Jinhui Yu

Automating Image Morphing Using Structural Similarity on a Halfway Domain

The main challenge in achieving good image morphs is to create a map that aligns corresponding image elements. Our aim is to help automate this often tedious task. We compute the map by optimizing the compatibility of corresponding warped image neighborhoods using an adaptation of structural similarity. The optimization is regularized by a thin-plate spline and may be guided by a few user-drawn points. We parameterize the map over a halfway domain and show that this representation offers many benefits. The map is able to treat the image pair symmetrically, model simple occlusions continuously, span partially overlapping images, and define extrapolated correspondences. Moreover, it enables direct evaluation of the morph in a pixel shader without mesh rasterization. We improve the morphs by optimizing quadratic motion paths and by seamlessly extending content beyond the image boundaries. We parallelize the algorithm on a GPU to achieve a responsive interface and demonstrate challenging morphs obtained with little effort.

Image-based synthesis of Chinese landscape painting

This paper describes a new framework for synthesizing Chinese landscape painting using an image-based approach. The framework involves two stages: a preprocessing phase, in which a few brush stroke texture primitivities (BSTP) are collected from samples of hand-made Chinese paintings, and the control picture is constructed to provide color IDs of mountains, and the on-line phases, in which the fog image is synthesized and mountains are “drawn” by mapping multiple layers of BSTP guided by the control picture. When more complex shading is needed, the shading picture is constructed and used during the BSTP mapping phase. Finally, the synthesized Chinese landscape paintings of a variety of styles are given and they look more close to the handmade work than those produced with previous modeling methods.

Restoration of Brick and Stone Relief from Single Rubbing Images

We present a two-level approach for height map estimation from single images, aiming at restoring brick and stone relief (BSR) from their rubbing images in a visually plausible manner. In our approach, the base relief of the low frequency component is estimated automatically with a partial differential equation (PDE)-based mesh deformation scheme. A few vertices near the central area of the object region are selected and assigned with heights estimated by an erosion-based contour map. These vertices together with object boundary vertices, boundary normals as well as the partial differential properties of the mesh are taken as constraints to deform the mesh by minimizing a least-squares error functional. The high frequency detail is estimated directly from rubbing images automatically or optionally with minimal interactive processing. The final height map for a restored BSR is obtained by blending height maps of the base relief and high frequency detail. We demonstrate that our method can not only successfully restore several BSR maps from their rubbing images, but also restore some relief-like surfaces from photographic images.

Generation of Kandinsky Art

ABSTRACT The authors present a programmed experiment to automatically generate art in the style of Kandinsky during his Bauhaus years. The program the authors developed analyzes the artist’s paintings based on his art theories and the authors’ own understanding and observations of his artworks. The authors describe the generation process in detail and share and discuss sample generated images styled according to four of Kandinsky’s paintings. By pseudorandomizing various parameters, the program is able to make each styled image it generates unique. The authors’ approach is highly scalable, limited only by the memory space set in the programming language Processing, which is used for the generation. Potential impacts of the authors’ approach are also discussed.

Outline font generating from images of ancient Chinese calligraphy

Chinese calligraphy is an art unique to Asian cultures. This paper presents a novel method for generating outline font from historical documents of Chinese calligraphy. The method consists of detecting feature points from character boundaries, and approximating contour segments. The feature-point-detection is based on statistical method considering the characteristics of a calligrapher. A database of basic strokes and some overlapping stroke components of Chinese characters extracted from the calligrapher are constructed in advance. And the relation between the noise level of stroke contours and the standard deviation of Gaussian kernel is retrieved from the database using linear regression. Thus, given an input character contour, the standard deviation for smoothing the noisy character contour can be calculated. Furthermore, a new method is employed to determine the feature points at the standard deviation. The feature points at a character contour subdivide the contour into segments. Each segment can be fitted by a parametric curve to obtain the outline font. Some experimental results and the comparisons to existing methods are also presented in the paper.

Capturing Character Contours from Images of Ancient Chinese Calligraphy

novel methodology is proposed to capture character contours from images of ancient Chinese calligraphy which mainly includes two steps: feature points detecting from character contours and contour segment approximating. A new feature-point-detection method called PCACSS (principle component analysis based curvature scale space) is thus proposed. Compared with several existing methods for feature-point-detection, it is robust to noise and can accurately detect feature points from character contours automatically. With feature points determined by PCACSS, the character contours are divided into some contour segments, each contour segment is then approximated with a straight line or Bezier curve depending on the least square error. The proposed method has been implemented and tested on a wide variety of calligraphy images. Resultant outline fonts captured from calligraphy images with our method are visual pleasing as demonstrated by the examples shown in this paper.

A Novel Method for Vectorizing Historical Documents of Chinese Calligraphy

We develop a novel method for feature point detection and employ it to generate outline font from historical document of Chinese calligraphy. The feature points at a character contour subdivide the contour into segments. Each segment can be then fitted by a parametric curve to obtain the outline font. Some experimental results are also presented in the paper.

Computational Aesthetic Evaluation of Logos

Computational aesthetics has become an active research field in recent years, but there have been few attempts in computational aesthetic evaluation of logos. In this article, we restrict our study on black-and-white logos, which are professionally designed for name-brand companies with similar properties, and apply perceptual models of standard design principles in computational aesthetic evaluation of logos. We define a group of metrics to evaluate some aspects in design principles such as balance, contrast, and harmony of logos. We also collect human ratings of balance, contrast, harmony, and aesthetics of 60 logos from 60 volunteers. Statistical linear regression models are trained on this database using a supervised machine-learning method. Experimental results show that our model-evaluated balance, contrast, and harmony have highly significant correlation of over 0.87 with human evaluations on the same dimensions. Finally, we regress human-evaluated aesthetics scores on model-evaluated balance, contrast, and harmony. The resulted regression model of aesthetics can predict human judgments on perceived aesthetics with a high correlation of 0.85. Our work provides a machine-learning-based reference framework for quantitative aesthetic evaluation of graphic design patterns and also the research of exploring the relationship between aesthetic perceptions of human and computational evaluation of design principles extracted from graphic designs.

Generating abstract paintings in Kandinsky style

This paper presents a recent project on automatic generation of Kandinsky style of abstract paintings using the programming language Processing. It first offers an analysis of Kandinskys paintings based on his art theories and the authors own understanding and observation. The generation process is described in details and sample generated images styled on four of Kandinskys paintings are also demonstrated and discussed. Our approach is highly scalable, limited only by the memory space set in Processing. Using random generation, every styled image generated can be unique. A selection of the images generated in the required resolution is also submitted and 70 images are made into a video companion.

Modeling ocean waves and interaction between objects and ocean water for cartoon animation

We present a scheme for the animation of ocean waves in a recognizable drawn animation cartoon style. This consists principally of two parts: the first part is the ocean surface generated by a procedural model which governs the dynamics of the ocean surface and rendered with stylized whitewater forms over the ocean wave crests; the second part is water effects caused by interactions between the ocean water and obstacle objects, such as waves crashing against obstacles‐like rocks or boats and forms surrounding objects, and those effects are simulated with different hierarchical models. The ocean surface can be used either alone or in conjunction with effects caused by the interactions according to the scene. With minimum user intervention, i.e., specification of a few parameters, our model is able to generate cartoon ocean wave animations using these procedural methods, as shown by examples given in the paper. Copyright