Abstract Automatic or semi‐automatic colorization can reduce the burden of illustrators in color illustration production, which is a research area with significant market demand. Texture details in eyes and hair influence the impression of character illustrations. Generally, these details are not expressed in line drawings. Many existing automatic or semi‐automatic colorization methods do not target hand‐drawn line drawings and it is difficult to paint texture details on such drawings. In this paper, we propose the semi‐automatic colorization of character line drawings around faces with texture details. Our method uses a reference image as a color hint and transfers the textures of the reference image to a line drawing. To achieve this, our method uses semantic segmentation masks to match parts of the line drawing with the same parts of the reference image. We create two types of segmentation datasets to train a segmentation network that creates segmentation masks. We transfer texture details to a hand‐drawn line drawing by mapping each part of the reference image to the corresponding part of the line drawing using segmentation masks. We show that our method is more effective for hand‐drawn line drawings than existing methods using qualitative and quantitative evaluations.

Abstract Our aim is to convert an object's appearance to an arbitrary colour considering the light scattering in the entire scene, which is often called the global illumination. Existing stylisation methods convert the colour of an object with a 1‐dimensional texture for 3‐dimensional computer graphics to reproduce a typical style used in illustrations and cel animations. However, they cannot express global illumination effects. We propose two individual methods to compute the global illumination and convert the shading to an arbitrary colour. The methods reproduce reflections in other objects with the converted colour. As a result, we can convert the colour of illumination effects that have not yet been reproduced, such as soft shadows and refractions.

Abstract Many studies have recently applied deep learning to the automatic colorization of line drawings. However, it is difficult to paint empty pupils using existing methods because the convolutional neural network are trained with pupils that have edges, which are generated from color images using image processing. Most actual line drawings have empty pupils that artists must paint in. In this paper, we propose a novel network model that transfers the pupil details in a reference color image to input line drawings with empty pupils. We also propose a method for accurately and automatically colorizing eyes. In this method, eye patches are extracted from a reference color image and automatically added to an input line drawing as color hints using our pupil position estimation network.

This paper presents a particle-based model for preserving fluid sheets of animated liquids with an adaptively sampled Fluid-Implicit-Particle (FLIP) method. In our method, we preserve fluid sheets by filling the breaking sheets with particle splitting in the thin regions, and by collapsing them in the deep water. To identify the critically thin parts, we compute the anisotropy of the particle neighborhoods, and use this information as a resampling criterion to reconstruct thin liquid surfaces. Unlike previous approaches, our method does not suffer from diffusive surfaces or complex remeshing operations, and robustly handles topology changes with the use of a meshless representation. We extend the underlying FLIP model with an anisotropic position correction to improve the particle spacing, and adaptive sampling to efficiently perform simulations of larger volumes. Due to the Lagrangian nature of our method, it can be easily implemented and efficiently parallelized. The results show that our method can produce visually complex liquid animations with thin structures and vivid motions.

We present a simple extension to PIC/FLIP(Particle-in-Cell/Fluid-Implicit-Particle) for animating liquid with enhanced behaviors such as pushing or eddying (Figure 1), which we call HFA/PIC/FLIP (High-frequency aware PIC/FLIP). As a fundamental approach we use PIC/FLIP [Brackbill and Ruppel 1986] and compute approximate low-frequency part and high-frequency part of particles velocities. Low-frequency velocities are entirely projected onto divergence free and high-frequency field is partially projected onto divergence free to achieve realistic liquid animations. In contrast to the PIC/FLIP proposed by Zhu and Bridson [Zhu and Bridson 2005]; our approach facilitates "pushing" or "curly" features whereas their approach disperses the momentum towards noisy directions, resulting in dispersion. Recently similar approach has been done with coarse to fine mesh grid [Lentine et al. 2010], we refined directly with particles. We produced several footage of liquid animation and compared with competitive alternatives to show the benefits of the proposed algorithm.

We present a new approach for rendering fluid with a style of vector graphics. The proposed method of vector art rendering of fluid exhibits a curly, beautiful and clear silhouette which deserves artistic expression. (See Figure 1) In computer vision, fluid has been mainly focused on capturing realistic behavior. Thanks to the much contribution of seminal works, large knowledge of photorealistic rendering of fluid have been amassed. On the other hand, although some approaches of the artistic rendering of fluid proposed [Selle et al. 2004; Eden et al. 2007], it seems that the area is not researched intensively and this kind of rendering technique still remains untouched. In order to achieve vector art rendering, we have to track the target region in fluid. However, popular methods of tracking interface model fluid with density field or collection of particles which introduce sudden topological changes and it is challenging to track them precisely. Instead, we consider the velocity field is infinitely continuous so that none of the part of fluid contour intersects each other at any state. Even though such consideration is questionable in physically derived model, this concept of non-intersection of contour leads a very simple algorithm and exhibits nice results. In this paper we first briefly describe the algorithm for vector fluid rendering and secondly we show some artworks created by our method to demonstrate the potential of our algorithm.

This paper describes a method for simulating and visualizing dyeing based on weave patterns and the physical parameters of the threads and the dye. We apply Fick's second law with a variable diffusion coefficient. We calculate the diffusion coefficient using the porosity, tortuosity, and the dye concentration based on the physical chemistry of dyeing. The tortuosity of the channel was incorporated in order to consider the effect of the weave patterns on diffusion. In this model, the total mass is conserved. We describe the cloth model using a two-layered cellular model that includes the essential factors required for representing the weft and warp. Our model also includes a simple dyeing technique that produces dyeing patterns by interrupting the diffusion of the dye in a cloth using a press. The results obtained using our model demonstrate that it is capable of modeling many of the characteristics of dyeing.

This paper presents an algorithm for generating realistic drawing strokes that can take on the appearance of pastels, charcoals, or crayons. The similarity between pigments deposited on a paper surface by a pastel stroke and on the texture of an illuminated paper surface was studied. Twelve paper textures were prepared and were illuminated from various directions in increments of 30 degrees for a stroke drawn in an arbitrary direction. After these textures were processed as if they could be used as a height field, pigments deposited on paper were calculated using the height field and pen pressure. The height field is determined by linear interpolation of multiple paper textures for drawing strokes in any direction. By using the paper texture instead of simulating the deposition of pigments, a realistic stroke with arbitrary parameters can be rendered in real time.