检索结果-内蒙古大学图书馆

Recent advances in zero-shot recognition

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Fu, Yanwei Xiang, Tao Jiang, Yu-Gang Xue, Xiangyang Sigal, Leonid Gong, Shaogang School of Data Science Fudan University Shanghai200433 China School of Computer Science Shanghai Key Lab of Intelligent Information Processing Fudan University Department of Computer Science University of British Columbia BC Canada School of Electronic Engineering and Computer Science Queen Mary University of London E1 4NS United Kingdom

With the recent renaissance of deep convolution neural networks, encouraging breakthroughs have been achieved on the supervised recognition tasks, where each class has sufficient training data and fully annotated training data. However, to scale the recognition to a large number of classes with few or now training samples for each class remains an unsolved problem. One approach to scaling up the recognition is to develop models capable of recognizing unseen categories without any training instances, or zero-shot recognition/learning. This article provides a comprehensive review of existing zero-shot recognition techniques covering various aspects ranging from representations of models, and from datasets and evaluation settings. We also overview related recognition tasks including one-shot and open set recognition which can be used as natural extensions of zero-shot recognition when limited number of class samples become available or when zero-shot recognition is implemented in a real-world setting. Importantly, we highlight the limitations of existing approaches and point out future research directions in this existing new research area. Copyright © 2017, The Authors. All rights reserved.

关键词： Machine learning

Learning label-Specific Features and Class-Dependent labels for Multi-label Classification

学校读者我要写书评

暂无评论

IEEE Transactions on Knowledge and Data Engineering 2016年第99期PP卷 3309-3323页

作者： Huang, Jun Li, Guorong Huang, Qingming Wu, Xindong School of Computer and Control Engineering University of Chinese Academy of Sciences Beijing101480 China Key Lab of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences Beijing100190 China School of Computer Science and Information Engineering Hefei University of Technology Hefei230009 China Department of Computer Science University of Vermont 33 Colchester Avenue BurlingtonVT05405 United States

Binary Relevance is a well-known framework for multi-label classification, which considers each class label as a binary classification problem. Many existing multi-label algorithms are constructed within this framework and utilize identical data representation in the discrimination of all the class labels. In multi-label classification, however, each class label might be determined by some specific characteristics of its own. In this paper, we seek to learn label-specific data representation for each class label, which is composed of label-specific features. Our proposed method LLSF can not only be utilized for multi-label classification directly, but also be applied as a feature selection method for multi-label learning and a general strategy to improve multi-label classification algorithms comprising a number of binary classifiers. Inspired by the research works on modeling high-order label correlations, we further extend LLSF to learn class-Dependent labels in a sparse stacking way, denoted as LLSFDL. It incorporates both second-order and high-order label correlations. A comparative study with the state-of-the-art approaches manifests the effectiveness and efficiency of our proposed methods. © 2016 IEEE.

关键词： Classification (of information)

One-shot fine-grained instance retrieval

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Yao, Hantao Zhang, Shiliang Zhang, Yongdong Li, Jintao Tian, Qi Key Lab of Intelligent Information Processing of CAS Institute of Computing Technology CAS Beijing100190 China University of Chinese Academy of Sciences Beijing100049 China School of Electronic Engineering and Computer Science Peking University Beijing100871 China Department of Computer Science University of Texas at San Antonio San Antonio United States

Fine-Grained Visual Categorization (FGVC) has achieved significant progress recently. However, the number of fine-grained species could be huge and dynamically increasing in real scenarios, making it difficult to recognize unseen objects under the current FGVC framework. This raises an open issue to perform large-scale fine-grained identification without a complete training set. Aiming to conquer this issue, we propose a retrieval task named One-Shot Fine-Grained Instance Retrieval (OSFGIR). "One-Shot" denotes the ability of identifying unseen objects through a fine-grained retrieval task assisted with an incomplete auxiliary training set. This paper first presents the detailed description to OSFGIR task and our collected OSFGIR-378K dataset. Next, we propose the Convolutional and Normalization Networks (CN-Nets) learned on the auxiliary dataset to generate a concise and discriminative representation. Finally, we present a coarse-to-fine retrieval framework consisting of three components, i.e., coarse retrieval, fine-grained retrieval, and query expansion, respectively. The framework progressively retrieves images with similar semantics, and performs fine-grained identification. Experiments show our OSFGIR framework achieves significantly better accuracy and efficiency than existing FGVC and image retrieval methods, thus could be a better solution for large-scale fine-grained object identification. Copyright © 2017, The Authors. All rights reserved.

关键词： Semantics

A robust convex formulation for ensemble clustering 25

学校读者我要写书评

暂无评论

A robust convex formulation for ensemble clustering

25th International Joint Conference on Artificial Intelligence, IJCAI 2016

作者： Gao, Junning Yamada, Makoto Kaski, Samuel Mamitsuka, Hiroshi Zhu, Shanfeng School of Computer Science Shanghai Key Lab of Intelligent Information Processing Fudan University Shanghai China Bioinformatics Center Institute for Chemical Research Kyoto University Uji Japan Department of Computer Science Aalto University Finland

We formulate ensemble clustering as a regularization problem over nuclear norm and cluster-wise group norm, and present an efficient optimization algorithm, which we call Robust Convex Ensemble Clustering (RCEC). A key feature of RCEC allows to remove anomalous cluster assignments obtained from component clustering methods by using the group-norm regularization. Moreover, the proposed method is convex and can find the globally optimal solution. We first showed that using synthetic data experiments, RCEC could learn stable cluster assignments from the input matrix including anomalous clusters. We then showed that RCEC outperformed state-of-the-art ensemble clustering methods by using real-world data sets.

关键词： Cluster analysis

GLAD: Global-local-alignment descriptor for pedestrian retrieval

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Wei, Longhui Zhang, Shiliang Yao, Hantao Gao, Wen Tian, Qi School of Electronic Engineering and Computer Science Peking University Beijing100871 China Key Lab of Intelligent Information Processing of CAS Institute of Computing Technology CAS Beijing100190 China University of Chinese Academy of Sciences Beijing100049 China Department of Computer Science University of Texas at San Antonio San AntonioTX78249-1604 United States

The huge variance of human pose and the misalignment of detected human images significantly increase the difficulty of person Re-Identification (Re-ID). Moreover, efficient Re-ID systems are required to cope with the massive visual data being produced by video surveillance systems. Targeting to solve these problems, this work proposes a Global-Local-Alignment Descriptor (GLAD) and an efficient indexing and retrieval framework, respectively. GLAD explicitly leverages the local and global cues in human body to generate a discriminative and robust representation. It consists of part extraction and descriptor learning modules, where several part regions are first detected and then deep neural networks are designed for representation learning on both the local and global regions. A hierarchical indexing and retrieval framework is designed to eliminate the huge redundancy in the gallery set, and accelerate the online Re-ID procedure. Extensive experimental results show GLAD achieves competitive accuracy compared to the state-of-the-art methods. Our retrieval framework significantly accelerates the online Re-ID procedure without loss of accuracy. Therefore, this work has potential to work better on person Re-ID tasks in real scenarios. Copyright © 2017, The Authors. All rights reserved.

关键词： Deep neural networks

Regional Gating neural Networks for Multi-label Image Classification 27

学校读者我要写书评

暂无评论

Regional Gating Neural Networks for Multi-label Image Classi...

27th British Machine Vision Conference, BMVC 2016

作者： Zhao, Rui-Wei Li, Jianguo Chen, Yurong Liu, Jia-Ming Jiang, Yu-Gang Xue, Xiangyang Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University Shanghai China Intel Labs China Beijing China Department of Control Science and Engineering Tongji University Shanghai China

This paper proposes a novel deep learning framework for multi-label image classification, namely regional gating neural networks (RGNN). The motivation is two folds. First, global image features (including CNN based features) ignore the underlying context information among different objects in an image. Consequently, people attempt to use information from objectness regions. However, current objectness region proposal algorithms usually produce several thousand region candidates, including many classification irrelevant or even noisy regions. This leads to the second problem: how to select useful contextual regions for image classification. RGNN is an end-to-end deep learning framework that can automatically select contextual region features with specially designed gate units, which are then fused for classification. Because the gate units and the classifier are integrated in the same deep neural network pipeline, we can learn parameters of the network simultaneously. We evaluate the proposed method on PASCAL VOC 2007/2012 and MS-COCO benchmarks, and results show that RGNN is superior to existing state-of-the-art methods. © 2016. The copyright of this document resides with its authors.

关键词： Image classification

Large-scale person re-identification as retrieval

学校读者我要写书评

暂无评论

Large-scale person re-identification as retrieval

IEEE International Conference on Multimedia and Expo (ICME)

作者： Hantao Yao Shiliang Zhang Dongming Zhang Yongdong Zhang Jintao Li Yu Wang Qi Tian University of Chinese Academy of Sciences Beijing China Key Lab of Intelligent Information Processing of Chinese Academy of Sciences (CAS) Institute of Computing Technology CAS Beijing China School of Electronic Engineering and Computer Science Peking University Beijing China National Computer Network Emergency Response Technical Team/Coordination Center of China Department of Computer Science University of Texas at San Antonio San Antonio USA

ISBN: (纸本)9781509060689

This paper targets to bring together the research efforts on two fields that are growing actively in the past few years: multicamera person Re-Identification (ReID) and large-scale image retrieval. We demonstrate that the essentials of image retrieval and person ReID are the same, i.e., measuring the similarity between images. However, person ReID requires more discriminative and robust features to identify the subtle differences of different persons and overcome the large variance among images of the same person. Specifically, we propose a coarse-to-fine (C2F) framework and a Convolutional neural Network structure named as Conv-Net to tackle the large-scale person ReID as an image retrieval task. Given a query person image, the C2F firstly employ Conv-Net to extract a compact descriptor and perform the coarse-level search. A robust descriptor conveying more spatial cues is hence extracted to perform the fine-level search. Extensive experimental results show that the proposed method outperforms existing methods on two public datasets. Further, the evaluation on a large-scale Person-520K dataset demonstrates that our work is significantly more efficient than existing works, e.g., only needs 180ms to identify a query person from 520K images.

关键词： Training Feature extraction Cameras Robustness Image retrieval Linear programming

P6. Toward optimized nonlinear deformation algorithms for subcortical DBS target regions in human brain MRI

学校读者我要写书评

暂无评论

Clinical Neurophysiology 2018年第8期129卷 e67-e68页

作者： S. Ewert A. Horn F. Finkel N. Li A. Kühn T.M. Herrington Charité – Universitätsmedizin Berlin Department of Neurology Movement Disorder and Neuromodulation Unit Berlin Germany Harvard Medical School Department of Neurology Massachusetts General Hospital Boston MA United States Northeastern University Program in Behavioral Neuroscience Boston MA United States Technische Universität Institute of Software Engineering and Theoretical Computer Science Neural Information Processing Group Berlin Germany

Objective 1. To evaluate and optimize commonly used nonlinear deformation algorithms when applied to human brain MRI. 2. To evaluate and optimize accuracy of automated atlas-based segmentation of the two most commonly used target regions for deep brain stimulation (DBS), the nucleus subthalamicus (STN) and the internal part of the globus pallidus (GPi). 3. To guide clinicians as to which preoperative MRI sequences to acquire in patients undergoing DBS. Background Translating single-subject imaging into a common reference frame such as the MNI space is a ‘core concept within the field of brain mapping’ (Evans et al., 2012) making an accurate transformation into this space necessary. Here, we evaluated and optimized 6 commonly used deformation algorithms and compared each outcome to manually labeled brain regions of 103 brains. Methods Algorithms evaluated were: New Segment, DARTEL and SHOOT (SPM); FNIRT (FSL); SyN and BSpline (ANTs). Target space was the MNI 2009b NLIN space. An atlas of the STN and GPi in MNI space (Ewert et al., 2017) was transformed to native space by inverting each deformation matrix. The resulting atlas-based segmentations were then compared to expert manual segmentations of the native brains. Overlap between the two was quantified using the Dice coefficient (Dice et al., 1945), mean surface distance and correlation of volumes. Two open source datasets were used: The IXI-dataset ( http://***/ixi-dataset/ ) and the HCP-dataset ( http://***/ ). Results The best performing algorithms were New Segment with a custom tissue probability map (TPM) (1) and ANTs SyN (2) with an additional step to refine subcortical coregistration. Performance is compared to inter-rater results (3) of manual segmentations. Fig. 1 shows results for the HCP data for STN (L, in red) and GPi (R, in blue) separately: median Dice for STN (1): 0.70 (std: 0.1); (2): 0.72 (0.06); (3): 0.76 (0.09). Median of mean surface distance in m

关键词：

Exact evaluation of the causal spectrum and localization properties of electronic states on a scale-free network

学校读者我要写书评

暂无评论

arXiv 2017年

作者： Xie, Pinchen Yang, Bingjia Zhang, Zhongzhi Andrade, Roberto F.S. Department of Physics Fudan University Shanghai200433 China Shanghai Key Lab of Intelligent Information Processing Fudan University Shanghai200433 China School of Computer Science Fudan University Shanghai200433 China Instituto de Fisica Universidade Federal da Bahia Salvador40210-210 Brazil

A nearest-neighbor tight-binding model on a tree structure is investigated. The full energy spectrum of the normalized Hamiltonian can be expressed in terms of successively increasing number of contributions at any finite step of construction of the tree, resulting in a causal chain. The degree of quantum localization of any eigenstate, measured by the inverse participation ratio (IPR), is also analytically expressed by means of terms in corresponding eigenvalue chain. The resulting IPR scaling behavior is expressed by the tails of eigenvalue chains as well. Copyright © 2017, The Authors. All rights reserved.

关键词： Eigenvalues and eigenfunctions