Learning Diverse Features with Part-Level Resolution for Person Re-Identification

Xie, Ben; Wu, Xiaofu; Zhang, Suofei; Zhao, Shiliang; Li, Ming

doi:10.48550/arxiv.2001.07442

Cited by 4 publications

(11 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, theses division strategies usually suffer from misalignment between corresponding parts -because of large variations in poses, viewpoints and scales. In order to avoid this misalignment, [38] suggest to concatenate the part-level feature vectors into a single vector and then apply a single loss to the concatenated vector. This strategy is more effective than applying individual loss to each part-level feature vector.…”

Section: Related Workmentioning

confidence: 99%

“…A global average pooling is then applied to each strip to obtain 4 local features vectors. The 4 local vectors are then concatenated yielding to an output vector f l as performed in [38].…”

Section: Local Branchmentioning

confidence: 99%

“…SAM focuses on the most relevant features within the spatial dimension which is essentially based on [7,38]. Indeed, SAM captures and aggregates related features in the spatial domain.…”

Section: Attention Modulesmentioning

confidence: 99%

“…Given a standard CNN backbone, several branches are created to enrich and diversify the features for the purpose of person reidentification. The proposed Deep Miner model is made of three types of branches: (i) The main branch G (in orange) is the original backbone and predicts the standard global features f g ; (ii) Several Input-Erased (IE) branches(in green) that takes as input erased feature maps and predict mined features f e1 and f e2 ; (iii) The local branch (in blue) that outputs local features f l and in which a uniform partition strategy is employed for part level feature resolution as proposed by[38]. In the global branch, and the bottom IE Branch, attention modules are used in order to improve their feature representation.…”

mentioning

confidence: 99%

See 3 more Smart Citations

Deep Miner: A Deep and Multi-branch Network which Mines Rich and Diverse Features for Person Re-identification

Benzine,

Seddik,

Desmarais

2021

Preprint

View full text Add to dashboard Cite

Most recent person re-identification approaches are based on the use of deep convolutional neural networks (CNNs). These networks, although effective in multiple tasks such as classification or object detection, tend to focus on the most discriminative part of an object rather than retrieving all its relevant features. This behavior penalizes the performance of a CNN for the re-identification task, since it should identify diverse and fine grained features. It is then essential to make the network learn a wide variety of finer characteristics in order to make the re-identification process of people effective and robust to finer changes.In this article, we introduce Deep Miner, a method that allows CNNs to "mine" richer and more diverse features about people for their re-identification. Deep Miner is specifically composed of three types of branches: a Global branch (G-branch), a Local branch (L-branch) and an Input-Erased branch (IE-branch). G-branch corresponds to the initial backbone which predicts global characteristics, while L-branch retrieves part level resolution features. The IE-branch for its part, receives partially suppressed feature maps as input thereby allowing the network to "mine" new features (those ignored by G-branch) as output. For this special purpose, a dedicated suppression procedure for identifying and removing features within a given CNN is introduced. This suppression procedure has the major benefit of being simple, while it produces a model that significantly outperforms state-of-the-art (SOTA) re-identification methods. Specifically, we conduct experiments on four standard person re-identification benchmarks and witness an absolute performance gain up to 6.5% mAP compared to SOTA.

show abstract

Section: Related Workmentioning

confidence: 99%

“…A global average pooling is then applied to each strip to obtain 4 local features vectors. The 4 local vectors are then concatenated yielding to an output vector f l as performed in [38].…”

Section: Local Branchmentioning

confidence: 99%

“…SAM focuses on the most relevant features within the spatial dimension which is essentially based on [7,38]. Indeed, SAM captures and aggregates related features in the spatial domain.…”

Section: Attention Modulesmentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

Deep Miner: A Deep and Multi-branch Network which Mines Rich and Diverse Features for Person Re-identification

Benzine,

Seddik,

Desmarais

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Recently, multiple-branch architectures have been proposed in particular [2][3][4][5][6]. These methods allow the network to focus on different person features in individual branches, e.g., on distinct spatial parts or channels.…”

Section: Introductionmentioning

confidence: 99%

Lightweight Multi-Branch Network for Person Re-Identification

Herzog,

Ji,

Teepe

et al. 2021

Preprint

View full text Add to dashboard Cite

Person Re-Identification aims to retrieve person identities from images captured by multiple cameras or the same cameras in different time instances and locations. Because of its importance in many vision applications from surveillance to human-machine interaction, person re-identification methods need to be reliable and fast. While more and more deep architectures are proposed for increasing performance, those methods also increase overall model complexity. This paper proposes a lightweight network that combines global, part-based, and channel features in a unified multi-branch architecture that builds on the resource-efficient OSNet backbone. Using a well-founded combination of training techniques and design choices, our final model achieves state-of-the-art results on CUHK03 labeled, CUHK03 detected, and Market-1501 with 85.1% mAP / 87.2% rank1, 82.4% mAP / 84.9% rank1, and 91.5% mAP / 96.3% rank1, respectively.

show abstract

Person Re‐identification Based on Feature Erasure and Diverse Feature Learning

Jidong

Zhang

et al. 2022

IET Computer Vision

View full text Add to dashboard Cite

Robust and comprehensive modelling of targets is the key to successful person re‐identification. However, some useful information may be ignored, since CNNs tend to learn from the most distinctive feature region of the human body. In the present study, a multi‐branch lightweight network structure that can enhance the ability of diverse feature retrieval is introduced. The proposed network consists of three branches. In the feature erasure branch, a drop block model is added to remove the horizontal region with the highest activation degree from feature vectors so as to allow the network to learn relatively low discrimination features. The global branch is used as an essential supplement to the feature erasure branch. A unified horizontal segmentation strategy is adopted in the local branch to avoid the influence of feature dislocation. Finally, diverse feature learning is achieved through the branch network structure. The proposed method can achieve state‐of‐the‐art results on Market‐1501, CUHK03 and DukeMTMC‐Reid data sets, thereby demonstrating the effectiveness of the method.

show abstract

Learning Diverse Features with Part-Level Resolution for Person Re-Identification

Cited by 4 publications

References 45 publications

Deep Miner: A Deep and Multi-branch Network which Mines Rich and Diverse Features for Person Re-identification

Deep Miner: A Deep and Multi-branch Network which Mines Rich and Diverse Features for Person Re-identification

Lightweight Multi-Branch Network for Person Re-Identification

Person Re‐identification Based on Feature Erasure and Diverse Feature Learning

Contact Info

Product

Resources

About