The Adaboost (Freund and Schapire, Eur. Conf. Comput. Learn. Theory 23-37, 1995) chooses a good set of weak classifiers in rounds. On each round, it chooses the optimal classifier (optimal feature and its threshold value) by minimizing the weighted error of classification. It also reweights training data so that the next round would focus on data that are difficult to classify. When determining the optimal feature and its threshold value, a process of classification is employed. The involved process of classification usually performs a hard decision (Viola and Jones, Rapid object detection using a boosted cascade of simple features, 2001; Joo et al., Sci. World J 2014: 1-17, 2014 Friedman et al., Ann. Stat 28:337-407, 2000). In this paper, we extend the process of classification to a soft fuzzy decision. We believe this extension could allow some flexibility to the Adaboost algorithm as well as a good performance especially when the size of a training data set is not large enough. The Adaboost algorithm, in general, assigns a same weight to each training datum on the first round of a boosting process (Freund and Schapire, Eur. Conf. Comput. Learn. Theory 23-37, 1995). We propose to assign different initial weights based on some statistical properties of involved features. In experimental results, we show that the proposed method yields higher performances compared to other ones.
A virtual reality is a virtual space constructed by a computer that provides users the opportunity to indirectly experience a situation they have not experienced in real life through the realization of information for virtual environments. Various studies have been conducted to realize virtual reality, in which the user interface is a major factor in maximizing the sense of immersion and usability. However, most existing methods have disadvantages, such as costliness or being limited to the physical activity of the user due to the use of special devices attached to the user's body. This paper proposes a new type of interface that enables the user to apply their intentions and actions to the virtual space directly without special devices, and test content is introduced using the new system. Users can interact with the virtual space by throwing an object in the space; to do this, moving object detectors are produced using infrared sensors. In addition, the users can control the virtual space with their own postures. The method can heighten interest and concentration, increasing the sense of reality and immersion and maximizing user's physical experiences.
Interactive display of complex scenes is a challenging problem in computer graphics. Such current approaches as z-buffer, level of detail and visibility culling have not fully used the temporal coherence between consecutive frames. When the viewing condition is fixed, the color and depth values of static polygons can be obtained from the result of the previous frame and only the remaining dynamic polygons require rendering. We present a method that enhances the speed of the conventional z-buffer algorithm by exploiting the above temporal coherence. This algorithm is simple to combine with existing graphics hardware that supports the conventional z-buffer algorithm. It can manipulate any scene suitable for the z-buffer algorithm without preprocessing or human intervention. The rendering time is proportional to the number of dynamic polygons in each frame. Experimental results show that our method is faster than the conventional z-buffer algorithm and the performance enhancement becomes higher as the fraction of static polygons increases.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.