When: Thursday 16th March 2017, 13:00-14:00.
Included in meeting:
Pubudu Thilan will be presenting on The Impact of Color to Gray-scale Conversion and Image Resolution in Face Recognition
In face recognition, with reliance on principle component analysis (PCA), oftentimes grayscale images are utilized because of their simplicity and achieved reduction in data load requirements especially in comparison to that of color images. This presentation compares eight different color-to-grayscale conversion algorithms and demonstrates the accuracy of each. This presentation details the specifics of an approach that uses PCA as a feature extraction method in face recognition and provides results that confirm that the PCA approach works well even for considerable reduction in image resolution which, for searches across large populations, can greatly minimize data loads and, consequently, be considerably more timely.
Brodie Lawson will be presenting on An Introduction to Gaussian Process Regression/Emulation
Gaussian Processes, the infinite dimensional extension of the normal distribution, can be used for many things including regression and the construction of emulators for complex models. I have recently discovered this technique and have been experimenting with the use to emulate models for the cardiac action potential. In this talk I’ll go over the basics of the technique, including special tricks regarding implementation and achieving smoothness between ‘neighbouring’ GPs, which are otherwise typically independent with no relationship with each other.
We look forward to seeing you there!
Matt and Shovanur