No abstract

No abstract

Describing certain types of spatial relationships between a pair of objects requires that the objects are assigned different “roles” in the relation, e.g., “A is above B” is different than “B is above A.” This asymmetric representation places one object in the “target” or “figure” role and the other in the “reference” or “ground” role. Here we provide evidence that this asymmetry may be present not just in spatial language, but also in perceptual representations. More specifically, we describe a model of visual spatial relationship judgment where the designation of the target object within such a spatial relationship is guided by the location of the “spotlight” of attention. To demonstrate the existence of this perceptual asymmetry, we cued attention to one object within a pair by briefly previewing it, and showed that participants were faster to verify the depicted relation when that object was the linguistic target. Experiment 1 demonstrated this effect for left-right relations, and Experiment 2 for above-below relations. These results join several other types of demonstrations in suggesting that perceptual representations of some spatial relations may be asymmetrically coded, and further suggest that the location of selective attention may serve as the mechanism that guides this asymmetry.

No abstract

The CCITT (Consultative Committee for International Telegraph and Telephone) has recently finalized a set of five Recommendations (H.261, H.221, H.242, H.230, and H.320) which will trigger a period of incredible growth for Video Teleconference (VTC) and Videophone (VP) services. These Recommendations are commonly referred to as P x 64 since they define systems operating at multiples of 64 Kbps.Although P x 64 systems manufactured by different vendors will be interoperable, they will not provide the same picture quality. For good reason, the P x 64 Recommendations do not specify system functions such as preprocessing, postprocessing, and encoding strategy, all of which significantly affect picture quality. It is therefore necessary to develop measures of picture quality for VTC/VP systems to insure a specifiable quality of service can be delivered to the user community. For static scenes, conventional measures used in the broadcast industry (e.g. spatial frequency response and S/N) provide good objective measures. However, VTC/VP systems exhibit visible distortion of moving scenes to a much greater degree than that experienced in broadcast TV. Therefore, it is necessary to develop new techniques to objectively measure the motion distortion in VTC/VP systems. This paper describes three techniques to measure this distortion: temporal frequency response, scene cut response, and transmitted frame rate. Results of tests performed on representative VTC systems are presented.