A saliency-based method of simulating visual attention in virtual scenes

Oyekoya, Oyewole; Steptoe, William; Steed, Anthony

doi:10.1145/1643928.1643973

Cited by 21 publications

(18 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When a frequently searched game object is modified to share perceptual features such as color or orientation with a target item, the item will attract attention [Bernhard et al 2011]. Saliency models have been employed to animate the gaze behavior of virtual characters [Oyekoya et al 2009] and crowds [Grillon and Thalmann 2009].…”

Section: Related Workmentioning

confidence: 99%

An Automated High-Level Saliency Predictor for Smart Game Balancing

Koulieris

Drettakis

Cunningham

et al. 2014

ACM Trans. Appl. Percept.

View full text Add to dashboard Cite

Successfully predicting visual attention can significantly improve many aspects of computer graphics: scene design, interactivity and rendering. Most previous attention models are mainly based on low-level image features, and fail to take into account highlevel factors such as scene context, topology, or task. Low-level saliency has previously been combined with task maps, but only for predetermined tasks. Thus, the application of these methods to graphics (e.g., for selective rendering) has not achieved its full potential. In this article, we present the first automated high-level saliency predictor incorporating two hypotheses from perception and cognitive science that can be adapted to different tasks. The first states that a scene is comprised of objects expected to be found in a specific context as well objects out of context which are salient (scene schemata) while the other claims that viewer's attention is captured by isolated objects (singletons). We propose a new model of attention by extending Eckstein's Differential Weighting Model. We conducted a formal eye-tracking experiment which confirmed that object saliency guides attention to specific objects in a game scene and determined appropriate parameters for a model. We present a GPUbased system architecture that estimates the probabilities of objects to be attended in real-time. We embedded this tool in a game level editor to automatically adjust game level difficulty based on object saliency, offering a novel way to facilitate game design. We perform a study confirming that game level completion time depends on object topology as predicted by our system.

show abstract

Section: Related Workmentioning

confidence: 99%

An Automated High-Level Saliency Predictor for Smart Game Balancing

Koulieris

Drettakis

Cunningham

et al. 2014

ACM Trans. Appl. Percept.

View full text Add to dashboard Cite

show abstract

“…In this paper, we extend a previous work on gaze modelling in two ways: we develop the attention model to drive both head and eye gaze and we integrate this model into SL. This provides an excellent platform for testing in general avatar encounters, and in the Experiments section, we discuss user experiments based on this platform.…”

Section: Methodsmentioning

confidence: 99%

“…The attention model is designed to adapt to the complex interaction within the scene and is based on data‐driven modelling of gaze behaviour.…”

Section: Methodsmentioning

confidence: 99%

“…In this paper, we study this approach in terms of gaze and head motion of the humanoid, as well as in terms of social behaviour. Initially designed as a gaze model, this paper extends a previous model of visual attention to animate the head. The model takes as input the intrinsic object properties and generates eye and head reactions to objects within the environment.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Modelling selective visual attention for autonomous virtual characters

Kokkinara

Oyekoya

Steed

2011

Computer Animation & Virtual

Self Cite

View full text Add to dashboard Cite

Autonomous virtual characters (AVCs) are becoming more prevalent both for real-time interaction and also as digital actors in film and TV production. AVCs require believable virtual human animations, accompanied by natural attention generation, and thus the software that controls the AVCs needs to model when and how to interact with the objects and other characters that exist in the virtual environment. This paper models automatic attention behaviour using a saliency model that generates plausible targets for combined gaze and head motions. The model was compared with the default behaviour of the Second Life (SL) system in an object observation scenario while it was compared with real actors' behaviour in a conversation scenario. Results from a study run within the SL system demonstrate a promising attention model that is not just believable and realistic but also adaptable to varying task, without any prior knowledge of the virtual scene. Copyright

show abstract

“…In this implementation, eye tracking is used to deliver gaze data, which acts as input to the lid saccade model. The operation of the lid saccade model may then be interrupted by blink signals: detected by the eye tracker; generated from a gaze model 34 ; or inferred from other behavioural tracking devices including microphones monitoring verbal utterances, or body tracking monitoring gestures. Following completion of a blink, the lid saccade model then resumes control.…”

Section: Blink Modelmentioning

confidence: 99%

Eyelid kinematics for virtual characters

Steptoe

Oyekoya²,

Steed³

2010

Computer Animation & Virtual

Self Cite

View full text Add to dashboard Cite

When compared to gaze, animation of the eyelids has been largely overlooked in the computer graphics literature. Eyelid movement plays an important part both in conveying accurate gaze direction and in improving the visual appearance of virtual characters. Eyelids have two major motion components: lid saccades that follow the vertical rotation the eyes, and blinking. Derived from literature in ophthalmology and psychology, this paper presents parametric models for both motion types, and emphasizes their dynamic temporal behaviour. Experimental validation classifies model-generated animation as similar to that encoded from expensive motion captured data, and significantly exceeding linearly interpolated animation. Lid saccades are bilateral shifts of the eyelids initiated by changes in the vertical rotation of the eyes, so that the eyelids follow the direction of gaze to ensure they do not obscure the pupils 2 . Blinking is the act of extremely brief closure of the eyelids, and may be voluntary or involuntary 3 . Physiologically, the eyelids serve to protect the eye from debris and rays of harmful intensity , and also to regularly spread the tears and other secretion on the eye surface to keep it continually moist 2 . Hence, eyelid movement is a ubiquitous element of human behaviour.As both real-time and offline character animation advance, there is an ever-growing need to consider the broad scope of what gives rise to visual and behavioural realism. Increasingly sophisticated geometrical and behavioural models of humans and components of humans have been presented for a applications including games, film and embodied conversational agents. When looking at a photograph or painting of a human, one typically spends 60% of the time looking at the eyes and 15% at the mouth 4 . It is no surprise then, that much of this research has focussed on the face 5 , and in particular the eyes 6 . Lee et al. 6 identify the high correlation between the motion of the eyes and the corresponding eyelid movements as a possible extension to their gaze model. While some subsequent gaze models have incorporated some form of eyelid animation to complement gaze motion 7-9 , eyelid animation has played a secondary role. Hence, the implementation details of realistic human eyelid dynamics has been left undefined. This paper presents two parametric models generating physiologically-accurate lid saccade and blink animations for virtual characters. The models are detailed algorithmically, together with an example implementation for avatar-mediated communication between users of networked immersive CAVE TM -like systems. The models are then validated against current methods of eyelid animation, including motion capture and linear interpolation, before discussion, limitations and future work are addressed.

show abstract

A saliency-based method of simulating visual attention in virtual scenes

Cited by 21 publications

References 32 publications

An Automated High-Level Saliency Predictor for Smart Game Balancing

An Automated High-Level Saliency Predictor for Smart Game Balancing

Modelling selective visual attention for autonomous virtual characters

Eyelid kinematics for virtual characters

Contact Info

Product

Resources

About