Multi-Task Learning with Low Rank Attribute Embedding for Person Re-Identification

Su, Chi; Yang, Fan; Zhang, Shiliang; Tian, Qi; Davis, Larry S.; Gao, Wen

doi:10.1109/iccv.2015.426

Cited by 159 publications

(84 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because of the semantic nature of attribute‐based representations, in the recent years, they have attracted considerable attention and have been applied to multifarious applications, such as image retrieval (Kovashka et al, ; Zhang, Zha, Yan, Bian, & Chua, ; Wang, Zhang, Tretter, & Lin, ; Choi, ; Chen, Chen, Kuo, & Hsu, ; Siddiquie, Feris, & Davis, ; F. X. Yu, Ji, Tsai, Ye, & Chang, ; Douze, Ramisa, & Schmid, ), image recognition and segmentation (Lu, Yang, Yang, & Rui, ; Shi, Yang, Hospedales, & Xiang, ; Cho, Kang, & Kim, ), clothing recommendation (Liu et al, ), visual object tracking (Danelljan, Khan, Felsberg, & van de Weijer, ), and person re‐identification (C. Su et al, ).…”

Section: Related Workmentioning

confidence: 99%

PSI: A probabilistic semantic interpretable framework for fine‐grained image ranking

et al. 2018

Asso for Info Science & Tech

View full text Add to dashboard Cite

Image Ranking is one of the key problems in information science research area. However, most current methods focus on increasing the performance, leaving the semantic gap problem, which refers to the learned ranking models are hard to be understood, remaining intact. Therefore, in this article, we aim at learning an interpretable ranking model to tackle the semantic gap in fine‐grained image ranking. We propose to combine attribute‐based representation and online passive‐aggressive (PA) learning based ranking models to achieve this goal. Besides, considering the highly localized instances in fine‐grained image ranking, we introduce a supervised constrained clustering method to gather class‐balanced training instances for local PA‐based models, and incorporate the learned local models into a unified probabilistic framework. Extensive experiments on the benchmark demonstrate that the proposed framework outperforms state‐of‐the‐art methods in terms of accuracy and speed.

show abstract

Section: Related Workmentioning

confidence: 99%

PSI: A probabilistic semantic interpretable framework for fine‐grained image ranking

et al. 2018

Asso for Info Science & Tech

View full text Add to dashboard Cite

show abstract

“…However this method does not take advantage of the fact that multiple tasks are learned simultaneously. The correlations between attributes are taken into account by [50], which also combines attribute labels with low-level features for re-identification. Although this approach achieves good performance, it uses handdesigned image features and several independent classification components.…”

Section: Deep Convolutional Neural Networkmentioning

confidence: 99%

Person Reidentification Using Deep Convnets With Multitask Learning

McLaughlin

Rincón

Miller

2017

IEEE Trans. Circuits Syst. Video Technol.

View full text Add to dashboard Cite

Abstract-Person re-identification involves recognizing a person across non-overlapping camera views, with different pose, illumination, and camera characteristics. We propose to tackle this problem by training a deep convolutional network to represent a person's appearance as a low-dimensional feature vector that is invariant to common appearance variations encountered in the re-identification problem. Specifically, a Siamese-network architecture is used to train a feature extraction network using pairs of similar and dissimilar images. We show that use of a novel multi-task learning objective is crucial for regularizing the network parameters in order to prevent over-fitting due to the small size the training dataset. We complement the verification task, which is at the heart of re-identification, by training the network to jointly perform verification, identification, and to recognise attributes related to the clothing and pose of the person in each image. Additionally, we show that our proposed approach performs well even in the challenging cross-dataset scenario, which may better reflect real-world expected performance.

show abstract

“…Umeda et al automatically discovered and learned new attributes that permit successful discrimination through a pair‐wise learning process. Su et al and Zhu et al introduced MTL‐LOREA and MLCNN methods, utilizing both low‐level features and semantic/data‐driven attributes for pedestrian re‐identification. Cheng et al proposed an approach to mine both human understandable and discriminative attributes based on data driving with the assist of naming the candidate attributes by an annotator.…”

Section: Introductionmentioning

confidence: 99%

Data‐driven pedestrian re‐identification based on hierarchical semantic representation

Cheng

Tao

et al. 2017

Concurrency and Computation

View full text Add to dashboard Cite

Summary Limited number of labeled data of surveillance video causes the training of supervised model for pedestrian re‐identification to be a difficult task. Besides, applications of pedestrian re‐identification in pedestrian retrieving and criminal tracking are limited because of the lack of semantic representation. In this paper, a data‐driven pedestrian re‐identification model based on hierarchical semantic representation is proposed, extracting essential features with unsupervised deep learning model and enhancing the semantic representation of features with hierarchical mid‐level ‘attributes’. Firstly, CNNs, well‐trained with the training process of CAEs, is used to extract features of horizontal blocks segmented from unlabeled pedestrian images. Then, these features are input into corresponding attribute classifiers to judge whether the pedestrian has the attributes. Lastly, with a table of ‘attributes‐classes mapping relations’, final result can be calculated. Under the premise of improving the accuracy of attribute classifier, our qualitative results show its clear advantages over the CHUK02, VIPeR, and i‐LIDS data set. Our proposed method is proved to effectively solve the problem of dependency on labeled data and lack of semantic expression, and it also significantly outperforms the state‐of‐the‐art in terms of accuracy and semanteme.

show abstract

Multi-Task Learning with Low Rank Attribute Embedding for Person Re-Identification

Cited by 159 publications

References 44 publications

PSI: A probabilistic semantic interpretable framework for fine‐grained image ranking

PSI: A probabilistic semantic interpretable framework for fine‐grained image ranking

Person Reidentification Using Deep Convnets With Multitask Learning

Data‐driven pedestrian re‐identification based on hierarchical semantic representation

Contact Info

Product

Resources

About