We introduce a fingertip touch recognition technique for general surfaces and a technique that estimates the direction of force on the basis of processed fingertip images captured by an overhead camera. When a finger presses on a hard surface, the distribution of color in the fingernail changes in accordance with the force applied. The first technique classifies the fingertip state (“press” or “non-press”) on the basis of the fingernail color information in the captured images. We use multiple EigenNail classifiers to achieve accurate classification of fingertip state under various lighting conditions. The second technique estimates the force direction by measuring the displacement of the fingertip. Through user studies, we demonstrated the feasibility of the proposed fingertip touch recognition techniques and were able to produce design guidelines for a pointing interface based on the proposed techniques.
In daily life, our behavior is guided by various visual stimuli such as the information on direction signs. However, our environmentally based perceptual capacity is often challenged in crowded circumstances, or more so, in emergency evacuation circumstances. In these situations, we often fail to pay attention to important signs. In order to achieve more effective direction guidance, we considered the use of unconscious reflexes in human walking action. In this study, we experimented with vision-guided walking direction control by optic flow stimulus combined with body sway. We observed a shift in subjects' walking direction and body sway and discuss the possible mechanism.
Visually impaired persons typically visualize graphical information through the tactile sensation of a pin matrix display. Unfortunately, the panel of a pin matrix display has a small gross area, the size of a postcard, so it is hard for several people to touch the panel and share the same information concurrently. Moreover, while the display is suitable for presenting a static figure, it is difficult to visualize a moving image with one's fingers. To solve these problems, we focused on the resolution of time in the sense of hearing, and conducted research on using sound from speakers arranged in a matrix to present both static and moving figures that can be shared by several people simultaneously. This paper describes an experiment we conducted to determine whether people could recognize the stroke order of the uppercase alphabet. Sound was broadcast from the various matrix speakers as if it were moving. Participants had a high average for correctly recognizing the alphabetic characters, demonstrating that people are able to recognize graphical information transmitted as moving sound.
Understanding, creating, and transmitting texts and figures is important in an information-based society. While systems that help visually impaired people understand, create, and transmit texts are available, ones to , help them create figures are still needed. We are developing a pilot system that enables them to draw tactile figures with their fingertips using a tactile display and computer vision techniques. Experimental evaluation of the usability of the system in comparison with that of a pen-input drawing system showed that is more effective?
We propose the a process concept of delayed tracking projection (DTP) for handheld projectors, in which (1) a pointer cursor sticks to the optical axis of the projector, (2) a user can moves the cursor in projected contents by moving the projector, and (3) the center of the contents tracks the cursor with a short delay. We combine the this DTP technique with a goal crossing interface. The aim objective of this paper study is to conduct an user experiment to understand determine how to design the an interface which that allows users to perform quick goal crossings at a low false selection rate. The Results of an experiment we conducted with a prototype system revealed the effects of delay, target distance, and target direction on goal crossing performances.
Motion graphics is a form of visual expression characterized by non-narrative, non-figurative based visuals that change over time. Due to the expansion of its application areas, a consideration of the affective quality of motion graphics is growing more important. This paper proposes an arousal estimation method that uses optical flow analysis as an affective quality assessment method for motion graphics. The primary objective is to verify two indexes: the total flow amount and the average magnitude of motion vectors. According to correlation analyses between each index and the arousal factor scores of video stimulus derived from an impression test with human participants, the correlation between the average magnitude of motion vectors and the arousal factor scores is significant. An analysis of the distribution of displacements showed that about three pixels per 33.3 msec is the border value: a higher ratio of movement slower than that border reduces the arousal level, and faster movement raises the level.
Eye fixation time distribution was obtained from eye fixation data in subjective selection test with paired food pictures by three opinions; “Like”, “Slightly like”, and “Neither like nor dislike”. And the diffusive and specific searches were discussed. Results show the frequency of eye fixation in the diffusive search is higher than that in the specific search, and the frequency of eye fixation for ‘selected’ pictures is higher than that for ‘not selected’ pictures in the specific search.
A super multi-view (SMV) 3D display has been proposed for reducing conflict between accommodation and vergence, which is a possible cause of visual fatigue. Previous studies reported that measured accommodative responses to SMV images were similar to those of real objects. We need to measure vergence response along with accommodative response to investigate the conflict between them. In this study, we measured accommodative and vergence responses while observers were viewing a static visual target presented on an SMV display. We also examined the responses to two-view displays and real objects for comparison. The results show that the SMV display induced a natural accommodative response relative to the two-view 3D display while no difference was found among the viewing conditions of the vergence response. This suggests that the SMV 3D display is effective for reducing conflict between accommodative and vergence responses.
It is known that excessive binocular disparity causes diplopia and weak impression of depth. However, recent studies have shown that a large and clear depth can be perceived for very large disparities when the targets move horizontally. In the present study, we compared the magnitudes of apparent depth obtained in dynamic conditions in which the targets oscillated horizontally and those obtained in static conditions in which the targets stayed stationary at eccentric locations. The results showed that the apparent depth was slightly larger when the eccentricity of the target was larger in static conditions and much larger depth was perceived in dynamic conditions than in static conditions. It appears that motion, not displacement, plays a major role in perceiving depth from very large disparities.
An auditory discrimination task is used to clarify the relationship between cognitive process and eyelid movement associated with eye closure. The timing of eyelid movement in response to a brief iterative brief sound stimulus is measured. Eyelid movement frequently occurs at a specific time after stimulus onset. This property is similar to that of eye blinking. In addition, an interesting tendency that the eyelid movement occurrence time is lower than that of eye blinking is obtained.
We have been developing a portable digital transmission system for next generation television called “Super Hi-Vision (SHV)” with hundreds of Mbps-class transmission and high link reliability. To fulfill these requirements, we have been researching a multiple-input multiple-output (MIMO) multiplexing technology using a millimeter-wave (42/55-GHz) band allocated for broadcasting organizations. In this paper, we propose a low-complexity MIMO detection scheme that uses block QR decomposition with M-algorithm to reduce the computational complexity of the maximum likelihood detection (MLD) method. In the proposed algorithm, a MIMO channel is converted into a block upper triangular matrix multiplied by a unitary matrix and divided into some small block matrices. Then, the transmit signals are selected by Manhattan metrics to reduce the complexity, and squared Euclidean metrics of the selected signal candidates are calculated in these block matrices. Computer simulations with a correlated 4 x 4 MIMO channel demonstrate that the proposed scheme shows almost the same BER performance as that of the MLD and better BER performance than that of the conventional QRM-MLD (QR decomposition with M-algorithm MLD) with less complexity.