Petaloid smocking is made of cloth by sewing and flattening pleats to create shapes over the surface. For dynamically visualizing the folding of pleats, this paper proposes an approach for creating crease patterns based on the rotation of shapes and pleat length. Dynamic visualization of petaloid smocking will contribute to avoiding the situation where the designed petaloid smocking cannot be actually produced due to the overlap of designed patterns, and to facilitating the shaping of flexible cloth. We show geometric characteristics of the crease patterns based on the proposed approach, and suggest a design framework with several examples of crease pattern. By designing a crease pattern with GeoGebra, the proposed approach is validated through the application to fashion goods. The shape of the produced bag can be transformed through folding, and fastened with magic tapes to enable variable size.

Recently, in addition to games, interaction with real virtual agents has been used for job training. Furthermore, multimodal dialog using both voice and gesture recognition is used. However, there is a challenge with conventional batch processing speech recognition: it is difficult to make the virtual agent respond while the user is speaking. Therefore, in this study, we confirm the possibility of enhancing social presence by generating responses using real-time speech recognition in spoken dialog with characters. While the user speaks, by extracting keywords from the speech content in real-time, we generate the emotional response of the virtual agent. The experimental results demonstrate that our approach improves social presence compared with conventional spoken dialog.

This paper introduces a novel color transformation method for recoloring an image to colors with a specific theme. To achieve such recoloring, we often need to adjust all colors in an image to keep the natural appearance of the original image. This task, however, is difficult and we usually determine the colors by tedious trial-and-error processes. We address this problem and propose an optimization-based approach that takes into account both the user-specified color theme and the color distribution of the original image. Users first specify a color of some regions in an input image, and then our system automatically adjusts the colors of the rest of the regions to keep the appearance of the input image. This automatic adjustment is formulated as a minimization problem. We accelerate the computation by solving the problem using a quantized color palette. In addition, we extend the method for a video. The color palette is built from the colors on some frames of the video since the calculation of all colors on all frames would be too much cost. We show several examples to demonstrate the effectiveness of our method.

Improvisation is a form of music in which the content of the performance is not specified in advance, but whether improvised or not, musical performance is considered to be dominated by feedforward control. In other words, in the process of general improvisation, phrases are selected from one’s own past input, suggesting that it is rare that a new phrase is created on the spot that deviates from one’s own musical experience. In this study, to support improvisation, we conducted an experiment in which an electric guitarist improvised on a consistent time line along the beginning and ending of a picture book, and the picture book, text, and recited voice brought about changes in the phrases. As a result of the experiment, the presence of the picture book, the text, and the reading voice was able to produce phrases other than the ones the performer normally uses.

Image colorization is a hot topic in the research of image processing and generation techniques. It plays an important role in restoring old photos, colorization medical images, and colorization anime manuscripts. In recent years, with the rapid development of deep learning, some colorization methods based on deep learning have been proposed, and satisfactory colorization results have been achieved. However, these colorization methods generally use unguided feature learning to achieve color prediction, it cannot predict, select, and intervene in image colors based on the subjective ideas of the creator. At the same time, the generated colorization results may have unreasonable phenomena such as color concentration, saturation and brightness distribution. Therefore, we propose a cross-domain color mapping from exemplar anime image colorization method, which is used for coloring a given anime image with a reference style. This method involves a supervised colorization network, which is divided into two structures: a domain semantic matching sub-network and a colorization sub-network. The domain semantic matching sub-network maps the pixels of the input grayscale image domain and the reference color image domain to an intermediate domain by calculating the similarity of the correlation matrix, thereby establishing a corresponding semantic relation and obtaining a coarse colorization map. We also employed an end-to-end CNN encoder and decoder network to further extract the feature information of the matched image colors, and used the Lab color space to select, propagate, and predict the color distribution features of the image. Experimental results have demonstrated that the anime coloring images generated by our network have high restoration similarity compared to the original reference images, reasonable coloration in the coloring area, and this colorization method solves the problem of prior color style transfer in anime images.

Numerous investigations have been conducted on the topic of color combinations, with a focus on generating color palettes to depict the patterns inherent in such combinations. These color palettes have proven to be immensely valuable in the realm of visual design. Our research also concerns color combinations; however, our approach involves the utilization of algorithms to demarcate visual center and non-visual center regions within game concept art. By adopting an artist's perspective, this research delves into the comprehension of color combinations. In light of our prior investigations, we summarize the colors associated with the visual and non-visual centers, using color segmentation, and subsequently compute the respective color proportions in these two areas for conclusive purposes. Furthermore, we ascertain the hue, saturation, and lightness characteristics of each color, thereby discerning the underlying laws of color combinations within the game concept art. The objective of this research paper is to provide digital artists with the means to produce superior and expeditious artworks that align with contemporary aesthetics.

In this paper, we use 3DCG to restore the old building of the National Museum of Modern Art (currently the National Museum of Modern Art, Tokyo), which opened in Kyobashi, Chuo-ku, Tokyo in 1952 as Japan's first national art museum. I would like to report on the creation of Exhibition reproduction 3D walkthrough (exhibition name "Reproduction VR") of the exhibition "Abstract and Illusion: How to Understand Non-Realistic Paintings." We recreated the exhibition room of the old building based on highly accurate historical research, and installed the exhibition based on digital images and records of exhibition photographs (glass dry plates) stored in the art library. Then, we created "Reproduction VR" for this exhibition. This is a highly academic immersive digital content experience.

From October 12, 2022, ～ February 5, 2023, "Reproduction VR" was exhibited in the collection exhibition "Playback "Abstract and Illusion" (1953-1954)" as the first attempt at the National Museum of Modern Art, Tokyo. It attracted attention as one of the methods of researching exhibitions and materials and was well received as a VR that could be enjoyed like a video game.