Hybrid algorithms for quantum image processing with quantum machine learning.


In image processing, the amount of data to be processed is growing faster than the speed and accuracy of analysis algorithms and computer hardware due to higher resolutions, higher dimensionality, wider availability of camera systems and new imaging techniques. Quantum computing (QC) promises exponential time and memory savings over classical systems.

We do not expect any real quantum advantage in image processing in the near future, but we want to achieve quantum readiness. Combining classical image processing algorithms with Quantum Machine Learning (QML) has the potential to process images of practical size in Noisy Intermediate-Scale Quantum (NISQ) computers already. As a first step, I will explore the connection between supervised QML, kernel-based methods and the Fourier representations of quantum kernels.