Search

In this paper, we proposed a serious game that not only for entertainment but also encourage people to do more exercise, in order to let them relax and refresh from the stressful working life and maintain a good health and well-being. Our game applied motion capture, muscle power sensing, action game factors and virtual reality technology (VR). In contrast to traditional VR games, our game provides a novel muscle-power sensitive gaming interface, such that special effects will be shown and a higher bonus scores could be received when the player do exercise harder and exert a greater muscle power. The preliminary experiment result reveals that the muscle sensing capability facilitates the players to pay more attention to their muscle activities when playing the game. The players also agreed that the proposed game could attract them to do exercise again.

Color and texture are basic elements in digital graphics. Selection of color with a picker is convenient in many of the image editing softwares. However, more organized and intelligent GUI for texture pattern selection is still missing. In this paper, we attempt to fill this gap with the introduction of several robust techniques in building an intuitive texture picking GUI. By arranging patterns according to their visual similarities, texture picker with plane and circular layout are presented. Additional functionality include content-based texture searching which can quickly find similar patterns of given sample. Preliminary response to the proposed interface is positive in general, while further improvements are required, for example, on building a hierarchy to facilitate high to low level selection for huge amount of texture patterns.

Due to the tediousness and labor intensive cost, some manga artists have already employed computer-assisted methods for converting color photographs to manga backgrounds. However, existing bitonal image generation methods usually produce unsatisfactory uniform screening results that are not consistent with traditional mangas, in which the artist employs a rich set of screens. In this paper, we propose a novel method for generating bitonal manga backgrounds from color photographs. Our goal is to preserve the visual richness in the original photograph by utilizing not only screen density, but also the variety of screen patterns. To achieve the goal, we select screens for different regions in order to preserve the tone similarity, texture similarity, and chromaticity distinguishability. The multi-dimensional scaling technique is employed in such a color-to-pattern matching for maintaining pattern dissimilarity of the screens. Users can control the mapping by a few parameters and interactively fine-tune the result. Several results are presented to demonstrate the effectiveness and convenience of the proposed method.