SGaze: A Data-Driven Eye-Head Coordination Model for Realtime Gaze Prediction

Zhang, Congyi; Li, Sheng; Wang, Guoping; Manocha, Dinesh

doi:10.1109/tvcg.2019.2899187

Cited by 50 publications

(27 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Static omnidirectional panoramas are one of the most widespread distribution formats for 360°content. Several works have attempted to analyze [17,26] and predict [1,10,21,39,40] user behavior and gaze scanpaths in such content. In the absence of eye-tracked gaze data, which is not typically accessible in VR devices, other works have analyzed head movements instead, showing that head orientation is a valuable proxy of gaze behavior [34,38].…”

Section: Related Workmentioning

confidence: 99%

Exploring the impact of 360° movie cuts in users’ attention

Marañes

Gutiérrez

Serrano

2020

2020 IEEE Conference on Virtual Reality and 3D User Interfaces (VR)

View full text Add to dashboard Cite

Figure 1: Example frame from the movie The People's House produced by Felix & Paul Studios used to study how people explore professionally edited 360°movies. For analyzing how users behave across movie cuts, we start from head orientation data (left), and we take into account visual attention priors in order to compute saliency maps (right) describing users' gaze. ABSTRACTVirtual Reality (VR) has grown since the first devices for personal use became available on the market. However, the production of cinematographic content in this new medium is still in an early exploratory phase. The main reason is that cinematographic language in VR is still under development, and we still need to learn how to tell stories effectively. A key element in traditional film editing is the use of different cutting techniques, in order to transition seamlessly from one sequence to another. A fundamental aspect of these techniques is the placement and control over the camera. However, VR content creators do not have full control of the camera. Instead, users in VR can freely explore the 360 • of the scene around them, which potentially leads to very different experiences. While this is desirable in certain applications such as VR games, it may hinder the experience in narrative VR. In this work, we perform a systematic analysis of users' viewing behavior across cut boundaries while watching professionally edited, narrative 360 • videos. We extend previous metrics for quantifying user behavior in order to support more complex and realistic footage, and we introduce two new metrics that allow us to measure users' exploration in a variety of different complex scenarios. From this analysis, (i) we confirm that previous insights derived for simple content hold for professionally edited content, and (ii) we derive new insights that could potentially influence VR content creation, informing creators about the impact of different cuts in the audience's behavior.

show abstract

Section: Related Workmentioning

confidence: 99%

Exploring the impact of 360° movie cuts in users’ attention

Marañes

Gutiérrez

Serrano

2020

2020 IEEE Conference on Virtual Reality and 3D User Interfaces (VR)

View full text Add to dashboard Cite

show abstract

“…The natural behavior or head, gaze and torso directions and timing differences were studied in Sidenmark and Gellersen (2019a) based on measurements on when the respective moves have started and how the fixation directions differ. Hu et al (2019) collected gaze and head orientation data from a number of participants viewing 360 °virtual worlds using HMD and a gaze tracker. The participants were able to freely navigate in the scenes.…”

Section: Prior Research On Head and Gaze Dynamicsmentioning

confidence: 99%

Head and Gaze Orientation in Hemispheric Image Viewing

Kangas¹,

Špakov²,

Raisamo³

et al. 2022

Front. Virtual Real.

View full text Add to dashboard Cite

Head mounted displays provide a good platform for viewing of immersive 360° or hemispheric images. A person can observe an image all around, just by turning his/her head and looking at different directions. The device also provides a highly useful tool for studying the observer’s gaze directions and head turns. We aimed to explore the interplay between participant’s head and gaze directions and collected head and gaze orientation data while participants were asked to view and study hemispheric images. In this exploration paper we show combined visualizations of both the head and gaze orientations and present two preliminary models of the relation between the gaze and the head orientations. We also show results of an analysis of the gaze and head behavior in relation to the given task/question.

show abstract

“…Combining Behavioral Data Sources in VR: Hu et al [12] observe a linear correlation between gaze positions and the head rotation's angular velocities. They implement a real-time gaze position prediction model.…”

Section: Related Workmentioning

confidence: 99%