Determining comprehension and quality of TV programs using eye-gaze tracking

Sawahata, Yasuhito; Khosla, Rajiv; Komine, Kazuteru; Hiruma, Nobuyuki; Itou, Takayuki; Watanabe, Seiji; Suzuki, Yuji; Hara, Yumiko; Issiki, Nobuo

doi:10.1016/j.patcog.2007.10.010

Cited by 29 publications

(21 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This would create a high degree of consistency in where viewers attend in a dynamic scene. Such clustering of gaze during dynamic scene viewing has been reported in several studies [48,[54][55][56][57][58][59][60][61][62] and is believed to be an integral part of film viewing (see [63] for a review). For example, Goldstein et al [54] showed 20 adults six long clips from Hollywood movies and found that for more than half of the viewing time the distribution of fixations from all viewers occupied less than 12% of the screen area.…”

Section: Introductionmentioning

confidence: 54%

“…For example, Goldstein et al [54] showed 20 adults six long clips from Hollywood movies and found that for more than half of the viewing time the distribution of fixations from all viewers occupied less than 12% of the screen area. This Attentional Synchrony effect is observed for feature films [48,[54][55][56][57][58][60][61][62][63], television [59], and videos of realworld scenes [53,60,64].…”

Section: Introductionmentioning

confidence: 94%

“…Reported success of density estimation via expectation/ maximization [103] of a parametric family distribution in Sawahata et al [59] prompted us to use mixture models as our clustering method. Sawahata et al investigated an entropy measure calculated from the log likelihood of a mixture model in order to determine viewer comprehension during dynamic scene viewing.…”

Section: Selection Of Visual Features During Dynamic Scene Viewingmentioning

confidence: 99%

“…zero covariance Gaussians centered at fixations, is dealt with in the BIC term which penalizes the number of clusters. We refer the reader to Bishop [103] and Sawahata et al [59] for more details on GMM modeling and model selection. A few example stills are shown in Fig.…”

Section: Selection Of Visual Features During Dynamic Scene Viewingmentioning

confidence: 99%

See 3 more Smart Citations

Clustering of Gaze During Dynamic Scene Viewing is Predicted by Motion

et al. 2010

View full text Add to dashboard Cite

Where does one attend when viewing dynamic scenes? Research into the factors influencing gaze location during static scene viewing have reported that low-level visual features contribute very little to gaze location especially when opposed by high-level factors such as viewing task. However, the inclusion of transient features such as motion in dynamic scenes may result in a greater influence of visual features on gaze allocation and coordination of gaze across viewers. In the present study, we investigated the contribution of low-to mid-level visual features to gaze location during free-viewing of a large dataset of videos ranging in content and length. Signal detection analysis on visual features and Gaussian Mixture Models for clustering gaze was used to identify the contribution of visual features to gaze location. The results show that mid-level visual features including corners and orientations can distinguish between actual gaze locations and a randomly sampled baseline. However, temporal features such as flicker, motion, and their respective contrasts were the most predictive of gaze location. Additionally, moments in which all viewers' gaze tightly clustered in the same location could be predicted by motion. Motion and mid-level visual features may influence gaze allocation in dynamic scenes, but it is currently unclear whether this influence is involuntary or due to correlations with higher order factors such as scene semantics.

show abstract

Section: Introductionmentioning

confidence: 54%

Section: Introductionmentioning

confidence: 94%

Section: Selection Of Visual Features During Dynamic Scene Viewingmentioning

confidence: 99%

Section: Selection Of Visual Features During Dynamic Scene Viewingmentioning

confidence: 99%

See 2 more Smart Citations

Clustering of Gaze During Dynamic Scene Viewing is Predicted by Motion

et al. 2010

View full text Add to dashboard Cite

show abstract

“…Also, being aware of the quality of service (QoS) transmission [16], such as delay, delay variation (jitter), and bandwidth consumed, is relevant to the operation and maintenance of the transmission network. All these situations need to be planned in order to get a closer view of the video quality users are receiving.…”

Section: Introductionmentioning

confidence: 99%

Evaluating the effectiveness of the use of teaching materials in video format in distance learning environments

Stiubiener

Silveira

Matushima

et al. 2012

2012 Frontiers in Education Conference Proceedings

View full text Add to dashboard Cite

Projections indicate that video will be responsible for 90% of network traffic until the end of the decade. Considering the high cost of media production and, mainly, the high consumption of network resources for applications that explore it, it will be extremely important to monitor such use, preventing against wrong or distorted views of this consumption. Within this scenario, it is necessary to develop mechanisms and tools that provide a clear vision of video effectiveness consumption, as well as the measuring of "viewer engagement". What tools can be developed to evaluate such consumption? What should be measured, i.e., which are the parameters and metrics necessary, and how can we obtain them, in order to measure the effectiveness of video applications use, for example in engineering courses and distance learning environments? These are some of the questions considered in this paper, especially evaluated within a scenario of video usage for teaching/learning purpose.

show abstract