Search

Your search keyword '"Kittler, Josef"' showing total 211 results
Start Over Author "Kittler, Josef" Publisher ieee
211 results on '"Kittler, Josef"'

2. GMML is All You Need

Author

Atito, Sara, primary, Awais, Muhammed, additional, Nandam, Srinivasa, additional, and Kittler, Josef, additional

Published
2023
Full Text
View/download PDF

6. The Ninth Visual Object Tracking VOT2021 Challenge Results

Author

Kristan, Matej, primary, Matas, Jiri, additional, Leonardis, Ales, additional, Felsberg, Michael, additional, Pflugfelder, Roman, additional, Kamarainen, Joni-Kristian, additional, Chang, Hyung Jin, additional, Danelljan, Martin, additional, Zajc, Luka Cehovin, additional, Lukezic, Alan, additional, Drbohlav, Ondrej, additional, Kapyla, Jani, additional, Hager, Gustav, additional, Yan, Song, additional, Yang, Jinyu, additional, Zhang, Zhongqun, additional, Fernandez, Gustavo, additional, Abdelpakey, Mohamed, additional, Bhat, Goutam, additional, Cerkezi, Llukman, additional, Cevikalp, Hakan, additional, Chen, Shengyong, additional, Chen, Xin, additional, Cheng, Miao, additional, Cheng, Ziyi, additional, Chiu, Yu-Chen, additional, Cirakman, Ozgun, additional, Cui, Yutao, additional, Dai, Kenan, additional, Dasari, Mohana Murali, additional, Deng, Qili, additional, Dong, Xingping, additional, Du, Daniel K., additional, Dunnhofer, Matteo, additional, Feng, Zhen-Hua, additional, Feng, Zhiyong, additional, Fu, Zhihong, additional, Ge, Shiming, additional, Gorthi, Rama Krishna, additional, Gu, Yuzhang, additional, Gunsel, Bilge, additional, Guo, Qing, additional, Gurkan, Filiz, additional, Han, Wencheng, additional, Huang, Yanyan, additional, Lawin, Felix Jaremo, additional, Jhang, Shang-Jhih, additional, Ji, Rongrong, additional, Jiang, Cheng, additional, Jiang, Yingjie, additional, Juefei-Xu, Felix, additional, Jun, Yin, additional, Ke, Xiao, additional, Khan, Fahad Shahbaz, additional, Hak Kim, Byeong, additional, Kittler, Josef, additional, Lan, Xiangyuan, additional, Lee, Jun Ha, additional, Leibe, Bastian, additional, Li, Hui, additional, Li, Jianhua, additional, Li, Xianxian, additional, Li, Yuezhou, additional, Liu, Bo, additional, Liu, Chang, additional, Liu, Jingen, additional, Liu, Li, additional, Liu, Qingjie, additional, Lu, Huchuan, additional, Lu, Wei, additional, Luiten, Jonathon, additional, Ma, Jie, additional, Ma, Ziang, additional, Martinel, Niki, additional, Mayer, Christoph, additional, Memarmoghadam, Alireza, additional, Micheloni, Christian, additional, Niu, Yuzhen, additional, Paudel, Danda, additional, Peng, Houwen, additional, Qiu, Shoumeng, additional, Rajiv, Aravindh, additional, Rana, Muhammad, additional, Robinson, Andreas, additional, Saribas, Hasan, additional, Shao, Ling, additional, Shehata, Mohamed, additional, Shen, Furao, additional, Shen, Jianbing, additional, Simonato, Kristian, additional, Song, Xiaoning, additional, Tang, Zhangyong, additional, Timofte, Radu, additional, Torr, Philip, additional, Tsai, Chi-Yi, additional, Uzun, Bedirhan, additional, Van Gool, Luc, additional, Voigtlaender, Paul, additional, Wang, Dong, additional, Wang, Guangting, additional, Wang, Liangliang, additional, Wang, Lijun, additional, Wang, Limin, additional, Wang, Linyuan, additional, Wang, Yong, additional, Wang, Yunhong, additional, Wu, Chenyan, additional, Wu, Gangshan, additional, Wu, Xiao-Jun, additional, Xie, Fei, additional, Xu, Tianyang, additional, Xu, Xiang, additional, Xue, Wanli, additional, Yan, Bin, additional, Yang, Wankou, additional, Yang, Xiaoyun, additional, Ye, Yu, additional, Yin, Jun, additional, Zhang, Chengwei, additional, Zhang, Chunhui, additional, Zhang, Haitao, additional, Zhang, Kaihua, additional, Zhang, Kangkai, additional, Zhang, Xiaohan, additional, Zhang, Xiaolin, additional, Zhang, Xinyu, additional, Zhang, Zhibin, additional, Zhao, Shaochuan, additional, Zhen, Ming, additional, Zhong, Bineng, additional, Zhu, Jiawen, additional, and Zhu, Xue-Feng, additional

Published
2021
Full Text
View/download PDF

8. Two-Stage Supervised Discrete Hashing for Cross-Modal Retrieval.

Author

Zhang, Donglin, Wu, Xiao-Jun, Xu, Tianyang, and Kittler, Josef

Subjects
BINARY codes, MULTIMODAL user interfaces, COMPUTER programming education, INSTRUCTIONAL systems
Abstract

Recently, hashing-based multimodal learning systems have received increasing attention due to their query efficiency and parsimonious storage costs. However, impeded by the quantization loss caused by numerical optimization, the existing cross-media hashing approaches are unable to capture all the discriminative information present in the original multimodal data. Besides, most cross-modal methods belong to the one-step paradigm, which learn the binary codes and hash function simultaneously, increasing the complexity of optimization. To address these issues, we propose a novel two-stage approach, named the two-stage supervised discrete hashing (TSDH) method. In particular, in the first phase, TSDH generates a latent representation for each modality. These representations are then mapped to a common Hamming space to generate the binary codes. In addition, TSDH directly endows the hash codes with the semantic labels, enhancing the discriminatory power of the learned binary codes. A discrete hash optimization approach is developed to learn the binary codes without relaxation, avoiding the large quantization loss. The proposed hash function learning scheme reuses the semantic information contained by the embeddings, endowing the hash functions with enhanced discriminability. Extensive experiments on several databases demonstrate the effectiveness of the developed TSDH, outperforming several recent competitive cross-media algorithms. [ABSTRACT FROM AUTHOR]

Published
2022
Full Text
View/download PDF

9. A Novel Ground Metric for Optimal Transport-Based Chronological Age Estimation.

Author

Akbari, Ali, Awais, Muhammad, Fatemifar, Soroush, Khalid, Syed Safwan, and Kittler, Josef

Abstract

Label distribution learning (LDL) is the state-of-the-art approach to dealing with a number of real-world applications, such as chronological age estimation from a face image, where there is an inherent similarity among adjacent age labels. LDL takes into account the semantic similarity by assigning a label distribution to each instance. The well-known Kullback-Leibler (KL) divergence is the widely used loss function for the LDL framework. However, the KL divergence does not fully and effectively capture the semantic similarity among age labels, thus leading to suboptimal performance. In this article, we propose a novel loss function based on the optimal transport theory for the LDL-based age estimation. A ground metric function plays an important role in the optimal transport formulation. It should be carefully determined based on the underlying geometric structure of the label space of the application in-hand. The label space in the age estimation problem has a specific geometric structure, that is, closer ages have more inherent semantic relationships. Inspired by this, we devise a novel ground metric function, which enables the loss function to increase the influence of highly correlated ages; thus exploiting the semantic similarity among ages more effectively than the existing loss functions. We then use the proposed loss function, namely, $\gamma $ –Wasserstein loss, for training a deep neural network (DNN). This leads to a notoriously computationally expensive and nonconvex optimization problem. Following the standard methodology, we formulate the optimization function as a convex problem and then use an efficient iterative algorithm to update the parameters of the DNN. Extensive experiments in age estimation on different benchmark datasets validate the effectiveness of the proposed method, which consistently outperforms state-of-the-art approaches. [ABSTRACT FROM AUTHOR]

Published
2022
Full Text
View/download PDF

14. Distribution Cognisant Loss for Cross-Database Facial Age Estimation With Sensitivity Analysis.

Author

Akbari, Ali, Awais, Muhammad, Feng, Zhen-Hua, Farooq, Ammarah, and Kittler, Josef

Subjects
SENSITIVITY analysis, DISTRIBUTION (Probability theory), AGE, SOCIAL problems, GENERALIZATION
Abstract

Existing facial age estimation studies have mostly focused on intra-database protocols that assume training and test images are captured under similar conditions. This is rarely valid in practical applications, where we typically encounter training and test sets with different characteristics. In this article, we deal with such situations, namely subjective-exclusive cross-database age estimation. We formulate the age estimation problem as the distribution learning framework, where the age labels are encoded as a probability distribution. To improve the cross-database age estimation performance, we propose a new loss function which provides a more robust measure of the difference between ground-truth and predicted distributions. The desirable properties of the proposed loss function are theoretically analysed and compared with the state-of-the-art approaches. In addition, we compile a new balanced large-scale age estimation database. Last, we introduce a novel evaluation protocol, called subject-exclusive cross-database age estimation protocol, which provides meaningful information of a method in terms of the generalisation capability. The experimental results demonstrate that the proposed approach outperforms the state-of-the-art age estimation methods under both intra-database and subject-exclusive cross-database evaluation protocols. In addition, in this article, we provide a comparative sensitivity analysis of various algorithms to identify trends and issues inherent to their performance. This analysis introduces some open problems to the community which might be considered when designing a robust age estimation system. [ABSTRACT FROM AUTHOR]

Published
2022
Full Text
View/download PDF

15. Graph2Net: Perceptually-Enriched Graph Learning for Skeleton-Based Action Recognition.

Author

Wu, Cong, Wu, Xiao-Jun, and Kittler, Josef

Subjects
INFORMATION modeling, HUMAN behavior, GRAPH algorithms, HUMAN skeleton, SKELETON
Abstract

Skeleton representation has attracted a great deal of attention recently as an extremely robust feature for human action recognition. However, its non-Euclidean structural characteristics raise new challenges for conventional solutions. Recent studies have shown that there is a native superiority in modeling spatiotemporal skeleton information with a Graph Convolutional Network (GCN). Nevertheless, the skeleton graph modeling normally focuses on the physical adjacency of the elements of the human skeleton sequence, which contrasts with the requirement to provide a perceptually meaningful representation. To address this problem, in this paper, we propose a perceptually-enriched graph learning method by introducing innovative features to spatial and temporal skeleton graph modeling. For the spatial information modeling, we incorporate a Local-Global Graph Convolutional Network (LG-GCN) that builds a multifaceted spatial perceptual representation. This helps to overcome the limitations caused by over-reliance on the spatial adjacency relationships in the skeleton. For temporal modeling, we present a Region-Aware Graph Convolutional Network (RA-GCN), which directly embeds the regional relationships conveyed by a skeleton sequence into a temporal graph model. This innovation mitigates the deficiency of the original skeleton graph models. In addition, we strengthened the ability of the proposed channel modeling methods to extract multi-scale representations. These innovations result in a lightweight graph convolutional model, referred to as Graph2Net, that simultaneously extends the spatial and temporal perceptual fields, and thus enhances the capacity of the graph model to represent skeleton sequences. We conduct extensive experiments on NTU-RGB+D 60&120, Northwestern-UCLA, and Kinetics-400 datasets to show that our results surpass the performance of several mainstream methods while limiting the model complexity and computational overhead. [ABSTRACT FROM AUTHOR]

Published
2022
Full Text
View/download PDF

18. Relaxed Block-Diagonal Dictionary Pair Learning With Locality Constraint for Image Recognition.

Author

Chen, Zhe, Wu, Xiao-Jun, and Kittler, Josef

Subjects
IMAGE representation, TEXT recognition, IMAGE recognition (Computer vision), IMAGE reconstruction, FEATURE extraction
Abstract

We propose a novel structured analysis–synthesis dictionary pair learning method for efficient representation and image classification, referred to as relaxed block-diagonal dictionary pair learning with a locality constraint (RBD-DPL). RBD-DPL aims to learn relaxed block-diagonal representations of the input data to enhance the discriminability of both analysis and synthesis dictionaries by dynamically optimizing the block-diagonal components of representation, while the off-block-diagonal counterparts are set to zero. In this way, the learned synthesis subdictionary is allowed to be more flexible in reconstructing the samples from the same class, and the analysis dictionary effectively transforms the original samples into a relaxed coefficient subspace, which is closely associated with the label information. Besides, we incorporate a locality-constraint term as a complement of the relaxation learning to enhance the locality of the analytical encoding so that the learned representation exhibits high intraclass similarity. A linear classifier is trained in the learned relaxed representation space for consistent classification. RBD-DPL is computationally efficient because it avoids both the use of class-specific complementary data matrices to learn discriminative analysis dictionary, as well as the time-consuming $l_{1}/l_{0}$ -norm sparse reconstruction process. The experimental results demonstrate that our RBD-DPL achieves at least comparable or better recognition performance than the state-of-the-art algorithms. Moreover, both the training and testing time are significantly reduced, which verifies the efficiency of our method. The MATLAB code of the proposed RBD-DPL is available at https://github.com/chenzhe207/RBD-DPL. [ABSTRACT FROM AUTHOR]

Published
2022
Full Text
View/download PDF

19. SymNet: A Simple Symmetric Positive Definite Manifold Deep Learning Method for Image Set Classification.

Author

Wang, Rui, Wu, Xiao-Jun, and Kittler, Josef

Subjects
DEEP learning, COVARIANCE matrices, SYMMETRIC matrices, DISCRIMINANT analysis, FEATURE selection, NONLINEAR functions
Abstract

By representing each image set as a nonsingular covariance matrix on the symmetric positive definite (SPD) manifold, visual classification with image sets has attracted much attention. Despite the success made so far, the issue of large within-class variability of representations still remains a key challenge. Recently, several SPD matrix learning methods have been proposed to assuage this problem by directly constructing an embedding mapping from the original SPD manifold to a lower dimensional one. The advantage of this type of approach is that it cannot only implement discriminative feature selection but also preserve the Riemannian geometrical structure of the original data manifold. Inspired by this fact, we propose a simple SPD manifold deep learning network (SymNet) for image set classification in this article. Specifically, we first design SPD matrix mapping layers to map the input SPD matrices into new ones with lower dimensionality. Then, rectifying layers are devised to activate the input matrices for the purpose of forming a valid SPD manifold, chiefly to inject nonlinearity for SPD matrix learning with two nonlinear functions. Afterward, we introduce pooling layers to further compress the input SPD matrices, and the log-map layer is finally exploited to embed the resulting SPD matrices into the tangent space via log-Euclidean Riemannian computing, such that the Euclidean learning applies. For SymNet, the (2-D)2principal component analysis (PCA) technique is utilized to learn the multistage connection weights without requiring complicated computations, thus making it be built and trained easier. On the tail of SymNet, the kernel discriminant analysis (KDA) algorithm is coupled with the output vectorized feature representations to perform discriminative subspace learning. Extensive experiments and comparisons with state-of-the-art methods on six typical visual classification tasks demonstrate the feasibility and validity of the proposed SymNet. [ABSTRACT FROM AUTHOR]

Published
2022
Full Text
View/download PDF

20. VisDrone-SOT2019: The Vision Meets Drone Single Object Tracking Challenge Results

Author

Du, Dawei, primary, Zhu, Pengfei, additional, Wen, Longyin, additional, Bian, Xiao, additional, Ling, Haibin, additional, Hu, Qinghua, additional, Zheng, Jiayu, additional, Peng, Tao, additional, Wang, Xinyao, additional, Zhang, Yue, additional, Bo, Liefeng, additional, Shi, Hailin, additional, Zhu, Rui, additional, Han, Bo, additional, Zhang, Chunhui, additional, Liu, Guizhong, additional, Wu, Han, additional, Wen, Hao, additional, Wang, Haoran, additional, Fan, Jiaqing, additional, Chen, Jie, additional, Gao, Jie, additional, Zhang, Jie, additional, Zhou, Jinghao, additional, Zhou, Jinliu, additional, Wang, Jinwang, additional, Wan, Jiuqing, additional, Kittler, Josef, additional, Zhang, Kaihua, additional, Huang, Kaiqi, additional, Yang, Kang, additional, Zhang, Kangkai, additional, Huang, Lianghua, additional, Zhou, Lijun, additional, Shi, Lingling, additional, Ding, Lu, additional, Wang, Ning, additional, Wang, Peng, additional, Hu, Qintao, additional, Laganiere, Robert, additional, Ma, Ruiyan, additional, Zhang, Ruohan, additional, Zou, Shanrong, additional, Zhao, Shengwei, additional, Li, Shengyang, additional, Zhu, Shengyin, additional, Li, Shikun, additional, Ge, Shiming, additional, Xuan, Shiyu, additional, Xu, Tianyang, additional, He, Ting, additional, Shi, Wei, additional, Song, Wei, additional, Hu, Weiming, additional, Zhang, Wenhua, additional, Zhu, Wenjun, additional, Yu, Xi, additional, Wang, Xianhai, additional, Wu, Xiaojun, additional, Li, Xiaotong, additional, Li, Xiaoxue, additional, Yin, Xiaoyue, additional, Zhang, Xin, additional, Zhao, Xin, additional, Xue, Xizhe, additional, Lei, Xu, additional, Yang, Xueyuan, additional, Gao, Yanjie, additional, Zhao, Yanyun, additional, Xu, Yinda, additional, Li, Ying, additional, Wang, Yong, additional, Yang, Yong, additional, Yang, Yuting, additional, Li, Yuxuan, additional, Wang, Zeyu, additional, Feng, Zhenhua, additional, Zhang, Zhipeng, additional, Yu, Zhiyong, additional, Duan, Zhizhao, additional, and Sun, Zhuojin, additional

Published
2019
Full Text
View/download PDF

25. The Seventh Visual Object Tracking VOT2019 Challenge Results

Author

Kristan, Matej, primary, Matas, Jiri, additional, Leonardis, Ales, additional, Felsberg, Michael, additional, Pflugfelder, Roman, additional, Kamarainen, Joni-Kristian, additional, Cehovin Zajc, Luka, additional, Drbohlav, Ondrej, additional, Lukezic, Alan, additional, Berg, Amanda, additional, Eldesokey, Abdelrahman, additional, Kapyla, Jani, additional, Fernandez, Gustavo, additional, Gonzalez-Garcia, Abel, additional, Memarmoghadam, Alireza, additional, Lu, Andong, additional, He, Anfeng, additional, Varfolomieiev, Anton, additional, Chan, Antoni, additional, Tripathi, Ardhendu Shekhar, additional, Smeulders, Arnold, additional, Pedasingu, Bala Suraj, additional, Chen, Bao Xin, additional, Zhang, Baopeng, additional, Wu, Baoyuan, additional, Li, Bi, additional, He, Bin, additional, Yan, Bin, additional, Bai, Bing, additional, Li, Bing, additional, Li, Bo, additional, Kim, Byeong Hak, additional, Ma, Chao, additional, Fang, Chen, additional, Qian, Chen, additional, Chen, Cheng, additional, Li, Chenglong, additional, Zhang, Chengquan, additional, Tsai, Chi-Yi, additional, Luo, Chong, additional, Micheloni, Christian, additional, Zhang, Chunhui, additional, Tao, Dacheng, additional, Gupta, Deepak, additional, Song, Dejia, additional, Wang, Dong, additional, Gavves, Efstratios, additional, Yi, Eunu, additional, Khan, Fahad Shahbaz, additional, Zhang, Fangyi, additional, Wang, Fei, additional, Zhao, Fei, additional, Ath, George De, additional, Bhat, Goutam, additional, Chen, Guangqi, additional, Wang, Guangting, additional, Li, Guoxuan, additional, Cevikalp, Hakan, additional, Du, Hao, additional, Zhao, Haojie, additional, Saribas, Hasan, additional, Jung, Ho Min, additional, Bai, Hongliang, additional, Yu, Hongyuan, additional, Peng, Houwen, additional, Lu, Huchuan, additional, Li, Hui, additional, Li, Jiakun, additional, Li, Jianhua, additional, Fu, Jianlong, additional, Chen, Jie, additional, Gao, Jie, additional, Zhao, Jie, additional, Tang, Jin, additional, Li, Jing, additional, Wu, Jingjing, additional, Liu, Jingtuo, additional, Wang, Jinqiao, additional, Qi, Jinqing, additional, Zhang, Jinyue, additional, Tsotsos, John K., additional, Lee, Jong Hyuk, additional, Weijer, Joost van de, additional, Kittler, Josef, additional, Lee, Jun Ha, additional, Zhuang, Junfei, additional, Zhang, Kangkai, additional, Wang, Kangkang, additional, Dai, Kenan, additional, Chen, Lei, additional, Liu, Lei, additional, Guo, Leida, additional, Zhang, Li, additional, Wang, Liang, additional, Wang, Liangliang, additional, Zhang, Lichao, additional, Wang, Lijun, additional, Zhou, Lijun, additional, Zheng, Linyu, additional, Rout, Litu, additional, Van Gool, Luc, additional, Bertinetto, Luca, additional, Danelljan, Martin, additional, Dunnhofer, Matteo, additional, Ni, Meng, additional, Kim, Min Young, additional, Tang, Ming, additional, Yang, Ming-Hsuan, additional, Paluru, Naveen, additional, Martinel, Niki, additional, Xu, Pengfei, additional, Zhang, Pengfei, additional, Zheng, Pengkun, additional, Zhang, Pengyu, additional, Torr, Philip H.S., additional, Wang, Qi Zhang Qiang, additional, Guo, Qing, additional, Timofte, Radu, additional, Gorthi, Rama Krishna, additional, Everson, Richard, additional, Han, Ruize, additional, Zhang, Ruohan, additional, You, Shan, additional, Zhao, Shao-Chuan, additional, Zhao, Shengwei, additional, Li, Shihu, additional, Li, Shikun, additional, Ge, Shiming, additional, Bai, Shuai, additional, Guan, Shuosen, additional, Xing, Tengfei, additional, Xu, Tianyang, additional, Yang, Tianyu, additional, Zhang, Ting, additional, Vojir, Tomas, additional, Feng, Wei, additional, Hu, Weiming, additional, Wang, Weizhao, additional, Tang, Wenjie, additional, Zeng, Wenjun, additional, Liu, Wenyu, additional, Chen, Xi, additional, Qiu, Xi, additional, Bai, Xiang, additional, Wu, Xiao-Jun, additional, Yang, Xiaoyun, additional, Chen, Xier, additional, Li, Xin, additional, Sun, Xing, additional, Chen, Xingyu, additional, Tian, Xinmei, additional, Tang, Xu, additional, Zhu, Xue-Feng, additional, Huang, Yan, additional, Chen, Yanan, additional, Lian, Yanchao, additional, Gu, Yang, additional, Liu, Yang, additional, Chen, Yanjie, additional, Zhang, Yi, additional, Xu, Yinda, additional, Wang, Yingming, additional, Li, Yingping, additional, Zhou, Yu, additional, Dong, Yuan, additional, Xu, Yufei, additional, Zhang, Yunhua, additional, Li, Yunkun, additional, Luo, Zeyu Wang Zhao, additional, Zhang, Zhaoliang, additional, Feng, Zhen-Hua, additional, He, Zhenyu, additional, Song, Zhichao, additional, Chen, Zhihao, additional, Zhang, Zhipeng, additional, Wu, Zhirong, additional, Xiong, Zhiwei, additional, Huang, Zhongjian, additional, Teng, Zhu, additional, and Ni, Zihan, additional

Published
2019
Full Text
View/download PDF

29. Visual Semantic Information Pursuit: A Survey.

Author

Liu, Daqi, Bober, Miroslaw, and Kittler, Josef

Subjects
DEEP learning, APPLICATION software
Abstract

Visual semantic information comprises two important parts: the meaning of each visual semantic unit and the coherent visual semantic relation conveyed by these visual semantic units. Essentially, the former one is a visual perception task while the latter corresponds to visual context reasoning. Remarkable advances in visual perception have been achieved due to the success of deep learning. In contrast, visual semantic information pursuit, a visual scene semantic interpretation task combining visual perception and visual context reasoning, is still in its early stage. It is the core task of many different computer vision applications, such as object detection, visual semantic segmentation, visual relationship detection, or scene graph generation. Since it helps to enhance the accuracy and the consistency of the resulting interpretation, visual context reasoning is often incorporated with visual perception in current deep end-to-end visual semantic information pursuit methods. Surprisingly, a comprehensive review for this exciting area is still lacking. In this survey, we present a unified theoretical paradigm for all these methods, followed by an overview of the major developments and the future trends in each potential direction. The common benchmark datasets, the evaluation metrics and the comparisons of the corresponding methods are also introduced. [ABSTRACT FROM AUTHOR]

Published
2021
Full Text
View/download PDF

30. Complementary Discriminative Correlation Filters Based on Collaborative Representation for Visual Object Tracking.

Author

Zhu, Xue-Feng, Wu, Xiao-Jun, Xu, Tianyang, Feng, Zhen-Hua, and Kittler, Josef

Subjects
OBJECT tracking (Computer vision), TRACKING algorithms, ART, FILTERS & filtration, ART objects, REGRESSION analysis
Abstract

In recent years, discriminative correlation filter (DCF) based algorithms have significantly advanced the state of the art in visual object tracking. The key to the success of DCF is an efficient discriminative regression model trained with powerful multi-cue features, including both hand-crafted and deep neural network features. However, the tracking performance is hindered by their inability to respond adequately to abrupt target appearance variations. This issue is posed by the limited representation capability of fixed image features. In this work, we set out to rectify this shortcoming by proposing a complementary representation of a visual content. Specifically, we propose the use of a collaborative representation between successive frames to extract the dynamic appearance information from a target with rapid appearance changes, which results in suppressing the undesirable impact of the background. The resulting collaborative representation coefficients are combined with the original feature maps using a spatially regularised DCF framework for performance boosting. The experimental results on several benchmarking datasets demonstrate the effectiveness and robustness of the proposed method, as compared with a number of state-of-the-art tracking algorithms. [ABSTRACT FROM AUTHOR]

Published
2021
Full Text
View/download PDF

31. Learning Low-Rank and Sparse Discriminative Correlation Filters for Coarse-to-Fine Visual Object Tracking.

Author

Xu, Tianyang, Feng, Zhen-Hua, Wu, Xiao-Jun, and Kittler, Josef

Subjects
OBJECT tracking (Computer vision), FILTERS & filtration, LAGRANGE multiplier, FEATURE extraction, TASK analysis
Abstract

Discriminative correlation filter (DCF) has achieved advanced performance in visual object tracking with remarkable efficiency guaranteed by its implementation in the frequency domain. However, the effect of the structural relationship of DCF and object features has not been adequately explored in the context of the filter design. To remedy this deficiency, this paper proposes a Low-rank and Sparse DCF (LSDCF) that improves the relevance of features used by discriminative filters. To be more specific, we extend the classical DCF paradigm from ridge regression to lasso regression, and constrain the estimate to be of low-rank across frames, thus identifying and retaining the informative filters distributed on a low-dimensional manifold. To this end, specific temporal-spatial-channel configurations are adaptively learned to achieve enhanced discrimination and interpretability. In addition, we analyse the complementary characteristics between hand-crafted features and deep features, and propose a coarse-to-fine heuristic tracking strategy to further improve the performance of our LSDCF. Last, the augmented Lagrange multiplier optimisation method is used to achieve efficient optimisation. The experimental results obtained on a number of well-known benchmarking datasets, including OTB2013, OTB50, OTB100, TC128, UAV123, VOT2016 and VOT2018, demonstrate the effectiveness and robustness of the proposed method, delivering outstanding performance compared to the state-of-the-art trackers. [ABSTRACT FROM AUTHOR]

Published
2020
Full Text
View/download PDF

32. Exploiting Deep Learning in Limited-Fronthaul Cell-Free Massive MIMO Uplink.

Author

Bashar, Manijeh, Akbari, Ali, Cumanan, Kanapathippillai, Ngo, Hien Quoc, Burr, Alister G., Xiao, Pei, Debbah, Merouane, and Kittler, Josef

Subjects
CONVOLUTIONAL neural networks, CENTRAL processing units, DEEP learning, CHANNEL estimation, GEOMETRIC programming
Abstract

A cell-free massive multiple-input multiple-output (MIMO) uplink is considered, where quantize-and-forward (QF) refers to the case where both the channel estimates and the received signals are quantized at the access points (APs) and forwarded to a central processing unit (CPU) whereas in combine-quantize-and-forward (CQF), the APs send the quantized version of the combined signal to the CPU. To solve the non-convex sum rate maximization problem, a heuristic sub-optimal scheme is exploited to convert the power allocation problem into a standard geometric programme (GP). We exploit the knowledge of the channel statistics to design the power elements. Employing large-scale-fading (LSF) with a deep convolutional neural network (DCNN) enables us to determine a mapping from the LSF coefficients and the optimal power through solving the sum rate maximization problem using the quantized channel. Four possible power control schemes are studied, which we refer to as i) small-scale fading (SSF)-based QF; ii) LSF-based CQF; iii) LSF use-and-then-forget (UatF)-based QF; and iv) LSF deep learning (DL)-based QF, according to where channel estimation is performed and exploited and how the optimization problem is solved. Numerical results show that for the same fronthaul rate, the throughput significantly increases thanks to the mapping obtained using DCNN. [ABSTRACT FROM AUTHOR]

Published
2020
Full Text
View/download PDF

33. MDLatLRR: A Novel Decomposition Method for Infrared and Visible Image Fusion.

Author

Li, Hui, Wu, Xiao-Jun, and Kittler, Josef

Subjects
IMAGE fusion, DECOMPOSITION method, INFRARED imaging, IMAGE processing, FEATURE extraction, MATRIX decomposition
Abstract

Image decomposition is crucial for many image processing tasks, as it allows to extract salient features from source images. A good image decomposition method could lead to a better performance, especially in image fusion tasks. We propose a multi-level image decomposition method based on latent low-rank representation(LatLRR), which is called MDLatLRR. This decomposition method is applicable to many image processing fields. In this paper, we focus on the image fusion task. We build a novel image fusion framework based on MDLatLRR which is used to decompose source images into detail parts(salient features) and base parts. A nuclear-norm based fusion strategy is used to fuse the detail parts and the base parts are fused by an averaging strategy. Compared with other state-of-the-art fusion methods, the proposed algorithm exhibits better fusion performance in both subjective and objective evaluation. [ABSTRACT FROM AUTHOR]

Published
2020
Full Text
View/download PDF

44. Robust One-Class Kernel Spectral Regression.

Author

Arashloo, Shervin Rahimzadeh and Kittler, Josef

Subjects
*TIKHONOV regularization, *HILBERT space, *DATA corruption, *LEARNING problems, *PRINCIPAL components analysis
Abstract

The kernel null-space technique is known to be an effective one-class classification (OCC) technique. Nevertheless, the applicability of this method is limited due to its susceptibility to possible training data corruption and the inability to rank training observations according to their conformity with the model. This article addresses these shortcomings by regularizing the solution of the null-space kernel Fisher methodology in the context of its regression-based formulation. In this respect, first, the effect of the Tikhonov regularization in the Hilbert space is analyzed, where the one-class learning problem in the presence of contamination in the training set is posed as a sensitivity analysis problem. Next, the effect of the sparsity of the solution is studied. For both alternative regularization schemes, iterative algorithms are proposed which recursively update label confidences. Through extensive experiments, the proposed methodology is found to enhance robustness against contamination in the training set compared with the baseline kernel null-space method, as well as other existing approaches in the OCC paradigm, while providing the functionality to rank training samples effectively. [ABSTRACT FROM AUTHOR]

Published
2021
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results