Honggang Zhang

Beijing University of Posts and Telecommunications

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Jun Guo

Beijing University of Posts and Telecommunications

Yi-Zhe Song

University of Surrey

Chun-Guang Li

Beijing University of Posts and Telecommunications

Kaili Zhao

Taiyuan University of Science and Technology

Tianming Du

Beijing University of Technology

Yonggang Qi

State Key Laboratory of Polymer Physics and Chemistry

Kaiyue Pang

University of Surrey

Tao Xiang

Zhejiang University

Xiaolin Song

Tianjin University of Technology

Timothy M. Hospedales

University of Edinburgh

Cooperative Institutions

Beijing University of Posts and Telecommunications

110

Chinese Academy of Sciences

Chinese Academy of Medical Sciences & Peking Union Medical College

Shanghai Jiao Tong University

Tsinghua University

Huazhong University of Science and Technology

Dalian University of Technology

University of Chinese Academy of Sciences

Harbin Institute of Technology

Sun Yat-sen University

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Size-self-adaptive recognition method of vehicle manufacturer logos based on feature extraction and SVM classifier

Frontiers of Electrical and Electronic Engineering in China (2009)

Wenting Lu Honggang Zhang Kunyan Lan Jun Guo

Logos Bible Software

Minutiae

Discriminative model

10.1007/s11460-009-0072-4

Cite

Citations (3)

Residual Attention Network for Image Classification

Fei Wang Mengqing Jiang Chen Qian Shuo Yang Cheng Li

In this work, we propose Residual Attention Network, a convolutional neural network using attention mechanism which can incorporate with state-of-art feed forward network architecture in an end-to-end training fashion. Our Residual Attention Network is built by stacking Attention Modules which generate attention-aware features. The attention-aware features from different modules change adaptively as layers going deeper. Inside each Attention Module, bottom-up top-down feedforward structure is used to unfold the feedforward and feedback attention process into a single feedforward process. Importantly, we propose attention residual learning to train very deep Residual Attention Networks which can be easily scaled up to hundreds of layers. Extensive analyses are conducted on CIFAR-10 and CIFAR-100 datasets to verify the effectiveness of every module mentioned above. Our Residual Attention Network achieves state-of-the-art object recognition performance on three benchmark datasets including CIFAR-10 (3.90% error), CIFAR-100 (20.45% error) and ImageNet (4.8% single model and single crop, top-5 error). Note that, our method achieves 0.6% top-1 accuracy improvement with 46% trunk depth and 69% forward FLOPs comparing to ResNet-200. The experiment also demonstrates that our network is robust against noisy labels.

Feed forward

Benchmark (surveying)

FLOPS

10.1109/cvpr.2017.683

Cite

Citations (3,403)

Regularized Instance Embedding for Deep Multi-Instance Learning

Applied Sciences (2019)

Yi Lin Honggang Zhang

In the era of Big Data, multi-instance learning, as a weakly supervised learning framework, has various applications since it is helpful to reduce the cost of the data-labeling process. Due to this weakly supervised setting, learning effective instance representation/embedding is challenging. To address this issue, we propose an instance-embedding regularizer that can boost the performance of both instance- and bag-embedding learning in a unified fashion. Specifically, the crux of the instance-embedding regularizer is to maximize correlation between instance-embedding and underlying instance-label similarities. The embedding-learning framework was implemented using a neural network and optimized in an end-to-end manner using stochastic gradient descent. In experiments, various applications were studied, and the results show that the proposed instance-embedding-regularization method is highly effective, having state-of-the-art performance.

Stochastic Gradient Descent

Regularization

Feature Learning

10.3390/app10010064

Cite

Citations (5)

Robust visual tracking by embedding combination and weighted-gradient optimization

Pattern Recognition (2020)

Jin Feng Peng Xu Shi Pu Kaili Zhao Honggang Zhang

BitTorrent tracker

Feature (linguistics)

Binary classification

Sample (material)

Robustness

10.1016/j.patcog.2020.107339

Cite

Citations (4)

Low-rank and structured sparse subspace clustering

Junjian Zhang Chun-Guang Li Honggang Zhang Jun Guo

High dimensional data often lie approximately in low dimensional subspaces corresponding to multiple classes or categories. Segmenting the high dimensional data into their corresponding low dimensional subspaces is referred as subspace clustering. State of the art methods solve this problem in two steps. First, an affinity matrix is built from data based on self-expressiveness model, in which each data point is expressed as a linear combination of other data points. Second, the segmentation is obtained by spectral clustering. However, solving two dependent steps separately is still suboptimal. In this paper, we propose a joint affinity learning and spectral clustering approach for low-rank representation based subspace clustering, termed Low-Rank and Structured Sparse Subspace Clustering (LRS ³ C), where a subspace structured norm that depends on subspace clustering result is introduced into the objective of low-rank representation problem. We solve it efficiently via a combination of Linearized Alternation Direction Method (LADM) with spectral clustering. Experiments on Hopkins 155 motion segmentation database and Extended Yale B data set demonstrated the effectiveness of our method.

Spectral Clustering

Data point

Clustering high-dimensional data

Rank (graph theory)

10.1109/vcip.2016.7805514

Cite

Citations (15)

Documents clustering based on max-correntropy nonnegative matrix factorization

International Conference on Machine Learning and Cybernetics (2014)

Le Li Jianjun Yang Yang Xu Zhen Qin Honggang Zhang

Nonnegative matrix factorization (NMF) has been successfully applied to many areas of both classification and clustering. Commonly used NMF algorithms mainly target on minimizing the l ₂ distance or the Kullback-Leibler (KL) divergence, which may not be suitable for nonlinear cases. In this paper, we propose a new decomposition method by maximizing the correntropy between the original and the product of two low-rank matrices for document clustering. This method also allows us to learn new basis vectors of the semantic feature space from data. To our knowledge, there is no existing work which clusters high dimensional document data by maximizing the correntropy in NMF. Our experimental results show the supremacy of the proposed method over other variants of NMF algorithms on Reuters21578 and TDT2 databasets.

Non-negative Matrix Factorization

Divergence (linguistics)

Rank (graph theory)

Matrix (chemical analysis)

10.1109/icmlc.2014.7009720

Cite

Citations (18)

Nonlinear estimation of missing ΔLSF parameters by a mixture of Dirichlet distributions

Zhanyu Ma Rainer Martin Jun Guo Honggang Zhang

In packet networks, a reliable scheme to handle packet loss during speech transmission is of great importance. As a common representation of the linear predictive coding (LPC) model, the line spectral frequency (LSF) parameters are widely used in speech quantization and transmission. In this paper, we propose a novel scheme to estimate the missing values occurring during LPC model transmission. In order to exploit the boundary and ordering properties of the LSF parameters, we utilize the ΔLSF representation and apply the Dirichlet mixture model (DMM) to capture the correlations among the elements in the ΔLSF vector. With the conditional distribution of the missing part given the received part, an optimal nonlinear minimum mean square error estimator for the missing values is proposed. Compared to the previously presented Gaussian mixture model based method, the proposed DMM based nonlinear estimator shows a convincing improvement.

10.1109/icassp.2014.6854943

Cite

Citations (4)

A new multi-scale fuzzy model for Histogram-Based Descriptors

Lunshao Chai Zhen Qin Honggang Zhang Jun Guo Bir Bhanu

In this paper, we first propose a general Multi-Scale Fuzzy Model (MSFM) which handles distortions at different scales in Histogram-Based Descriptors (HBDs). This model can be applied both on one-dimensional HBDs and multi-dimensional HBDs. We then focus on applying MSFM on the widely used Shape Context for a Simplified Multi-scale Fuzzy Shape Context (SMFSC) descriptor. Fuzzy models are barely used in multi-dimensional HBDs due to the significant increase of computational complexity. We show that by introducing an intra-bin point location approximation and an approximate iterative fuzzification approach, the algorithm can be simplified and thus SMFSC hardly increases computational complexity. Experiments on standard shape dataset show that SMFSC improves upon the Inner Distance Shape Context. We also applied SMFSC on Content-Based Product Image Retrieval and the experimental results further validate the effectiveness of our model.

10.1109/icmew.2013.6618383

Cite

Citations (0)

Detection of Vehicle Manufacture Logos Using Contextual Information

Lecture notes in computer science (2010)

Wenting Lu Honggang Zhang Kunyan Lan Jun Guo

Logos Bible Software

Discriminative model

10.1007/978-3-642-12304-7_51

Cite

Citations (9)

An Algorithm of Image Restoration Based on Blur Parameter Identification with Cepstrum

Honggang Zhang

At present,the restoration of motion-blurred image mainly depends on accurate identification of blur parameters and good restoration algorithms.In order to identify the blur parameters accurately,an identification method based on cepstrum domain was presented.Histogram equalization and threshold segmentation were employed for the cepstrum to extract the blur information.Then,Radon transformation of the cepstrum was used to identify the blur direction,and mathematical morphology filtering was employed to extract the blur scale.An algorithm of total variation model based on PDE was proposed to restore the motion-blurred image by choosing different λ values to acquire the best restoration effect.Experimental results using artificially blurred images and real images show that the proposed methods are accurate and robust.

Cepstrum

Motion blur

Gaussian blur

Identification

Cite

Citations (1)