This project strives to achieve high performance object localization with Convolutional Neural Networks (CNNs) - in particular for pupil centers in the context of remote eye tracking systems. Three different network architectures suitable to the task are developed, evaluated and compared - one based on regression using fully connected layers, one Fully Convolutional Network and one Deconvolutional Network. The best performing model achieves a mean error of only 0.52 pixel distance and a median error of 0.42 pixel distance compared to the ground truth annotations. The 95th percentile lies at 1.12 pixel error. This exceeds the performance of current state-of-the-art pupil center detection algorithms by an order of magnitude, a result that can...
A deep learning model for video saliency detection is proposed and trained. The neural network archi...
Various methods have been used to estimate the pupil location within an image or a real-time video f...
Various methods have been used to estimate the pupil location within an image or a real-time video f...
This project strives to achieve high performance object localization with Convolutional Neural Netwo...
Remote eye tracking technology has suffered an increasing growth in recent years due to its applicab...
Remote eye tracking technology has suffered an increasing growth in recent years due to its applicab...
Deep learning has opened new doors to things that were only imaginable before. When it comes to eye ...
Deep learning has opened new doors to things that were only imaginable before. When it comes to eye ...
Eye centre localisation is critical to eye tracking systems of various forms and with applications i...
Remote eye tracking technology has suffered an increasing growth in recent years due to its applicab...
This thesis empirically studies transfer learning as a calibration framework for Convolutional Neura...
This thesis empirically studies transfer learning as a calibration framework for Convolutional Neura...
Eye centre localisation is critical to eye tracking systems of various forms and with applications i...
Eye centre localisation is critical to eye tracking systems of various forms and with applications i...
A deep learning model for video saliency detection is proposed and trained. The neural network archi...
A deep learning model for video saliency detection is proposed and trained. The neural network archi...
Various methods have been used to estimate the pupil location within an image or a real-time video f...
Various methods have been used to estimate the pupil location within an image or a real-time video f...
This project strives to achieve high performance object localization with Convolutional Neural Netwo...
Remote eye tracking technology has suffered an increasing growth in recent years due to its applicab...
Remote eye tracking technology has suffered an increasing growth in recent years due to its applicab...
Deep learning has opened new doors to things that were only imaginable before. When it comes to eye ...
Deep learning has opened new doors to things that were only imaginable before. When it comes to eye ...
Eye centre localisation is critical to eye tracking systems of various forms and with applications i...
Remote eye tracking technology has suffered an increasing growth in recent years due to its applicab...
This thesis empirically studies transfer learning as a calibration framework for Convolutional Neura...
This thesis empirically studies transfer learning as a calibration framework for Convolutional Neura...
Eye centre localisation is critical to eye tracking systems of various forms and with applications i...
Eye centre localisation is critical to eye tracking systems of various forms and with applications i...
A deep learning model for video saliency detection is proposed and trained. The neural network archi...
A deep learning model for video saliency detection is proposed and trained. The neural network archi...
Various methods have been used to estimate the pupil location within an image or a real-time video f...
Various methods have been used to estimate the pupil location within an image or a real-time video f...