Sight-seeing in the eyes of deep neural networks

We address the interpretability of convolutional neural networks (CNNs) for predicting a geo-location from an image. In a pilot experiment we classify images of Pittsburgh vs Tokyo and visualize the learned CNN filters. We found that varying the CNN architecture leads to variating in the visualized filters. This calls for further...

Deep Learning for Pixelwise Classification of Hyperspectral Images: A generalizing model for a fixed scene subject to temporally changing weather, lighting and seasonal conditions

In hyperspectral (HS) imaging, for every pixel a spectrum of wavelengths is captured. These spectra represent material properties, i.e. the spectral signatures. So, classification of HS imagery is based on material properties. This thesis describes a framework to perform pixelwise classification of HS images of a fixed scene subject to varying...

Learning the scale of image features in Convolutional Neural Networks

The millions of filter weights in Convolutional Neural Networks (CNNs), all have a well-defined and analytical expression for the partial derivative to the loss function. Therefor these weights can be learned from data with a technique called gradient descent optimization. While the filter weights have a well-defined derivative, the filter size...

Quantifying and Predicting Urban Attractiveness with Street-View Data and Convolutional Neural Networks

Analysing attractiveness of places in a region is beneficial to support urban planning and policy making. However, the attractiveness of a place is a subjective and high-level concept which is difficult to quantify. The existing methods rely on traditional surveys which may require high cost and have low scalability. This thesis attempts to...

Generating virtual CT from MRI using fully convolutional neural networks with improved structure quality

<br/>

Reinforcement Learning of Visual Features

The digital environment has an ever increasing amount smart programs. Programs that also get smarter every day. They help us filtering spam e-mail and they adjust to show us personalized advertisements. These smart programs observe people and serve (other) people. A robot can be seen as a program with a body. Make the program smart enough and it...