Quantized Fisher Discriminant Analysis
This addresses a gap in integrating machine learning with information theory for image compression and classification, though it appears incremental as it builds on existing FDA methods.
The paper tackles the problem of combining machine learning and information theory by proposing Quantized Fisher Discriminant Analysis (QFDA), a subspace learning method that discriminates quantized images in the DCT domain as effectively as non-quantized images with FDA, enabling the use of compressed images without significant loss in classification accuracy, as demonstrated on AT&T face and Fashion MNIST datasets.
This paper proposes a new subspace learning method, named Quantized Fisher Discriminant Analysis (QFDA), which makes use of both machine learning and information theory. There is a lack of literature for combination of machine learning and information theory and this paper tries to tackle this gap. QFDA finds a subspace which discriminates the uniformly quantized images in the Discrete Cosine Transform (DCT) domain at least as well as discrimination of non-quantized images by Fisher Discriminant Analysis (FDA) while the images have been compressed. This helps the user to throw away the original images and keep the compressed images instead without noticeable loss of classification accuracy. We propose a cost function whose minimization can be interpreted as rate-distortion optimization in information theory. We also propose quantized Fisherfaces for facial analysis in QFDA. Our experiments on AT&T face dataset and Fashion MNIST dataset show the effectiveness of this subspace learning method.