A Graph Based Unsupervised Feature Aggregation for Face Recognition

Cheng, Yu; Li, Yanfeng; Liu, Qiankun; Yao, Yuan; Pedapudi, Venkata Sai Vijay Kumar; Fan, Xiaotian; Su, Chi; Shen, Shengmei

doi:10.1109/iccvw.2019.00332

Cited by 4 publications

(5 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the other hand, very few works have considered the context information for FR. By utilizing the mutual information between pairs in the testing dataset, [3] proposes an effective graph-based unsupervised feature aggregation method for FR. This method reveals the discriminative power of the local context, however, it is only applicable for offline evaluation.…”

Section: Related Workmentioning

confidence: 99%

“…As far as we know, few approaches consider local context modelling in the field of FR. Yu et al [3] propose a feature aggregation method to utilize the mutual information between pairs in an offline FR testing environment. However, it is impossible to acquire the full information of the test set during online evaluation.…”

Section: Introductionmentioning

confidence: 99%

“…2) Latent. For online evaluation, IDA does not need full access of the whole test sets explicitly compared with [3]. Instead, the context information can be estimated implicitly with the help of SSE.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Inter-class Discrepancy Alignment for Face Recognition

Liu¹,

Wu²,

Wu³

et al. 2021

Preprint

View full text Add to dashboard Cite

The field of face recognition (FR) has witnessed great progress with the surge of deep learning. Existing methods mainly focus on extracting discriminative features, and directly compute the cosine or L2 distance by the point-topoint way without considering the context information. In this study, we make a key observation that the local context represented by the similarities between the instance and its inter-class neighbors 1 plays an important role for FR. Specifically, we attempt to incorporate the local information in the feature space into the metric, and propose a unified framework called Inter-class Discrepancy Alignment (IDA), with two dedicated modules, Discrepancy Alignment Operator (IDA-DAO) and Support Set Estimation (IDA-SSE). IDA-DAO is used to align the similarity scores considering the discrepancy between the images and its neighbors, which is defined by adaptive support sets on the hypersphere. For practical inference, it is difficult to acquire support set during online inference. IDA-SSE can provide convincing inter-class neighbors by introducing virtual candidate images generated with GAN. Furthermore, we propose the learnable IDA-SSE, which can implicitly give estimation without the need of any other images in the evaluation process. The proposed IDA can be incorporated into existing FR systems seamlessly and efficiently. Extensive experiments demonstrate that this framework can 1) significantly improve the accuracy, and 2) make the model robust to the face images of various distributions. Without bells and whistles, our method achieves state-ofthe-art performance on multiple standard FR benchmarks.* means equal contribution 1 In this paper, we refer inter-class neighbors to as support set.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Inter-class Discrepancy Alignment for Face Recognition

Liu¹,

Wu²,

Wu³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…It aims to learn effective representation of nodes by using graph topological structures or other characteristics (e.g., node attributes). Nowadays, graph embedding has become a crucial graph analysis technology in multiple practical applications, including user recommendation system (Silva et al 2010), social network analysis (Orsini, Baracchi, and Frasconi 2018), face recognition (Cheng et al 2019;Jiang et al 2017) and protein function prediction (Borgwardt et al 2007).…”

Section: Introductionmentioning

confidence: 99%

Graph Game Embedding

Hong

Zhang

Cui

et al. 2021

AAAI

View full text Add to dashboard Cite

Graph embedding aims to encode nodes/edges into low-dimensional continuous features, and has become a crucial tool for graph analysis including graph/node classification, link prediction, etc. In this paper we propose a novel graph learning framework, named graph game embedding, to learn discriminative node representation as well as encode graph structures. Inspired by the spirit of game learning, node embedding is converted to the selection/searching process of player strategies, where each node corresponds to one player and each edge corresponds to the interaction of two players. Then, a utility function, which theoretically satisfies the Nash Equilibrium, is defined to measure the benefit/loss of players during graph evolution. Furthermore, a collaboration and competition mechanism is introduced to increase the discriminant learning ability. Under this graph game embedding framework, considering different interaction manners of nodes, we propose two specific models, named paired game embedding for paired nodes and group game embedding for group interaction. Comparing with existing graph embedding methods, our algorithm possesses two advantages: (1) the designed utility function ensures the stable graph evolution with theoretical convergence and Nash Equilibrium satisfaction; (2) the introduced collaboration and competition mechanism endows the graph game embedding framework with discriminative feature leaning ability by guiding each node to learn an optimal strategy distinguished from others. We test the proposed method on three public datasets about citation networks, and the experimental results verify the effectiveness of our method.

show abstract

“…However, this may not be the best solution, since, for each sample, a different combination of the graphs might be a good option. For instance, in the task of person identification, we can have graphs constructed based on fingerprints [22] and face [23] features. While for some people their fingerprint could be more discriminative, for some people their face might be more discriminative for identification.…”

Section: Introductionmentioning

confidence: 99%

A Sample Dependent Decision Fusion Algorithm for Graph-based Semi-supervised Learning

Namjoy

Bosaghzadeh

2020

IJE

View full text Add to dashboard Cite

On many occasions, the evaluation of a phenomenon based on a single feature could not solely be resulted in comprehensive and accurate results. Moreover, even if we have several features, we don't know in advance, which feature offers a better description of the phenomenon. Thus, selecting the best features and especially their combination could lead to better results. An affinity graph is a tool that can describe the relationship between the samples. In this paper, we proposed a graph-based samplebased ranking method that sorts the graphs based on six proposed parameters. The sorting is performed such that the graphs at the top of the list have better performance compared to the graphs at the bottom. Furthermore, we propose a fusion method to merge the information of various features and improve the accuracy of label propagation. Moreover, a method is proposed for parameter optimizations and the ultimate decision fusion. The experimental results indicate that the proposed scheme, apart from correctly ranking the graphs according to their accuracy, in the fusion step, increases the accuracy compared to the use of a single feature.

show abstract

A Graph Based Unsupervised Feature Aggregation for Face Recognition

Cited by 4 publications

References 42 publications

Inter-class Discrepancy Alignment for Face Recognition

Inter-class Discrepancy Alignment for Face Recognition

Graph Game Embedding

A Sample Dependent Decision Fusion Algorithm for Graph-based Semi-supervised Learning

Contact Info

Product

Resources

About