Purchase this article with an account.
Takahiro Kawabe, Kazushi Maruya, Shin'ya Nishida; What do human observers see in dynamic image deformation?. Journal of Vision 2014;14(10):261. doi: 10.1167/14.10.261.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Image information of an object often deforms for various reasons. For example, a non-rigid object such as a rubber ball or a fabric can deform due to an external force and, as a consequence, produce an image deformation. On the other hand, image deformation can occur without the physical deformation of objects themselves. For example, an under-liquid scene is distorted due to refraction at the surface of the flowing liquid. In this case, a refractive transparent material intervening between the object and an observer causes image deformation of the underlying object. How does the visual system discriminate image deformation due to physical object deformation from image deformation due to the intervention of a transparent layer? The present study examined what human observers see during dynamic image deformation. We sinusoidally deformed the central area of two-dimensional 1/f noise. The spatial and drift frequencies of deformation were manipulated. Observers were asked to report the appearance of the deformed central area using three alternatives ('global object deformation', 'drifting grating', and 'transparent layer'), which were derived from a preliminary observation. Object deformation was often reported when the spatial frequency of deformation was low. Meanwhile, a transparent layer was predominantly reported as the spatiotemporal frequencies of deformation increased (i.e., dynamic transparency). Importantly, however, dynamic transparency was not observed when one-dimensional noise was employed as a background. In one-dimensional noise, image deformation occurred along an edge with consistent luminance polarity, and consequently enhanced perception of a global shape deformation. The visual system likely utilizes spatiotemporal frequencies of deformation and global shape as diagnostic features to discriminate image deformation due to physical object deformation from image deformation due to the intervention of a transparent material.
Meeting abstract presented at VSS 2014
This PDF is available to Subscribers Only