Filters








389 Hits in 10.4 sec

Rethinking the Distribution Gap of Person Re-identification with Camera-based Batch Normalization [article]

Zijie Zhuang, Longhui Wei, Lingxi Xie, Tianyu Zhang, Hengheng Zhang, Haozhe Wu, Haizhou Ai, Qi Tian
2020 arXiv   pre-print
With an effective operator named Camera-based Batch Normalization (CBN), we force the image data of all cameras to fall onto the same subspace, so that the distribution gap between any camera pair is largely  ...  The fundamental difficulty in person re-identification (ReID) lies in learning the correspondence among individual cameras.  ...  Acknowledgements This work was supported by National Science Foundation of China under grant No. 61521002.  ... 
arXiv:2001.08680v3 fatcat:cfa3mwbtkbeafamyk52sr3pzm4

Global Distance-distributions Separation for Unsupervised Person Re-identification [article]

Xin Jin, Cuiling Lan, Wenjun Zeng, Zhibo Chen
2020 arXiv   pre-print
Supervised person re-identification (ReID) often has poor scalability and usability in real-world deployments due to domain gaps and the lack of annotations for the target domain data.  ...  Distribution-based hard mining is proposed to further promote the separation of the two distributions. We validate the effectiveness of the GDS constraint in unsupervised ReID networks.  ...  .: Person transfer GAN to bridge domain gap for person re-identification. In: CVPR (2018) 40. Wojke, N., Bewley, A.: Deep cosine metric learning for person re-identification.  ... 
arXiv:2006.00752v3 fatcat:zqkcryiqdfayxg3zn2jqz3yuve

Dual Gaussian-based Variational Subspace Disentanglement for Visible-Infrared Person Re-Identification [article]

Nan Pu, Wei Chen, Yu Liu, Erwin M. Bakker, Michael S. Lew
2020 arXiv   pre-print
Except for the intra-modality variance that RGB-RGB person re-identification mainly overcomes, VI-ReID suffers from additional inter-modality variance caused by the inherent heterogeneous gap.  ...  Visible-infrared person re-identification (VI-ReID) is a challenging and essential task in night-time intelligent surveillance systems.  ...  [40] introduce a challenging RGB-infrared cross-modality person re-identification task, i.e., visible-infrared person re-identification (VI-ReID).  ... 
arXiv:2008.02520v1 fatcat:5l3bwptyy5fd5bztxzfq5xamna

Temporal-Contextual Attention Network for Video-Based Person Re-identification [chapter]

Di Chen, Zheng-Jun Zha, Jiawei Liu, Hongtao Xie, Yongdong Zhang
2018 Lecture Notes in Computer Science  
Then, rethinking person ReID as a zero-shot learning problem, we propose the Mixed High-Order Attention Network (MHN) to further enhance the discrimination and richness of attention knowledge in an explicit  ...  Attention has become more attractive in person reidentification (ReID) as it is capable of biasing the allocation of available resources towards the most informative parts of an input signal.  ...  Acknowledgments: This work was partially supported by the National Natural Science Foundation of China under Grant Nos. 61871052, 61573068, 61471048, and BUPT Excellent Ph.D.  ... 
doi:10.1007/978-3-030-00776-8_14 fatcat:x3b66o57abayxcfkw6oujoboqq

Transferable, Controllable, and Inconspicuous Adversarial Attacks on Person Re-identification With Deep Mis-Ranking [article]

Hongjun Wang, Guangrun Wang, Ya Li, Dongyu Zhang, Liang Lin
2020 arXiv   pre-print
The success of DNNs has driven the extensive applications of person re-identification (ReID) into a new era. However, whether ReID inherits the vulnerability of DNNs remains unexplored.  ...  The code is available at https://github.com/whj363636/Adversarial-attack-on-Person-ReID-With-Deep-Mis-Ranking.  ...  Acknowledgement This work was supported in part by the State Key Development Program (No. 2018YFC0830103), in part by NSFC  ... 
arXiv:2004.04199v1 fatcat:7rjm3pm5cnesfm4a2463d3om4m

Mixed High-Order Attention Network for Person Re-Identification [article]

Binghui Chen, Weihong Deng, Jiani Hu
2019 arXiv   pre-print
Then, rethinking person ReID as a zero-shot learning problem, we propose the Mixed High-Order Attention Network (MHN) to further enhance the discrimination and richness of attention knowledge in an explicit  ...  Attention has become more attractive in person reidentification (ReID) as it is capable of biasing the allocation of available resources towards the most informative parts of an input signal.  ...  Acknowledgments: This work was partially supported by the National Natural Science Foundation of China under Grant Nos. 61871052, 61573068, 61471048, and BUPT Excellent Ph.D.  ... 
arXiv:1908.05819v1 fatcat:hu3cs4jsozcj3b4wuqhwnxzwom

Rethinking of Pedestrian Attribute Recognition: Realistic Datasets with Efficient Method [article]

Jian Jia, Houjing Huang, Wenjie Yang, Xiaotang Chen, Kaiqi Huang
2020 arXiv   pre-print
in train and test set, which is not consistent with practical application.  ...  Experiments on existing and proposed datasets verify the superiority of our method by achieving state-of-the-art performance.  ...  Recently, pedestrian attribute recognition has drawn increasing attention due to its great potential in real world application such as person retrieval [18] , person search [9] and person re-identification  ... 
arXiv:2005.11909v2 fatcat:d5lbpy3htrbo7dbobgncpcpc5u

Rethinking of Pedestrian Attribute Recognition: A Reliable Evaluation under Zero-Shot Pedestrian Identity Setting [article]

Jian Jia, Houjing Huang, Xiaotang Chen, Kaiqi Huang
2021 arXiv   pre-print
Second, based on the proposed definition, we expose the limitations of the existing datasets, which violate the academic norm and are inconsistent with the essential requirement of practical industry application  ...  We review and rethink the recent progress from three perspectives.  ...  In order to improve global feature representation of person re-identification, Lin et al.  ... 
arXiv:2107.03576v2 fatcat:jjabaul7o5fcrcvol7qwvnjj5u

Farewell to Mutual Information: Variational Distillation for Cross-Modal Person Re-Identification [article]

Xudong Tian, Zhizhong Zhang, Shaohui Lin, Yanyun Qu, Yuan Xie, Lizhuang Ma
2021 arXiv   pre-print
To verify our theoretically grounded strategies, we apply our approaches to cross-modal person Re-ID, and conduct extensive experiments, where the superior performance against state-of-the-art methods  ...  Our intriguing findings highlight the need to rethink the way to estimate mutual  ...  This work is supported by the National Natural Science  ... 
arXiv:2104.02862v1 fatcat:dhct4i62l5gbfnopxfty7rtqoy

von Mises-Fisher Loss: An Exploration of Embedding Geometries for Supervised Learning [article]

Tyler R. Scott and Andrew C. Gallagher and Michael C. Mozer
2021 arXiv   pre-print
An interesting property observed for the spherical losses lead us to propose a probabilistic classifier based on the von Mises-Fisher distribution, and we show that it is competitive with state-of-the-art  ...  methods while producing improved out-of-the-box calibration.  ...  domain (e.g., face verification, person drawn from a distribution very similar to that of the train- re-identification, image classification).  ... 
arXiv:2103.15718v4 fatcat:brsxb3so5ngchemwj3pffrvkeq

A Survey of Unsupervised Domain Adaptation for Visual Recognition [article]

Youshan Zhang
2021 arXiv   pre-print
Secondly, we overview the state-of-the-art methods for different categories of UDA from both traditional methods and deep learning based methods.  ...  Unsupervised DA (UDA) deals with a labeled source domain and an unlabeled target domain.  ...  Batch Normalization Batch Normalization (BN) [83] has been widely used in deep 5.2 Adversarial-based methods networks to decrease the covariance shift.  ... 
arXiv:2112.06745v1 fatcat:65ey4xuygrh4fphb5cqwvqi5fq

A Survey of Deep Active Learning [article]

Pengzhen Ren, Yun Xiao, Xiaojun Chang, Po-Yao Huang, Zhihui Li, Brij B. Gupta, Xiaojiang Chen, Xin Wang
2021 arXiv   pre-print
This article is to fill this gap, we provide a formal classification method for the existing work, and a comprehensive and systematic overview.  ...  In this way, DL has aroused strong interest of researchers and has been rapidly developed. Compared with DL, researchers have relatively low interest in AL.  ...  Exploit the Unknown Gradually: One-Shot Video-Based Person Re-Identification by Stepwise Learning.  ... 
arXiv:2009.00236v2 fatcat:zuk2doushzhlfaufcyhoktxj7e

Label-Noise Robust Generative Adversarial Networks [article]

Takuhiro Kaneko, Yoshitaka Ushiku, Tatsuya Harada
2019 arXiv   pre-print
with no reliance on any classifier.  ...  To remedy this, we propose a novel family of GANs called label-noise robust GANs (rGANs), which, by incorporating a noise transition model, can learn a clean label conditional generative distribution even  ...  This work was supported by JSPS KAKENHI Grant Number JP17H06100, partially supported by JST CREST Grant Number JPMJCR1403, Japan, and partially supported by the Ministry of Education, Culture, Sports,  ... 
arXiv:1811.11165v2 fatcat:6ybdr2ar2vap5cvr42qjibbkoq

Exploring Data Aggregation and Transformations to Generalize across Visual Domains [article]

Antono D'Innocente
2021 arXiv   pre-print
With through experimentation, we show how our proposed solutions outperform competitive state-of-the-art approaches in established DG and DA benchmarks.  ...  We also design an algorithm that adapts an object detection model to any out of distribution sample at test time.  ...  We use a batch size of 1, keep batch normalization layers fixed for both pretraining and adaptation phases and freeze the first 2 blocks of ResNet50.  ... 
arXiv:2108.09208v1 fatcat:acushuyacbcydmzqcdtremuf5e

Learning Neural Textual Representations for Citation Recommendation

Binh Thanh Kieu, Inigo Jauregi Unanue, Son Bao Pham, Hieu Xuan Phan, Massimo Piccardi
2021 2020 25th International Conference on Pattern Recognition (ICPR)  
Re-Identification DAY 4 -Jan 15, 2021 Liu, Hong; Miao, Ziling; Yang, Bing; Ding, Runwei 2104 A Base-Derivative Framework for Cross-Modality RGB-Infrared Person Re-Identification DAY 4 -Jan 15, 2021 Tomei  ...  LIME with feature dependency sampling DAY 3 -Jan 14, 2021 Xu, Simin; Luo, Lingkun; Hu, Shiqiang 2527 Attention-Based Model with Attribute Classification for Cross- Domain Person Re-Identification  ... 
doi:10.1109/icpr48806.2021.9412725 fatcat:3vge2tpd2zf7jcv5btcixnaikm
« Previous Showing results 1 — 15 out of 389 results