Why can LDA be performed in PCA transformed space?

By Jian Yang, Jing-yu Yang. Pattern Recognition, 2003

616 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/5c87b9a1fec66f629da3c9b7f1e1985b5bb765a7

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Lei Zhang
2. Meng Yang
3. David Dapeng Zhang
4. Weihong Deng
5. Jian Yang
6. Luc Van Gool
7. Shuicheng Yan
8. Wangmeng Zuo

In Media Technology

1. Jon Atli Benediktsson
2. Jocelyn Chanussot
3. Xudong Kang
4. Pedram Ghamisi
5. Danfeng Hong
6. Lei Zhang
7. Shutao Li
8. David Dapeng Zhang

Citations in the corpus

• Feature Extraction for Hyperspectral Imagery: The Evolution From Shallow to Deep: Overview and Toolbox (2020, methodology): “Please refer to [62]–[64] that might be useful for those who are interested in more types of W. Similar to SLDA that first projects the original data into a subspace and then LDA is performed in the transformed subspace, FSDA starts with maximizing the between-spectral scatter matrix (Sf ) to enhanc”
• A General Exponential Framework for Dimensionality Reduction (2014, methodology): “To tackle the SSS problem, many variants of LDA have been proposed in the recent years, such as Fisherface [3], null LDA [6], LDA/QR [7], LDA/GSVD [8], LDA/FKT [9], DLA [10][11], Direct LDA and its variants [12][13][6][14].”
• KPCA plus LDA: a complete kernel Fisher discriminant framework for feature extraction and recognition (2005, background): “…J È b ð'Þ ¼ ' T S È b '; ðjj'jj ¼ 1Þ: ð12Þ As a special case of the Fisher criterion, the criterion given in (12) is very intuitive since it is reasonable to use the betweenclass scatter to measure the discriminatory ability of a projection axis when the within-class scatter is zero [22], [24].”
• Globally Maximizing, Locally Minimizing: Unsupervised Discriminant Projection with Applications to Face and Palm Biometrics (2007, neutral): “Ç”
• Comments on "Globally Maximizing, Locally Minimizing: Unsupervised Discriminant Projection with Application to Face and Palm Biometrics" (2007, methodology): “PCA and LDA are two well-known linear subspace learning methods which have been extensively used in pattern recognition and computer vision areas and have become the most popular techniques for face recognition and other biometrics [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14]”
• Generalizing discriminant analysis using the generalized singular value decomposition (2004, methodology): “Index Terms—Linear discriminant analysis, latent semantic indexing, principal component analysis, generalized singular value decomposition, QR decomposition, trace optimization.”
• Gabor Feature Based Sparse Representation for Face Recognition with Gabor Occlusion Dictionary (2010, methodology): “Therefore, subspace learning and manifold learning methods have been dominantly and successfully used in appearance based FR [4], [5], [6], [7], [8], [9], [10], [11].”
• Metaface learning for sparse representation based face recognition (2010, methodology): “Therefore, subspace learning and manifold learning methods have been dominantly and successfully used in appearance based FR [1-8], which includes Eigenface, Fisherface [1-3], locality preserving projection (LPP) [6], local discriminant embedding (LDE) [7], unsupervised discriminant projection (UDP)”
• Eigenfeature Regularization and Extraction in Face Recognition (2008, methodology): “Subsequent work [28] extracts features separately from the principal and null spaces of the within-class scatter matrix.”