A Probabilistic Time-Evolving Approach to Scanpath Prediction
This work addresses the challenge of predicting human visual attention scanpaths, which is important for applications in human-computer interaction and vision science, representing a strong incremental improvement over existing methods.
The paper tackles the problem of scanpath prediction by introducing a probabilistic time-evolving approach using Bayesian deep learning and a novel spatio-temporal loss function, achieving results that outperform state-of-the-art methods and closely match human baseline performance.
Human visual attention is a complex phenomenon that has been studied for decades. Within it, the particular problem of scanpath prediction poses a challenge, particularly due to the inter- and intra-observer variability, among other reasons. Besides, most existing approaches to scanpath prediction have focused on optimizing the prediction of a gaze point given the previous ones. In this work, we present a probabilistic time-evolving approach to scanpath prediction, based on Bayesian deep learning. We optimize our model using a novel spatio-temporal loss function based on a combination of Kullback-Leibler divergence and dynamic time warping, jointly considering the spatial and temporal dimensions of scanpaths. Our scanpath prediction framework yields results that outperform those of current state-of-the-art approaches, and are almost on par with the human baseline, suggesting that our model is able to generate scanpaths whose behavior closely resembles those of the real ones.