Locality Preserving Discriminant Projection

Shikkenawis, Gitam; Mitra, Suman K.

doi:10.1109/isba.2015.7126365

Cited by 6 publications

(7 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Locality Preserving Discriminant Projection [12] Locality Preserving Discriminant Projection (LPDP) [12] aims at achieving maximum separability between different classes keeping the local structure of the data intact. Objective function of LPDP is a combination of maximization and minimization problems as follows: (1) here, b i = w T a i is the projection of the data point a i on the learned transformation matrix w. The objective function ensures that if the data points belong to same class, they should be mapped close to each other in the projection domain whereas the data points belonging to different classes should be mapped far apart assuring maximum separability between classes.…”

Section: Kernelization Of Locality Preserving Discriminant Projecmentioning

confidence: 99%

“…Objective function of LPDP is a combination of maximization and minimization problems as follows: (1) here, b i = w T a i is the projection of the data point a i on the learned transformation matrix w. The objective function ensures that if the data points belong to same class, they should be mapped close to each other in the projection domain whereas the data points belonging to different classes should be mapped far apart assuring maximum separability between classes. S and D are the similarity and dissimilarity matrices, details about which can be found in [12]. Weight in the similarity matrix is assigned only if the data points belong to same class, on the other hand, in case of dissimilarity, data points from different classes are considered, thus making the approach supervised.…”

Section: Kernelization Of Locality Preserving Discriminant Projecmentioning

confidence: 99%

“…work on the assumption that the data is linear. Locality Preserving Projection (LPP) [6], Neighborhood Preserving Embedding (NPE) [5], Orthogonal Neighborhood Preserving Projections (ONPP) [9], Locality Preserving Discriminant Projection (LPDP) [12] and some of their variants emphasize on local structure of the data points and aim at preserving the same in the lower dimensional representation as well. LPP formulation ensures that neighboring data points in the original space remain neighbors in learned subspace as well.…”

Section: Introductionmentioning

confidence: 99%

“…Hence, explicit manipulation of distances is required. Locality Preserving Discriminant Projection (LPDP) [12], works on this principle i.e. neighboring data points should remain closer whereas the data points belonging to different classes should be projected distant from each other.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Kernelization of locality preserving discriminant projection

Shikkenawis

Mitra

2015

2015 Fifth National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics (NCVPRIPG)

Self Cite

View full text Add to dashboard Cite

In recent years, non-linear dimensionality reduction approaches have become popular due to their suitability to capture the non-linearities present in the data. However, these are not applicable for recognition related applications as representation of new data points is not defined in the reduced subspace. Hence, to explore the non-linearities of the data, local structure preserving approaches received considerable attention. These approaches keep the local information of the data points intact in the lower dimensional space as well. Locality Preserving Discriminant Projection (LPDP) not only preserves the neighborhood information for each data point, but also tries to discriminate data points from different classes using their class labels. The performance of such neighborhood information preserving dimensionality reduction techniques do not guarantee to capture complex non-linearities present in the data. To address this issue, kernel functions that map the data in a nonlinear feature space before applying the dimensionality reduction approach are widely used. This article discusses Kernelization of LPDP which explores complex non-linear changes of face images due to facial expression, illumination, pose and environmental changes. The proposed approach is applied for face recognition on some of the benchmark databases with such variability.

show abstract

Section: Kernelization Of Locality Preserving Discriminant Projecmentioning

confidence: 99%

Section: Kernelization Of Locality Preserving Discriminant Projecmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Kernelization of locality preserving discriminant projection

Shikkenawis

Mitra

2015

2015 Fifth National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics (NCVPRIPG)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Discriminant information also plays vital role in forming distinct clusters [25–27], which is completely ignored in both the cases. Locality preserving discriminant projection (LPDP) [28] takes into account both similarity and dissimilarity between the data points while learning the basis, which results in enhanced grouping of data. Analogues to PCA and OLPP, this approach also processes data in vector format and the attained basis are also not orthogonal.…”

Section: Introductionmentioning

confidence: 99%