This document describes the work performed the master thesis on a new approach to visualisation for music. We investigate how visualisation of (latent) characteristics of a song can improve user-guided search and exploration. The need for such visualisation is described, followed by a formulation of the research questions, literature reviews, our method, design and evaluation, a discussion of our results and an outlook on further development of this project.

This report documents the DaisyRiot global illumination renderer. The renderer applies the radiosity method and uses spectral rendering in order to simulate fluorescent materials: materials can reflect energy of a given frequency at another frequency. Rendering is accelerated by NVIDIA OptiX and CUDA parallelization. In this document you will find details of the design of the renderer, its architecture, a discussion of its results and performance and a reflection on its future. All source code is released online: github.com/asylunatic/DaisyRiot.

Before you lies the report that resulted from the Individual Research Project of Vera Hoveling. For this project, a neural network has been developed to complete partially rendered images: BobRossNet. In this document you will find the details of the project, the architecture of BobRossNet, its training, results and a reflection on its future.