Single-trial EEG classification of artifacts in videos

Abstract

In this article we use an ElectroEncephaloGraph (EEG) to explore the perception of artifacts that typically appear during rendering and determine the perceptual quality of a sequence of images. Although there is an emerging interest in using an EEG for image quality assessment, one of the main impediments to the use of an EEG is the very low Signal-to-Noise Ratio (SNR) which makes it exceedingly difficult to distinguish neural responses from noise. Traditionally, event-related potentials have been used for analysis of EEG data. However, they rely on averaging and so require a large number of participants and trials to get meaningful data. Also, due the the low SNR ERP's are not suited for single-trial classification.We propose a novel wavelet-based approach for evaluating EEG signals which allows us to predict the perceived image quality from only a single trial. Our wavelet-based algorithm is able to filter the EEG data and remove noise, eliminating the need for many participants or many trials. With this approach it is possible to use data from only 10 electrode channels for single-trial classification and predict the presence of an artifact with an accuracy of 85%. We also show that it is possible to differentiate and classify a trial based on the exact type of artifact viewed. Our work is particularly useful for understanding how the human visual system responds to different types of degradations in images and videos. An understanding of the perception of typical image-based rendering artifacts forms the basis for the optimization of rendering and masking algorithms.

Cognitive, affective, and experience correlates of speech quality perception in complex listening conditions

“…); this is similar to what is done with listening quality assessment of text-to-speech systems. As in previous literature, we have shown that EEG can be used to gather cognitive, distortion, and quality-of-experience insights ( [6], [8], [9], [16]). …”

“…For example, [6] showed that simple degradations in the speech signal (e.g., multiplicative noise) could be processed by humans at an unconscious level; similar findings were also observed for artificial audiovisual stimuli [7]. Recently, P300 peak amplitudes were shown to be significantly correlated with the level of distortions added to audiovisual stimuli [8] and classifiers were developed to discriminate between clean and distorted visual stimuli based only on EEG [9]. Lastly, EEG has also been used to characterize user emotional state while watching Youtube videos [10].…”

Review on using physiology in quality of experience

“…However, those measures did not show any significant change with the change of quality [28]. Mustafa et al [13] used a 5.6 s long low-complexity video that showed a person walking. They introduced different kinds of distortion, such as partial freezing and blurring (on the person, and on a greater area), ghosting of the movement plus the original video, resulting into six different video conditions.…”

“…2. Using classification techniques this can be transferred to online analysis of incoming physiological signals, such as deciding whether the brain activity of the proceeding stimulus was evoked by one special class of stimuli [13]. To address this in the case of Quality of Experience an exemplary class of degradation can be used.…”

“…the person) the P300 is larger compared to distortions that contain a greater area. In addition, power frequency analysis was performed [13] on the same data set. When analyzing the frequency range of 10 Hz to 20 Hz, they could show that artifacts on the person evoke again the greatest change compared to the reference condition.…”