IEEE Transactions on Pattern Analysis and Machine Intelligence · 2004 · 3.6K citations · 19 references
ComputingComputer ScienceComputer VisionFeature EngineeringMachine VisionDimensionality ReductionFace RecognitionFeature ExtractionPattern RecognitionPrincipal Component AnalysisRepresentation LearningCovariance Matrix DecompositionFace Feature ExtractionFace RepresentationImage Feature ExtractionImage Feature RepresentationImage RepresentationPrincipal Components
As opposed to PCA, 2DPCA operates directly on 2‑D image matrices, eliminating the need to vectorize images before feature extraction. This paper develops two‑dimensional principal component analysis (2DPCA) for image representation. 2DPCA constructs an image covariance matrix directly from the original image matrices, derives its eigenvectors for feature extraction, and its performance was evaluated through experiments on the ORL, AR, and Yale face databases. 2DPCA achieved higher recognition rates and more efficient feature extraction than PCA across all trials.
In this paper, a new technique coined two-dimensional principal component analysis (2DPCA) is developed for image representation. As opposed to PCA, 2DPCA is based on 2D image matrices rather than 1D vectors so the image matrix does not need to be transformed into a vector prior to feature extraction. Instead, an image covariance matrix is constructed directly using the original image matrices, and its eigenvectors are derived for image feature extraction. To test 2DPCA and evaluate its performance, a series of experiments were performed on three face image databases: ORL, AR, and Yale face databases. The recognition rate across all trials was higher using 2DPCA than PCA. The experimental results also indicated that the extraction of image features is computationally more efficient using 2DPCA than PCA.
19
Matthew Turk, Alex Pentland · Journal of Cognitive Neuroscience · 1991 · 13.8K citations · Full text
Aleix M. Martı́nez · CTIT technical reports series · 1998 · 2.8K citations