Refine
Has Fulltext
- yes (12) (remove)
Year of publication
Document Type
- Doctoral Thesis (12) (remove)
Language
- English (12)
Is part of the Bibliography
- yes (12) (remove)
Keywords
- eye movements (12) (remove)
Eye movements serve as a window into ongoing visual-cognitive processes and can thus be used to investigate how people perceive real-world scenes. A key issue for understanding eye-movement control during scene viewing is the roles of central and peripheral vision, which process information differently and are therefore specialized for different tasks (object identification and peripheral target selection respectively). Yet, rather little is known about the contributions of central and peripheral processing to gaze control and how they are coordinated within a fixation during scene viewing. Additionally, the factors determining fixation durations have long been neglected, as scene perception research has mainly been focused on the factors determining fixation locations. The present thesis aimed at increasing the knowledge on how central and peripheral vision contribute to spatial and, in particular, to temporal aspects of eye-movement control during scene viewing. In a series of five experiments, we varied processing difficulty in the central or the peripheral visual field by attenuating selective parts of the spatial-frequency spectrum within these regions. Furthermore, we developed a computational model on how foveal and peripheral processing might be coordinated for the control of fixation duration. The thesis provides three main findings. First, the experiments indicate that increasing processing demands in central or peripheral vision do not necessarily prolong fixation durations; instead, stimulus-independent timing is adapted when processing becomes too difficult. Second, peripheral vision seems to play a prominent role in the control of fixation durations, a notion also implemented in the computational model. The model assumes that foveal and peripheral processing proceed largely in parallel and independently during fixation, but can interact to modulate fixation duration. Thus, we propose that the variation in fixation durations can in part be accounted for by the interaction between central and peripheral processing. Third, the experiments indicate that saccadic behavior largely adapts to processing demands, with a bias of avoiding spatial-frequency filtered scene regions as saccade targets. We demonstrate that the observed saccade amplitude patterns reflect corresponding modulations of visual attention. The present work highlights the individual contributions and the interplay of central and peripheral vision for gaze control during scene viewing, particularly for the control of fixation duration. Our results entail new implications for computational models and for experimental research on scene perception.
A central insight from psychological studies on human eye movements is that eye movement patterns are highly individually characteristic. They can, therefore, be used as a biometric feature, that is, subjects can be identified based on their eye movements. This thesis introduces new machine learning methods to identify subjects based on their eye movements while viewing arbitrary content. The thesis focuses on probabilistic modeling of the problem, which has yielded the best results in the most recent literature. The thesis studies the problem in three phases by proposing a purely probabilistic, probabilistic deep learning, and probabilistic deep metric learning approach. In the first phase, the thesis studies models that rely on psychological concepts about eye movements. Recent literature illustrates that individual-specific distributions of gaze patterns can be used to accurately identify individuals. In these studies, models were based on a simple parametric family of distributions. Such simple parametric models can be robustly estimated from sparse data, but have limited flexibility to capture the differences between individuals. Therefore, this thesis proposes a semiparametric model of gaze patterns that is flexible yet robust for individual identification. These patterns can be understood as domain knowledge derived from psychological literature. Fixations and saccades are examples of simple gaze patterns. The proposed semiparametric densities are drawn under a Gaussian process prior centered at a simple parametric distribution. Thus, the model will stay close to the parametric class of densities if little data is available, but it can also deviate from this class if enough data is available, increasing the flexibility of the model. The proposed method is evaluated on a large-scale dataset, showing significant improvements over the state-of-the-art. Later, the thesis replaces the model based on gaze patterns derived from psychological concepts with a deep neural network that can learn more informative and complex patterns from raw eye movement data. As previous work has shown that the distribution of these patterns across a sequence is informative, a novel statistical aggregation layer called the quantile layer is introduced. It explicitly fits the distribution of deep patterns learned directly from the raw eye movement data. The proposed deep learning approach is end-to-end learnable, such that the deep model learns to extract informative, short local patterns while the quantile layer learns to approximate the distributions of these patterns. Quantile layers are a generic approach that can converge to standard pooling layers or have a more detailed description of the features being pooled, depending on the problem. The proposed model is evaluated in a large-scale study using the eye movements of subjects viewing arbitrary visual input. The model improves upon the standard pooling layers and other statistical aggregation layers proposed in the literature. It also improves upon the state-of-the-art eye movement biometrics by a wide margin. Finally, for the model to identify any subject — not just the set of subjects it is trained on — a metric learning approach is developed. Metric learning learns a distance function over instances. The metric learning model maps the instances into a metric space, where sequences of the same individual are close, and sequences of different individuals are further apart. This thesis introduces a deep metric learning approach with distributional embeddings. The approach represents sequences as a set of continuous distributions in a metric space; to achieve this, a new loss function based on Wasserstein distances is introduced. The proposed method is evaluated on multiple domains besides eye movement biometrics. This approach outperforms the state of the art in deep metric learning in several domains while also outperforming the state of the art in eye movement biometrics.