Predicting Eye Fixations Under Distortion Using Bayesian Observers
This work is incrementally important for researchers studying human perception and visual quality, as it explores how image distortions affect visual attention.
This paper investigates the impact of image distortions, specifically JPEG compression artifacts like blocking and ringing, on human visual attention. By modifying Bayesian visual search models (MAP and ELM), the study observed that these compression artifacts indeed influence eye fixation movements.
Visual attention is very an essential factor that affects how human perceives visual signals. This report investigates how distortions in an image could distract human's visual attention using Bayesian visual search models, specifically, Maximum-a-posteriori (MAP) \cite{findlay1982global}\cite{eckstein2001quantifying} and Entropy Limit Minimization (ELM) \cite{najemnik2009simple}, which predict eye fixation movements based on a Bayesian probabilistic framework. Experiments on modified MAP and ELM models on JPEG-compressed images containing blocking or ringing artifacts were conducted and we observed that compression artifacts can affect visual attention. We hope this work sheds light on the interactions between visual attention and perceptual quality.