Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
This lecture covers the application of the Karhunen Loève transform (KLT) in face recognition using eigenfaces, the discrete Fourier transform (DFT) and circulant matrices, and the optimality of DFT for stationary processes. It also delves into classical 2D transforms like the discrete cosine transform (DCT) and other separable transforms. The lecture further explores the implementation of separable transforms, the basis functions of DCT, and other transforms like the sine transform and Hadamard transform. Additionally, it discusses wavelet transforms, multirate operations, perfect reconstruction filterbanks, and the history of wavelets. The presentation concludes with signal processing developments from the 'DCT era' to the emergence of perfect-reconstruction filterbanks in the early 1980s.