Prototype Discriminative Learning for Face Image Set Classification

@inproceedings{Wang2016PrototypeDL,
  title={Prototype Discriminative Learning for Face Image Set Classification},
  author={Wen Wang and Ruiping Wang and S. Shan and Xilin Chen},
  booktitle={ACCV},
  year={2016}
}
This paper presents a novel Prototype Discriminative Learning (PDL) method to solve the problem of face image set classification. We aim to simultaneously learn a set of prototypes for each image set and a linear discriminative transformation to make projections on the target subspace satisfy that each image set can be optimally classified to the same class with its nearest neighbor prototype. For an image set, its prototypes are actually “virtual” as they do not certainly appear in the set but… 

Prototype Discriminative Learning for Image Set Classification

TLDR
To optimize the prototypes and the projection jointly, a specific gradient descent mechanism is designed by updating the projection on Stiefel manifold and the prototypes in Euclidean space in an alternative optimization manner.

Latent Sparse Discriminative Learning for Face Image Set Classification

TLDR
A new term is proposed to exploit the relations between different gallery sets, which can boost the set discrimination so as to improve performance, and a latent sparse constraint is used to reduce the trade-off parameters and computational cost.

Image Set-Oriented Dual Linear Discriminant Regression Classification and Its Kernel Extension

TLDR
A novel discriminative framework to exploit the superiority of discriminant regression mechanism by learning a projection matrix to force the represented image points from the same class to be close and those from different class are better separated is proposed.

Query set centered sparse projection learning for set based image classification

TLDR
This work can learn a discriminate and sparse projection for image set feature extraction via sparse projection learning by employing ℓ 2,1 norm from the perspective of the query set and an alternative updating algorithm to solve the optimization problem is proposed.

PMAL: Open Set Recognition via Robust Prototype Mining

TLDR
This work proposes a novel Prototype Mining And Learning (PMAL) framework, which has a prototype mining mechanism before the phase of optimizing embedding space, explicitly considering two crucial properties, namely high-quality and diversity of the prototype set.

Incremental few-shot learning via vector quantization in deep embedded space

TLDR
The proposed learning vector quantization in deep embedded space can be customized as a kernel smoother to handle incremental few-shot regression tasks and outperforms other state-of-the-art methods in incremental learning.

References

SHOWING 1-10 OF 40 REFERENCES

Prototype based feature learning for face image set classification

  • Mingbo MaMing ShaoXu ZhaoY. Fu
  • Computer Science
    2013 10th IEEE International Conference and Workshops on Automatic Face and Gesture Recognition (FG)
  • 2013
TLDR
A novel framework to extract image set's features through hyperplane supervised by max-margin criterion between any image set and prototype image set is proposed and shown to be superior to the state-of-the-art methods in terms of both time complexity and recognition accuracy.

Simultaneous Feature and Dictionary Learning for Image Set Based Face Recognition

TLDR
To better exploit the nonlinearity of face samples from different image sets, a deep SFDL (D-SFDL) method is proposed by jointly learning hierarchical non-linear transformations and class-specific dictionaries to further improve the recognition performance.

Face recognition based on image sets

  • Hakan CevikalpB. Triggs
  • Computer Science
    2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition
  • 2010
TLDR
A novel method for face recognition from image sets that combines kernel trick and robust methods to discard input points that are far from the fitted model, thus handling complex and nonlinear manifolds of face images.

Probabilistic nearest neighbor search for robust classification of face image sets

TLDR
The proposed Probabilistic Nearest Neighbor (ProNN) search method is evaluated on three widely-studied public databases, Honda/UCSD, YouTube celebrities and Multiple Biometric Grand Challenge, under two kinds of experimental settings where image sets are contaminated either with false positive faces or images of other subjects.

Image sets alignment for Video-Based Face Recognition

TLDR
This work proposes to bridge the two sets with a reference image set that is well-defined and pre-structured to a number of local models offline that can be computed by comparing only the corresponded local models rather than considering all the pairs.

Matching image sets via adaptive multi convex hull

TLDR
Experiments show that the proposed novel approach to enhance nearest points based methods by integrating affine/convex hull classification with an adapted multi-model approach outperforms single model approaches and other recent techniques, such as Sparse Approximated Nearest Points, Mutual Subspace Method and Manifold Discriminant Analysis.

Manifold-Manifold Distance with application to face recognition based on image set

TLDR
The proposed MMD method outperforms the competing methods on the task of Face Recognition based on Image Set, and a novel manifold learning approach is proposed, which expresses a manifold by a collection of local linear models, each depicted by a subspace.

Sparse approximated nearest points for image set classification

TLDR
This paper introduces a novel between-set distance called Sparse Approximated Nearest Point (SANP) distance, which enforces sparsity on the sample coefficients rather than the model coefficients and jointly optimizes the nearest points as well as their sparse approximations.

Joint prototype and metric learning for set-to-set matching: Application to biometrics

TLDR
This paper proposes a method that jointly learns prototypes and a Mahalanobis distance that represents the gallery image sets using fewer samples, while the classification accuracy is maintained or improved.

Improved Image Set Classification via Joint Sparse Approximated Nearest Subspaces

TLDR
This work proposes to constrain the clustering of each query image set by forcing the clusters to have resemblance to the clusters in the gallery image sets, and shows that the proposed method consistently outperforms several other recent techniques, such as Affine Hull based Image Set Distance, Sparse Approximated Nearest Points (SANP) and Manifold Discriminant Analysis (MDA).