-Reum Ryu, Han-Cheol Luo, Jinting Jiang, Chengzhi Han, Mingyan Wu, Qi Lin, Wenjie Yu, Lei Li, Xinpeng Jiang, Ting Fan, Haoqiang Liu, Shuaicheng Xu, Shuning Song, Binbin Chen, Xiangyu Zhang, Shile Zhou, Jiantao Zhang, Zhao Zhao, Suiyi Zheng, Huan Gao, Yangcheng Wei, Yanyan Wang, Bo Ren, Jiahuan Luo, Yan Kondo, Yuki Miyata, Riku Yasue, Fuma Naruki, Taito Ukita, Norimichi Chang, Hua-En Yang, Hao-Hsiang Chen, Yi-Chung Chiang, Yuan-Chun Huang, Zhi-Kai Chen, Wei-Ting Chen, I-Hsiang Hsieh, Chia-Hsuan Kuo, Sy-Yen Xianwei, Li Fu, Huiyuan Liu, Chunlin Ma, Huadong Fu, Binglan He, Huiming Wang, Mengjia She, Wenxuan Liu, Yu Nathan, Sabari Kansal, Priya Zhang, Zhongjian Yang, Huabin Wang, Yan Zhang, Yanru Phutke, Shruti S. Kulkarni, Ashutosh Khan, Md Raqib Murala, Subrahmanyam Vipparthi, Santosh Kumar Ye, Heng Liu, Zixi Yang, Xingyi Liu, Songhua Wu, Yinwei Jing, Yongcheng Yu, Qianhao Zheng, Naishan Huang, Jie Long, Yuhang Yao, Mingde Zhao, Feng Zhao, Bowen Ye, Nan Shen, Ning Cao, Yanpeng Xiong, Tong Xia, Weiran Li, Dingwen Xia, Shuchen Computer Vision Lab Ifi Caidas University of Würzburg Germany Computer Vision Lab Eth Zürich Switzerland Meituan Group China Department of Information Technology Uppsala University Sweden Institute of Automation Chinese Academy of Sciences Beijing China Nanjing China Maicro Nanjing China Department of Artificial Intelligence Convergence Sahmyook University Seoul Korea Republic of Megvii Technology China University of Electronic Science and Technology of China China University of Macau China China Toyota Technological Institute Japan Graduate Institute of Electronics Engineering National Taiwan University Taiwan Department of Electrical Engineering National Taiwan University Taiwan Graduate Institute of Communication Engineering National Taiwan University Taiwan ServiceNow United States Beijing University of Post and Teleconmunication Beijing China Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education China Couger Inc. Computer Vision and Pattern Recognition Lab Indian Institute of Technology Ropar Punjab Rupnagar India Research Institute Singapore National University of Singapore Singapore Research Institute Singapore University of Sydney Australia Brain-Inspired Vision Laboratory Information Science and Technology Institution University of Science and Technology of China China State Key Laboratory of Fluid Power and Mechatronic Systems School of Mechanical Engineering Zhejiang University Hangzhou310027 China Key Laboratory of Advanced Manufacturing Technology of Zhejiang Province School of Mechanical Engineering Zhejiang University Hangzhou310027 China South China University of Technology China

ISBN: (纸本)9798350302493

This work reviews the results of the NTIRE 2023 Challenge on Image Shadow Removal. The described set of solutions were proposed for a novel dataset, which captures a wide range of object-light interactions. It consists of 1200 roughly pixel aligned pairs of real shadow free and shadow affected images, captured in a controlled environment. The data was captured in a white-box setup, using professional equipment for lights and data acquisition sensors. The challenge had a number of 144 participants registered, out of which 19 teams were compared in the final ranking. The proposed solutions extend the work on shadow removal, improving over the performance level describing state-of-the-art methods. © 2023 IEEE.

关键词： Data acquisition

USED: Universal Speaker Extraction and Diarization

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Ao, Junyi Yıldırım, Mehmet Sinan Tao, Ruijie Ge, Meng Wang, Shuai Qian, Yanmin Li, Haizhou Shenzhen Research Institute of Big Data School of Data Science The Chinese University of Hong Kong Shenzhen518172 China Department of Electrical and Computer Engineering National University of Singapore Singapore119077 Singapore Saw Swee Hock School of Public Health National University of Singapore Singapore117549 Singapore Shenzhen Research Institute of Big Data Shenzhen518172 China Auditory Cognition and Computational Acoustics Lab Department of Computer Science and Engineering China MoE Key Laboratory of Artificial Intelligence AI Institute Shanghai Jiao Tong University Shanghai200240 China

Speaker extraction and diarization are two enabling techniques for real-world speech applications. Speaker extraction aims to extract a target speaker’s voice from a speech mixture, while speaker diarization demarcates speech segments by speaker, annotating ‘who spoke when’. Previous studies have typically treated the two tasks independently. In practical applications, it is more meaningful to have knowledge about ‘who spoke what and when’, which is captured by the two tasks. The two tasks share a similar objective of disentangling speakers. Speaker extraction operates in the frequency domain, whereas diarization is in the temporal domain. It is logical to believe that speaker activities obtained from speaker diarization can benefit speaker extraction, while the extracted speech offers more accurate speaker activity detection than the speech mixture. In this paper, we propose a unified model called Universal Speaker Extraction and Diarization (USED) to address output inconsistency and scenario mismatch issues. It is designed to manage speech mixtures with varying overlap ratios and variable number of speakers. We show that the USED model significantly outperforms the competitive baselines for speaker extraction and diarization tasks on LibriMix and SparseLibriMix datasets. We further validate the diarization performance on CALLHOME, a dataset based on real recordings, and experimental results indicate that our model surpasses recently proposed approaches. Copyright © 2023, The Authors. All rights reserved.

关键词： Frequency domain analysis

Model-Driven Deep Learning for Non-Coherent Massive Machine-Type Communications

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Ma, Zhe Wu, Wen Gao, Feifei Shen, Xuemin The Institute for Artificial Intelligence Tsinghua University State Key Lab of Intelligent Technologies and Systems Beijing National Research Center for Information Science and Technology Department of Automation Tsinghua University Beijing100084 China The Frontier Research Center Peng Cheng Laboratory Guangdong Shenzhen518055 China The Department of Electrical and Computer Engineering University of Waterloo WaterlooONN2L 3G1 Canada

In this paper, we investigate the joint device activity and data detection in massive machine-type communications (mMTC) with a one-phase non-coherent scheme, where data bits are embedded in the pilot sequences and the base station simultaneously detects active devices and their embedded data bits without explicit channel estimation. Due to the correlated sparsity pattern introduced by the noncoherent transmission scheme, the traditional approximate message passing (AMP) algorithm cannot achieve satisfactory performance. Therefore, we propose a deep learning (DL) modified AMP network (DL-mAMPnet) that enhances the detection performance by effectively exploiting the pilot activity correlation. The DL-mAMPnet is constructed by unfolding the AMP algorithm into a feedforward neural network, which combines the principled mathematical model of the AMP algorithm with the powerful learning capability, thereby benefiting from the advantages of both techniques. Trainable parameters are introduced in the DL-mAMPnet to approximate the correlated sparsity pattern and the large-scale fading coefficient. Moreover, a refinement module is designed to further advance the performance by utilizing the spatial feature caused by the correlated sparsity pattern. Simulation results demonstrate that the proposed DL-mAMPnet can significantly outperform traditional algorithms in terms of the symbol error rate performance. © 2023, CC BY-NC-SA.

关键词： Message passing

NTIRE 2023 Challenge on 360° Omnidirectional Image and Video Super-Resolution: Datasets, Methods and Results

学校读者我要写书评

暂无评论

NTIRE 2023 Challenge on 360° Omnidirectional Image and Vide...

2023 IEEE/CVF Conference on computer Vision and Pattern Recognition Workshops, CVPRW 2023

作者： Cao, Mingdeng Mou, Chong Yu, Fanghua Wang, Xintao Zheng, Yinqiang Zhang, Jian Dong, Chao Li, Gen Shan, Ying Timofte, Radu Sun, Xiaopeng Li, Weiqi Sheng, Xuhan Chen, Bin Ma, Haoyu Cheng, Ming Zhao, Shijie Huang, Huaibo Zhou, Xiaoqiang Ai, Yuang He, Ran Wu, Renlong Yang, Yi Zhang, Zhilu Zhang, Shuohao Li, Junyi Chen, Yunjin Ren, Dongwei Zuo, Wangmeng Yang, Hao-Hsiang Chen, Yi-Chung Huang, Zhi-Kai Chen, Wei-Ting Chiang, Yuan-Chun Chang, Hua-En Chen, I.-Hsiang Hsieh, Chia-Hsuan Kuo, Sy-Yen Zhang, Zebin Zhang, Jiaqi Wang, Yuhui Cui, Shuhao Huang, Junshi Zhu, Li Tian, Shuman Yu, Wei Luo, Bingchun Cui, Wanwan Xu, Tianyu Li, Chunyang Bao, Long Sun, Heng Zhang, Zhenyu Wang, Qian The University of Tokyo Japan Arc Lab Tencent Pcg China Peking University China Shenzhen Institute of Advanced Technology Cas China Platform Technologies Tencent Online Video China Computer Vision Lab Ifi & Caidas University of Würzburg Germany ByteDance China Peking University Shenzhen Graduate School China Mais&cripac Institute of Automation Chinese Academy of Sciences China School of Artificial Intelligence University of Chinese Academy of Sciences China University of Science and Technology of China China Beijing Institute of Technology China School of Information Science and Technology ShanghaiTech University China Harbin Institute of Technology China Graduate Institute of Electronics Engineering National Taiwan University Taiwan Department of Electrical Engineering National Taiwan University Taiwan Graduate Institute of Communication Engineering National Taiwan University Taiwan ServiceNow United States ShanghaiTech University China Meituan China Xiaomi Inc

ISBN: (纸本)9798350302493

This report introduces two high-quality datasets Flickr360 and ODV360 for omnidirectional image and video super-resolution, respectively, and reports the NTIRE 2023 challenge on 360° omnidirectional image and video super-resolution. Unlike ordinary 2D images/videos with a narrow field of view, omnidirectional images/videos can represent the whole scene from all directions in one shot. There exists a large gap between omnidirectional image/video and ordinary 2D image/video in both the degradation and restoration processes. The challenge is held to facilitate the development of omnidirectional image/video super-resolution by considering their special characteristics. In this challenge, two tracks are provided: one is the omnidirectional image super-resolution and the other is the omnidirectional video super-resolution. The task of the challenge is to super-resolve an input omnidirectional image/video with a magnification factor of ×4. Realistic omnidirectional downsampling is applied to construct the datasets. Some general degradation(e.g., video compression) is also considered for the video track. The challenge has 100 and 56 registered participants for those two tracks. In the final testing stage, 7 and 3 participating teams submitted their results, source codes, and fact sheets. Almost all teams achieved better performance than baseline models by integrating omnidirectional characteristics, reaching compelling performance on our newly collected Flickr360 and ODV360 datasets. © 2023 IEEE.

关键词： Image compression

LocalViT: Analyzing Locality in Vision Transformers

学校读者我要写书评

暂无评论

LocalViT: Analyzing Locality in Vision Transformers

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

作者： Yawei Li Kai Zhang Jiezhang Cao Radu Timofte Michele Magno Luca Benini Luc Van Goo Computer Vision Lab D-ITET ETH Zurich Switzerland Center for Artificial Intelligence and Data Science (CAIDAS) University of Wurzburg Germany Center for Project-Based Learning D-ITET ETH Zurich Switzerland Integrated Systems Laboratory D-ITET ETH Zurich Switzerland Department of Electrical Electronic and Information Engineering University of Bologna Italy Processing Speech and Images (PSI) KU Leuven Belgium

The aim of this paper is to study the influence of locality mechanisms in vision transformers. Transformers originated from machine translation and are particularly good at modelling long-range dependencies within a long sequence. Although the global interaction between the token embeddings could be well modelled by the self-attention mechanism of transformers, what is lacking is a locality mechanism for infor-mation exchange within a local region. In this paper, locality mechanism is systematically investigated by carefully designed controlled experiments. We add locality to vision transformers into the feed-forward network. This seemingly simple solution is inspired by the comparison between feed-forward networks and inverted residual blocks. The importance of locality mechanisms is validated in two ways: 1) A wide range of design choices (activation function, layer placement, expansion ratio) are available for incorporating locality mechanisms and proper choices can lead to a performance gain over the baseline, and 2) The same locality mechanism is successfully applied to vision transformers with different architecture designs, which shows the generalization of the locality concept. For ImageNet2012 classification, the locality-enhanced transformers outperform the baselines Swin-T [1], DeiT-T [2] and PVT-T [3] by 1.0%, 2.6 % and 3.1 % with a negligible increase in the number of parameters and computational effort. Code is available at https://***/ofsoundof/LocalViT.

关键词：

Retinal Disorder Diagnosis Based on Hybrid Deep Learning Models

学校读者我要写书评

暂无评论

SSRN

SSRN 2022年

作者： Sedik, Ahmed El-Hag, Noha A. El-Hoseny, Heba M. El-Banby, Ghada M. Khalaf, Ashraf A.M. Abd El-Samie, Fathi E. El-Shafai, Walid Department of the Robotics and Intelligent Machines Faculty of Artificial Intelligence Kafrelsheikh University Egypt Menoufia University Egypt Department of Electronics and Electrical Communication Engineering Al-Obour High Institute for Engineering and Technology Egypt Department of Industrial Electronics and Control Engineering Faculty of Electronic Engineering Menoufia University Menouf32952 Egypt Deptartment of Electronics and Electrical Communication Engineering Faculty of Electronic Engineering Menoufia University Menouf32952 Egypt Department of Information Technology College of Computer and Information Sciences Princess Nourah Bint Abdulrahman University Riyadh84428 Saudi Arabia Security Engineering Lab Computer Science Department Prince Sultan University Riyadh11586 Saudi Arabia

Anomaly detection from medical images is badly needed for automated diagnosis. For example, medical images obtainedwith several modalities such as magnetic resonance (MR) and confocal microscopy need to be classified for anomaly *** MR images may contain tumors that need to be detected with anomaly detection techniques. In MR imaging, braintumors are represented as areas composed of abnormal cells that need to be detected. In addition, confocal microscopy can be used for retinal imaging. Hence, there is a need to apply classification techniques for diagnosing diseases of the eye such as Diabetic Retinopathy (DR), which leads to exudates in different regions of the retina. The detection of exudates is necessary for the diagnosis of DR. In this paper, two models are presented for the classification of MR and retinal images. Much research has been done in this area, but it did not achieve high accuracy of classification in an acceptable time. In addition, the previous models did not consider both types of image modalities considered in this paper. The suggested models are Convolutional Neural Network (CNN) and Convolutional Long Short-Term Memory (Conv-LSTM). They are used to classify © 2022, The Authors. All rights reserved.

关键词： Ophthalmology

A robust and interpretable deep learning framework for multi-modal registration via keypoints

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Wang, Alan Q. Yu, Evan M. Dalca, Adrian V. Sabuncu, Mert R. School of Electrical and Computer Engineering Cornell University Cornell Tech New YorkNY10044 United States Department of Radiology Weill Cornell Medical School New YorkNY10065 United States Iterative Scopes CambridgeMA02139 United States Computer Science and Artificial Intelligence Lab The Massachusetts Institute of Technology CambridgeMA02139 United States A.A. Martinos Center for Biomedical Imaging The Massachusetts General Hospital CharlestownMA02129 United States

We present KeyMorph, a deep learning-based image registration framework that relies on automatically detecting corresponding keypoints. State-of-the-art deep learning methods for registration often are not robust to large misalignments, are not interpretable, and do not incorporate the symmetries of the problem. In addition, most models produce only a single prediction at test-time. Our core insight which addresses these shortcomings is that corresponding keypoints between images can be used to obtain the optimal transformation via a differentiable closed-form expression. We use this observation to drive the end-to-end learning of keypoints tailored for the registration task, and without knowledge of ground-truth keypoints. This framework not only leads to substantially more robust registration but also yields better interpretability, since the keypoints reveal which parts of the image are driving the final alignment. Moreover, KeyMorph can be designed to be equivariant under image translations and/or symmetric with respect to the input image ordering. Finally, we show how multiple deformation fields can be computed efficiently and in closed-form at test time corresponding to different transformation variants. We demonstrate the proposed framework in solving 3D affine and spline-based registration of multi-modal brain MRI scans. In particular, we show registration accuracy that surpasses current state-of-the-art methods, especially in the context of large displacements. Our code is available at https://***/alanqrwang/keymorph. © 2023, CC0.

关键词： Image registration