A Behavioral Analysis of Computational Models of Visual Attention

Robots often incorporate computational models of visual attention to streamline processing. Even though the number of visual attention systems employed on robots has increased dramatically in recent years, the evaluation of these systems has remained primarily qualitative and subjective. We introduce quantitative methods for evaluating computational models of visual attention by direct comparison with gaze trajectories acquired from humans. In particular, we focus on the need for metrics based not on distances within the image plane, but that instead operate at the level of underlying features. We present a framework, based on dimensionality-reduction over the features of human gaze trajectories, that can simultaneously be used for both optimizing a particular computational model of visual attention and for evaluating its performance in terms of similarity to human behavior. We use this framework to evaluate the Itti et al. (1998) model of visual attention, a computational model that serves as the basis for many robotic visual attention systems.

[1]  K. Turano,et al.  Oculomotor strategies for the direction of gaze tested with a real-world activity , 2003, Vision Research.

[2]  Edward H. Adelson,et al.  The Laplacian Pyramid as a Compact Image Code , 1983, IEEE Trans. Commun..

[3]  L. Itti.  Quantitative modelling of perceptual salience at human eye position , 2006 .

[4]  Masahiro Fujita,et al.  AIBO: Toward the Era of Digital Creatures , 2001, Int. J. Robotics Res..

[5]  J. Gallant,et al.  Goal-Related Activity in V4 during Free Viewing Visual Search Evidence for a Ventral Stream Visual Salience Map , 2003, Neuron.

[6]  D. Robinson,et al.  Shared neural control of attentional shifts and eye movements , 1996, Nature.

[7]  Bruce A. Draper,et al.  Evaluation of selective attention under similarity transformations , 2005, Comput. Vis. Image Underst..

[8]  John K. Tsotsos,et al.  Modeling Visual Attention via Selective Tuning , 1995, Artif. Intell..

[9]  Laurent Itti,et al.  Causal saliency effects during natural vision , 2006, ETRA.

[10]  Christof Koch,et al.  Control of Selective Visual Attention: Modeling the Where Pathway , 1995, NIPS.

[11]  John K. Tsotsos.  A ‘complexity level’ analysis of immediate vision , 2004, International Journal of Computer Vision.

[12]  S. Petersen,et al.  Contributions of the pulvinar to visual spatial attention , 1987, Neuropsychologia.

[13]  Laurent Itti,et al.  Realistic avatar eye and head animation using a neurobiological model of visual attention , 2004, SPIE Optics + Photonics.

[14]  S. Treue.  Visual attention: the where, what, how and why of saliency , 2003, Current Opinion in Neurobiology.

[15]  B. Scassellati.  Imitation and mechanisms of joint attention: a developmental structure for building social skills on a humanoid robot , 1999 .

[16]  S Ullman,et al.  Shifts in selective visual attention: towards the underlying neural circuitry. , 1985, Human neurobiology.

[17]  David J. Heeger,et al.  Optical flow using spatiotemporal filters , 2004, International Journal of Computer Vision.

[18]  L. Itti.  Author address: , 1999 .

[19]  Steven S. Beauchemin,et al.  The computation of optical flow , 1995, CSUR.

[20]  S. Petersen,et al.  The pulvinar and visual salience , 1992, Trends in Neurosciences.

[21]  V. Lakshminarayanan,et al.  Basic and Clinical Applications of Vision Science , 1997, Documenta Ophthalmologica Proceedings Series.

[22]  M. Goldberg,et al.  The representation of visual salience in monkey parietal cortex , 1998, Nature.

[23]  Minoru Asada,et al.  Developmental learning model for joint attention , 2002, IEEE/RSJ International Conference on Intelligent Robots and Systems.

[24]  Jeremy M. Wolfe,et al.  Guided Search 3.0 , 1997 .

[25]  John K. Tsotsos,et al.  Attending to visual motion , 2005, Comput. Vis. Image Underst..

[26]  F. Volkmar,et al.  Visual fixation patterns during viewing of naturalistic social situations as predictors of social competence in individuals with autism. , 2002, Archives of general psychiatry.

[27]  Wolfram Burgard,et al.  The Interactive Museum Tour-Guide Robot , 1998, AAAI/IAAI.

[28]  Illah R. Nourbakhsh,et al.  A survey of socially interactive robots , 2003, Robotics Auton. Syst..

[29]  C. Koch,et al.  Attention activates winner-take-all competition among visual filters , 1999, Nature Neuroscience.

[30]  Zhaoping Li.  A saliency map in primary visual cortex , 2002, Trends in Cognitive Sciences.

[31]  Anders Eriksson,et al.  Learning in Visual Attention , 2004, ICPR 2004.

[32]  J. Wolfe,et al.  Guided Search 2.0 A revised model of visual search , 1994, Psychonomic bulletin & review.

[33]  Joseph H. Goldberg,et al.  Identifying fixations and saccades in eye-tracking protocols , 2000, ETRA.

[34]  Iain D. Gilchrist,et al.  Visual correlates of fixation selection: effects of scale and time , 2005, Vision Research.

[35]  Stephanie Rosenthal,et al.  Designing robots for long-term social interaction , 2005, 2005 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[36]  Brian Scassellati,et al.  A Context-Dependent Attention System for a Social Robot , 1999, IJCAI.

[37]  John K. Tsotsos,et al.  Neurobiology of Attention , 2005 .

[38]  J. Wolfe,et al.  A Model of Visual Search Catches Up With Jay Enoch 40 Years Later , 1999 .

[39]  Heinz Hügli,et al.  Empirical Validation of the Saliency-based Model of Visual Attention , 2003 .

[40]  Rangachar Kasturi,et al.  Machine vision , 1995 .

[41]  Antonio Torralba,et al.  Modeling global scene factors in attention. , 2003, Journal of the Optical Society of America. A, Optics, image science, and vision.

[42]  Derrick J. Parkhurst,et al.  Modeling the role of salience in the allocation of overt visual attention , 2002, Vision Research.

[43]  T. Kanda,et al.  Robot mediated round table: Analysis of the effect of robot's gaze , 2002, Proceedings. 11th IEEE International Workshop on Robot and Human Interactive Communication.