A unified scheme for adaptive stroke-based rendering

This paper presents a comprehensive scheme for automatically generating a broad class of artistic illustrations from photographs. Using strokes as the major building blocks, our system optimizes the stroke attributes subject to the desired rendering style. The stroke attributes are computed adaptively to enable importance-based control of the abstraction level at each pixel. We propose a novel outline detection and refinement paradigm called edge painting to construct an outline map, and from which to derive the pixel-wise importance. We also introduce an adaptive bilateral filter to adaptively guide the curved stroke directions based on the importance map. Given the outline, importance, and direction maps, the system creates the illustration via selecting the representative colors, setting the style parameters, and optimizing the stroke attributes based on simulated annealing. The experimental results show that our scheme facilitates automatic production of artistic illustrations in a wide range of rendering styles.     

Multiresolution modeling techniques in CAD

This paper investigates new wavelet-based multiresolution modeling techniques in CAD. Models in the system are represented by endpoint-interpolating B-spline functions. Wavelet deformation techniques are extended to include synthesis editing and detail blending from previous sweep editing and fractional editing methods. Application of multiresolution editing to multiple patches of complex topology is investigated. An algorithm for multi-patch deformation is developed by enforcement of G⁰ continuity between adjacent patches. After editing G¹ continuity may be recaptured by numerical optimization of the error vectors along boundary seams. A simple CAD interface is implemented for B-spline patch acquisition based on reverse engineered data and layer based modeling and a moving cursor plane is used for three-dimensional (3D) editing. Design examples are presented to illustrate the 3D wavelet editing capability. The developed wavelet deformation techniques can be a useful complement to current CAD systems by providing powerful new modeling and editing capabilities.     

Pencil-like sketch rendering of 3D scenes using trajectory planning and dynamic tracking

Objective: We present a new non-photorealistic rendering method to render 3D scenes in the form of pencil-like sketches.Methods: This work is based on the observation that the dynamic feedback mechanism involving the human visual system and the motor control of the hand collectively generates the visual characteristics unique to hand-drawn sketches. At the heart of our approach is a trajectory planning and tracking algorithm that generate the sketch in multiple layers using a dynamic pen model. On each layer, a set of target strokes are generated from the silhouette lines, edges, and shaded regions which serve as the target trajectory for a closed-loop dynamic pen model. The pen model then produces the rendered sketch, whose characteristics can be adjusted with a set of trajectory and tracking parameters. This process continues in several layers until the tonal difference between the sketch and the original 3D render is minimized.Results: We demonstrate our approach with examples that are created by controlling the parameters of our sketch rendering algorithms.Conclusion: The examples not only show typical sketching artifacts that are common to human-drawn sketches but also demonstrate that it is capable of producing multiple sketching styles.     

Object's contour detection scheme based on image fusion

Three novel object's contour detection schemes based on image fusion are proposed in this paper. In these schemes an active contour model is applied to detect the object's contour edge. Since an object's contour in an infrared (IR) image is usually clearer than that in a visible image, the convergent active contour in a visible image is improved with that in an IR image. The first contour detection scheme is realized by revising the shape-preserving active contour model. The second scheme minimizes the B-spline L ₂ norm's square of the difference of the B-spline control point vectors in two modal images. Contour tracking and extraction experiments indicate that the first scheme outperforms the second one. Moreover, a third scheme based on the active contour and pixel-level image fusion is proposed for images with incomplete but complementary scene information. An example using contour extraction of a partially hidden tank proves its efficacy.     

基于信息熵的风格绘画分类研究

针对艺术风格绘画分类算法中存在的精度和效率不高等问题,提出一种基于信息 熵的艺术风格绘画分类算法。首先选取西方漫画、素描、油画、水彩画,以及国内烙画、水墨 画、壁画具有代表性的 7 种艺术绘画风格作为研究对象,对图像进行去噪、归一化等预处理。 其次,提取绘画艺术作品风格特征,分别求取图像的颜色熵、分块熵、轮廓熵,并合并构成不 同绘画风格的信息熵。信息熵求取时,将色彩空间转换为 Lab 颜色空间,通过 a、b 通道颜色值 及加权函数获得图像的颜色熵;通过对艺术图像分块求取信息熵,求取分块的信息熵均值获得 分块熵;通过 Contourlet 变换,求取艺术图像的轮廓信息,获得轮廓熵。接着,合并提取的颜 色熵、分块熵、轮廓熵,利用支持向量机(SVM)对艺术风格图像学习训练,获得艺术绘画风格 的分类模型;最后,提取待识别绘画风格样本的熵特征,通过 SVM 分类识别获得最终的分类 结果。该方法具有特征维数少、运算速度快、尺度不变性等优势,实验结果表明,其能提高不 同绘画风格的分类精度和效率。     

Lit-Sphere extension for artistic rendering

The Lit-Sphere model proposed by Sloan et al. (Proceedings of Graphics Interface 2001, pp. 143–150, 2001) is a method for emulating expressive artistic shading styles for 3D scenes. Assuming that artistic shading styles are described by view space normals, this model produces a variety of stylized shading scenes beyond traditional 3D lighting control. However, it is limited to the static lighting case: the shading effect is only dependent on the camera view. In addition, it cannot support small-scale brush stroke styles. In this paper, we propose a scheme to extend the Lit-Sphere model based on light space normals rather than view space normals. Owing to the light space representation, our shading model addresses the issues of the original Lit-Sphere approach, and allows artists to use a light source to obtain dynamic diffuse and specular shading. Then the shading appearance can be refined using stylization effects including highlight shape control, sub-lighting effects, and brush stroke styles. Our algorithms are easy to implement on GPU, so that our system allows interactive shading design.     

Adversarial Interactive Cartoon Sketch Colourization with Texture Constraint and Auxiliary Auto-Encoder

Colouring cartoon sketches can help children develop their intellect and inspire their artistic creativity. Unlike photo colourization or anime line art colourization, cartoon sketch colourization is challenging due to the scarcity of texture information and the irregularity of the line structure, which is mainly reflected in the phenomenon of colour-bleeding artifacts in generated images. We propose a colourization approach for cartoon sketches, which takes both sketches and colour hints as inputs to produce impressive images. To solve the problem of colour-bleeding artifacts, we propose a multi-discriminator colourization framework that introduces a texture discriminator in the conditional generative adversarial network (cGAN). Then we combined this framework with a pre-trained auxiliary auto-encoder, where an auxiliary feature loss is designed to further improve colour quality, and a condition input is introduced to increase the generalization ability over hand-drawn sketches. We present both quantitative and qualitative evaluations, which prove the effectiveness of our proposed method. We test our method on sketches of varying complexity and structure, then build an interactive programme based on our model for user study. Experimental results demonstrate that the method generates natural and consistent colour images in real time from sketches drawn by non-professionals.     

Drawing for Illustration and Annotation in 3D

We present a system for sketching in 3D, which strives to preserve the degree of expression, imagination, and simplicity of use achieved by 2D drawing. Our system directly uses user-drawn strokes to infer the sketches representing the same scene from different viewpoints, rather than attempting to reconstruct a 3D model. This is achieved by interpreting strokes as indications of a local surface silhouette or contour. Strokes thus deform and disappear progressively as we move away from the original viewpoint. They may be occluded by objects indicated by other strokes, or, in contrast, be drawn above such objects. The user draws on a plane which can be positioned explicitly or relative to other objects or strokes in the sketch. Our system is interactive, since we use fast algorithms and graphics hardware for rendering. We present applications to education, design, architecture and fashion, where 3D sketches can be used alone or as an annotation of an existing 3D model.     

Control point adjustment for B-spline curve approximation

Pottmann et al. propose an iterative optimization scheme for approximating a target curve with a B-spline curve based on square distance minimization, or SDM. The main advantage of SDM is that it does not need a parameterization of data points on the target curve. Starting with an initial B-spline curve, this scheme makes an active B-spline curve converge faster towards the target curve and produces a better approximating B-spline curve than existing methods relying on data point parameterization. However, SDM is sensitive to the initial B-spline curve due to its local nature of optimization. To address this, we integrate SDM with procedures for automatically adjusting both the number and locations of the control points of the active spline curve. This leads to a method that is more robust and applicable than SDM used alone. Furthermore, it is observed that the most time consuming part of SDM is the repeated computation of the foot-point on the target curve of a sample point on the active B-spline curve. In our implementation, we speed up the foot-point computation by pre-computing the distance field of the target curve using the Fast Marching Method. Experimental examples are presented to demonstrate the effectiveness of our method. Problems for further research are discussed.     

Automatic Anterior Lamina Cribrosa Surface Depth Measurement Based on Active Contour and Energy Constraint

The lamina cribrosa is affected by intraocular pressure, which is the major risk of glaucoma. However, the capability to evaluate the lamina cribrosa in vivo has been limited until recently due to poor image quality and the posterior laminar displacement of glaucomatous eyes. In this study, we propose an automatic method to measure the anterior lamina cribrosa surface depth (ALCSD), including a method for detecting Bruch’s membrane opening (BMO) based on k-means and region-based active contour. An anterior lamina cribrosa surface segmentation method based on energy constraint is also proposed. In BMO detection, we initialize the Chan-Vese active contour model by using the segmentation map of the k-means cluster. In the segmentation of anterior lamina cribrosa surface, we utilize the energy function in each A-scan to establish a set of candidates. The points in the set that fail to meet the constraints are removed. Finally, we use the B-spline fitting method to obtain the results. The proposed automatic method can model the posterior laminar displacement by measuring the ALCSD. This method achieves a mean error of 45.34 μm in BMO detection. The mean errors of the anterior lamina cribrosa surface are 94.1% within five pixels and 76.1% within three pixels.     

Regularization of B-Spline Objects

By a d-dimensional B-spline object (denoted as Od), we mean a B-spline curve (d=1), a B-spline surface (d=2) or a B-spline volume (d=3). By regularization of a B-spline object Od we mean the process of relocating the control points of Od such that it approximates an isometric map of its definition domain in certain directions and is shape preserving. In this paper we develop an efficient regularization method for Od, d=1,2,3, based on solving weak form L²-gradient flows constructed from the minimization of certain regularizing energy functionals. These flows are integrated via the finite element method using B-spline basis functions. Our experimental results demonstrate that our new regularization methods are very effective.     

Approximate merging of B-spline curves via knot adjustment and constrained optimization

This paper addresses the problem of approximate merging of two adjacent B-spline curves into one B-spline curve. The basic idea of the approach is to find the conditions for precise merging of two B-spline curves, and perturb the control points of the curves by constrained optimization subject to satisfying these conditions. To obtain a merged curve without superfluous knots, we present a new knot adjustment algorithm for adjusting the end k knots of a kth order B-spline curve without changing its shape. The more general problem of merging curves to pass through some target points is also discussed.     

结合图元与感知哈希的手写输入简笔画识别

目的 为了克服手写输入中随意性强和自由度大的缺陷,同时兼顾简笔画的整体属性和局部特征,提出一种基于图元识别与感知哈希技术相结合的手写输入简笔画二级识别算法。方法 首先提取笔画的几何特征、笔序特征及结构特征且进行识别,然后查找由图元信息、笔画结构信息和笔序信息构成的简笔画语义库,完成由规则的几何图元构成的简笔画识别;若未被识别,则生成简笔画图像,利用感知哈希技术完成简笔画图像的识别。结果 基于本文提出的简笔画识别方法,实现了对样本库中150种简笔画对象的识别,平均识别率为82.6%。结论 实验结果表明,对于不同用户手写输入的任意样本库中的简笔画,该方法具有较高的识别率,此外,还可以通过在简笔画语义库和样本库中增加简笔画的种类等方式实现对更多种类简笔画的扩展识别。     

Wavelet-Based Multiresolution with

Multiresolution methods are a common technique used for dealing with large-scale data and representing it at multiple levels of detail. We present a multiresolution hierarchy construction based on subdivision, which has all the advantages of a regular data organization scheme while reducing the drawback of coarse granularity. The -subdivision scheme only doubles the number of vertices in each subdivision step regardless of dimension n. We describe the construction of 2D, 3D, and 4D hierarchies representing surfaces, volume data, and time-varying volume data, respectively. The 4D approach supports spatial and temporal scalability. For high-quality data approximation on each level of detail, we use downsampling filters based on n-variate B-spline wavelets. We present a B-spline wavelet lifting scheme for -subdivision steps to obtain small or narrow filters. Narrow filters support adaptive refinement and out-of-core data exploration techniques.     

Discriminative Sketch‐based 3D Model Retrieval via Robust Shape Matching

We propose a sketch‐based 3D shape retrieval system that is substantially more discriminative and robust than existing systems, especially for complex models. The power of our system comes from a combination of a contour‐based 2D shape representation and a robust sampling‐based shape matching scheme. They are defined over discriminative local features and applicable for partial sketches; robust to noise and distortions in hand drawings; and consistent when strokes are added progressively. Our robust shape matching, however, requires dense sampling and registration and incurs a high computational cost. We thus devise critical acceleration methods to achieve interactive performance: precomputing kNN graphs that record transformations between neighboring contour images and enable fast online shape alignment; pruning sampling and shape registration strategically and hierarchically; and parallelizing shape matching on multi‐core platforms or GPUs. We demonstrate the effectiveness of our system through various experiments, comparisons, and user studies.     

Improving signal prediction performance of neural networks through multiresolution learning approach.

In this paper, we extend the original work on multiresolution learning for neural networks, and present new developments on the multiresolution learning paradigm. The contributions of this paper include: (1) proposing a new concept and method of adjustable neural activation functions in multiresolution learning to improve neural network learning efficacy and generalization performance for signal predictions; (2) providing new insightful explanations for the multiresolution learning paradigm from a multiresolution optimization perspective; (3) exploring underlying ideas why the multiresolution learning scheme associated with adjustable activation functions would be more appropriate for the multiresolution learning paradigm; and (4) providing rigorous validations to evaluate the multiresolution learning paradigm with adjustable activation functions and comparing it with the schemes of multiresolution learning with fixed activation functions and traditional learning. This paper presents systematically new analytical and experimental results on the multiresolution learning approach for training an individual neural network model, demonstrates our integral solution on neural network learning efficacy, and illustrates the significant improvements on neural networks' generalization performance and robustness for nonlinear signal predictions.     

Enhanced lane: interactive image segmentation by incremental path map construction

Live-wire type techniques for interactive image segmentation are of practical use for various applications such as medical image analysis, digital image composition, etc. Intelligent scissors [5] and live wire [8] are the representative techniques of this type, which are based on a graph search over an entire image. Another technique called live lane [8] is also based on a graph search but localizes the search domain to give an interactive feedback. Compared to the live wire, the live lane trades off the repeatability of segmentation for its time efficiency. In this paper, we present a novel image segmentation technique called enhanced lane, a modified version of the live lane that ensures both efficiency and repeatability. By restricting the search domain and updating the path map incrementally, the enhanced lane can extract objects from an image interactively with its efficiency comparable to that of the live lane while also keeping its repeatability comparable to that of the live wire. Based on the live lane paradigm, our technique also differs from the time-efficient version of live wire called live wire on the fly [9]: the enhanced lane always guarantees strictly bounded response time regardless of the image size and follows the target boundary with little digression which leads to better repeatability.     

VARIANT: A System for Terrain Modeling at Variable Resolution

We describe VARIANT (VAriable Resolution Interactive ANalysis of Terrain), an extensible system for processing and visualizing terrains represented through Triangulated Irregular Networks (TINs), featuring the accuracy of the representation, possibly variable over the terrain domain, as a further parameter in computation.VARIANT is based on a multiresolution terrain model, which we developed in our earlier research. Its architecture is made of a kernel, which provides primitive operations for building and querying the multiresolution model; and of application programs, which access a terrain model based on the primitives in the kernel.VARIANT directly supports basic queries (e.g., windowing, buffering, computation of elevation at a given point, or along a given line) as well as high-level operations (e.g., fly-over visualization, contour map extraction, viewshed analysis). However, the true power of VARIANT lies in the possibility of extending it with new applications that can exploit its multiresolution features in a transparent way.     

基于多分辨率方法的主动轮廓线跟踪算法

由Ｋａｓｓ等１９８７年首次提出的主动次轮廓线模型，在数字图像分析和计算机视觉领域得到了越来越广泛的应用，基于单分辨率的主动轮廓线跟踪算法在一个共同的缺点，即：要求初始轮廓线离目标的真实轮廓线很近，这样的算法用于跟踪将只能跟踪缓慢运动的目标，从而限制了主动轮廓线跟踪算法的应用范围，本文提出了一种基于多分辨率的主动轮廓线跟踪算法，它不但继承了基于单分辨率的主动轮廓线算法的优点，而且降低了对初始轮廓线的