教育经历：

1992.09-1996.06 西安交通大学无线电技术专业学士学位
1999.09-2001.09 西安交通大学信号与信息处理专业硕士学位
2002.03-2006.06 西安交通大学信号与信息处理专业博士学位

工作经历

1996.7－1999.8 西安交通大学科技处 科员
2001.8－2002.8 西安交通大学电信学院信通系助教
2002.9－2007.7 西安交通大学电信学院信通系讲师
2007.8 西安交通大学电信学院信通系副教授
2008.5 西安交通大学电信学院信通系博士生导师
2008.9 入选西安交通大学“青年骨干教师培养计划"
2010.2 入选2009年教育部新世纪优秀人才支持计划
2011.1 晋升教授

学术兼职：

中国计算机学会青年科技论坛（YOCSEF）AC委员
中国计算机学会多媒体专委会委员
中国计算机学会人工智能与模式识别专委会委员
中国计算机学会学术工委委员
中国图像图形学会多媒体专委会委员
中国图像图形学会青年工委委员
陕西省信号处理学会理事

发表论文：

2011

_____________________________________________________________________________________________________
Arcimboldo-like Collage Using Internet Images ACM Transactions on Graphics, 2011, accepted Hua Huang, Lei Zhang, Hongchao Zhang
Collage is a composite artwork made from assemblage of different material forms. In this work, we present a novel approach for creating a fantastic collage artform, namely Arcimboldo-like collage, which represents an input image with multiple thematically-related cutouts from the filtered Internet images. Due to the massive data of Internet images, competent image cutouts can almost always be discovered to match the segmented components of the input image. The selected cutouts are purposefully arranged such that as a whole assembly, they can represent the input image with disguise in both shape and color; but separately, individual cutout is still recognizable as its own being. Experimental results and user study show that our algorithm can effectively produce the entertaining Arcimboldolike collages.
_____________________________________________________________________________________________________
Web-Image Driven Best Views of 3D Shapes The Visual Computer. 2011, Accepted Hong Liu, Lei Zhang, Hua Huang
The rapid advance of the Internet provides available huge database of web images. In this paper, we introduce a novel approach for automatically computing the best views of 3D shapes based on their web images. Best view selection is generally an intuitive task of getting the most information of a 3D shape. The novelty of our approach is to directly explore human perception on observing 3D shapes from the relevant web images. Those images are captured from biased views of different people, thus sufficiently reflecting view choice when observing the 3D shapes. By collecting web images possibly captured from the similar views, the best view is selected as the one possessing the most web images. We experiment our method with the shapes in Princeton Shape Benchmark (PSB), as well make comparisons with traditional geometric descriptor based approaches. The results demonstrate that our method is not only robust but also able to produce more canonical views in accordance with human perception.
_____________________________________________________________________________________________________
What’re in a scene? - Towards the Computational Best View of 3D Scene Advanced Science Letters, 2011, Accepted Lei Zhang, Hong Liu, Hua Huang
A 3D scene is usually composed of groups of multiple objects with different materials. From the best viewpoint, people should easily observe what are in the scene. In this paper, we present a method to automatically compute the best view for a 3D scene. By clustering on geometric features of different orders, a novel scene view descriptor is proposed to enable faithful measurement on the concurrent importance of scene objects. Then the view cost function based on the descriptor is optimized to select the best view from a set of candidate views. From the best view computed by our method, people can quickly get a complete picture on the target scene. We experiment our method with a series of 3D scene models from Google 3D Warehouse, and make comparisons with previous view selection methods as well as some user studies on the computed best views. The results demonstrate that our method is efficient and robust towards the computation of the best views of 3D scenes.
_____________________________________________________________________________________________________
Fast feature-based mode decision for 4×4 intra prediction in H.264/AVC Science in China Series F: Information Sciences. 2011, Accepted Wang Ping, Huang Hua, Tan Zheng
_____________________________________________________________________________________________________
RepSnapping: efficient image cutout for repeated scene elements Computer Graphics Forum. 2011, Accepted Hua Huang, Lei Zhang, Hong-Chao Zhang
Repeated scene elements are copious and ubiquitous in natural images. Cutout of those repeated elements usually involves tedious and laborious user interaction by previous image segmentation methods. In this paper, we present RepSnapping, a novel method oriented to cutout of repeated scene elements with much less user interaction. By exploring inherent similarity between repeated elements, a new optimization model is introduced to thread correlated elements in the segmentation procedure. The model proposed here enables efficient solution using maxflow/ min cut on an extended graph. Experiments indicate that RepSnapping facilitates cutout of repeated elements better than the state-of-the-art interactive image segmentation and repetition detection methods.
_____________________________________________________________________________________________________
Fast Facial Image Super-resolution via Local Linear Transformations for Resource-limited Applications IEEE Transactions on Circuits and Systems for Video Technology, 2011. Onlie Hua Huang, Ning Wu
Most of popular learning-based super-resolution (SR) approaches suffer from complicated learning structures and highly intensive computation, especially in resource-limited applications. We propose a novel frontal facial image SR approach by using multiple local linear transformations to approximate the non-linear mapping between low-resolution (LR) and highresolution (HR) images in the pixel domain. We adopt Procrustes analysis to obtain orthogonal matrices representing the learned linear transformations, which can not only well capture appearance variations in facial patches but also greatly simplify the transformation computation to matrices manipulation. An HR image can be directly reconstructed from a single LR image without need of the large training data, thus avoiding the use of a large redundant LR and HR patch database. Experimental results show that our approach is computationally fast as well the SR quality compares favorably with the state-of-the-art approaches from both subjective and objective evaluations. Besides, our approach is insensitive to the size of training data and robust to a wide range of facial variations like occlusions. More importantly, the proposed method is also much more effective than other comparative methods to reconstruct real world images captured from the Internet and webcams.
_____________________________________________________________________________________________________
Internet image voting for best view selection of 3D shapes Computer Graphics International Conference. 2011 Hong Liu, Lei Zhang, Hua Huang
Based on the huge image database on the In- ternet, we introduce a new approach to automatically select the best view for a 3D shape by Internet image voting (IIV). The problem of selecting the best view is mainly considered to be a task of getting the most information of a 3D shape, which is much intuitive and deeply related to the nature of human perception. The novelty of our IIV approach is to encourage common hu- man perception to directly determine the best view. To collect the human perception on a given 3D shape, our approach rst searches relevant images on the Internet. By carefully ltering out some cluttered images, we ob- tain a database where the images are feasible to analyze as well possibly contains candidate views captured from dierent people. Then, for each sampled view of the 3D shape, images tally votes for the best matching view by considering both shape and saliency similarity. Finally, the candidate view with the most votes is selected as the best view. We experiment our method with the shapes in Princeton Shape Benchmark, and also make compar- isons with traditional geometric descriptor based ap- proaches. The results demonstrate that our method is not only robust but also able to produce better view in accordance with human perception.
_____________________________________________________________________________________________________
Painterly rendering with content-dependent natural paint strokes The Visual Computer. 2011, 27(9):861-871 Hua Huang, TianNan Fu
We present a new painterly rendering method that simulates artists’ content-dependent painting process and the natural variation of hand-painted strokes. First, a new stroke layout strategy is proposed to enhance the contrast between large and small paint strokes, which is an important characteristic of hand-painted paintings. Specifically, the input image is partitioned into nonuniform grids according to its importance map, and determined by the grid size, an individually constructed paint stroke is applied in each grid. Second, an anisotropic digital brush is designed to simulate a real paint brush. In particular, each bristle of the digital brush has an individual color, so that strokes rendered by the new brush can have multiple colors and naturally varied textures. Finally, we present a novel method to add lighting effects to the canvas. This lighting imitation method is robust and very easy to implement, and it can significantly improve the quality of rendering. Comparing with traditional painterly rendering approaches, the new method simulates more closely the real painting procedure, and our experimental results show that it can produce vivid paintings with fewer artifacts.
_____________________________________________________________________________________________________
A fast two-step block type decision algorithm for intra prediction in H.264/AVC high profile Multimedia Tools and Applications. 2011, Online Ping Wang, Hua Huang, Zheng Tan
Intra coding in H.264/AVC can significantly improve the compression efficiency but at the cost of high computational complexity. To reduce the complexity, this paper presents an efficient block type decision algorithm for intra prediction. In H.264/AVC high profile, three kinds of block types are supported. This algorithm determines the optimal block type by two steps. The first step is based on the fact that the block type of intra prediction is highly dependent on the smoothness of macroblock. An edge-based feature is introduced to characterize the smoothness, by comparing it with two thresholds impossible block type is firstly filtered out. Then the second step is based on the correlation of block type chosen for different chroma modes. Experimental results show that the proposed fast algorithm can achieve 72.8% time saving on average for encoding the all intra-frame sequence with average 0.75% bit rate increase and 0.05dB PSNR degradation when comparing with the reference software.
_____________________________________________________________________________________________________
EXCOL: an extract-and-complete approach to cartoon animation reusing IEEE Transactions on Visualization and Computer Graphics. 2011, Online Lei Zhang, Hua Huang, Hong-Bo Fu
We introduce the EXCOL method (EXtract-and-COmplete Layering) — a novel cartoon animation processing technique to convert a traditional animated cartoon video into multiple semantically meaningful layers. Our technique is inspired by vision-based layering techniques but focuses on shape cues in both the extraction and completion steps to reflect the unique characteristics of cartoon animation. For layer extraction, we define a novel similarity measure incorporating both shape and color of automatically segmented regions within individual frames and propagate a small set of user-specified layer labels among similar regions across frames. By clustering regions with the same labels, each frame is appropriately partitioned into different layers, with each layer containing semantically meaningful content. Then a warping-based approach is used to fill missing parts caused by occlusion within the extracted layers to achieve a complete representation. EXCOL provides a flexible way to effectively reuse traditional cartoon animations with only a small amount of user interaction. It is demonstrated that our EXCOL method is effective and robust, and the layered representation benefits a variety of applications in cartoon animation processing.
_____________________________________________________________________________________________________
Super-Resolution Method for Face Recognition Using Nonlinear Mappings on Coherent Features IEEE Transaction on Neural Networks. 2011, 22(1): 121-130 Hua Huang, Huiting He
Low-resolution (LR) of face images significantly decreases the performance of face recognition. To address this problem, we present a super-resolution method that uses nonlinear mappings to infer coherent features that favor higher recognition of the nearest neighbor (NN) classifiers for recognition of single LR face image. Canonical correlation analysis is applied to establish the coherent subspaces between the principal component analysis (PCA) based features of high-resolution (HR) and LR face images. Then, a nonlinear mapping between HR/LR features can be built by radial basis functions (RBFs) with lower regression errors in the coherent feature space than in the PCA feature space. Thus, we can compute super-resolved coherent features corresponding to an input LR image according to the trained RBF model efficiently and accurately. And, face identity can be obtained by feeding these super-resolved features to a simple NN classifier. Extensive experiments on the Facial Recognition Technology, University of Manchester Institute of Science and Technology, and Olivetti Research Laboratory databases show that the proposed method outperforms the state-of-the-art face recognition algorithms for single LR image in terms of both recognition rate and robustness to facial variations of pose and expression.

_____________________________________________________________________________________________________

返回顶部

2010

_____________________________________________________________________________________________________
Manifold Learning for Visualizing and Analyzing High- dimensional Data IEEE Intelligent Systems. 2010, 25(4): 54-61 Junping Zhang, Hua Huang, Jue Wang
_____________________________________________________________________________________________________
Super-resolution of Human Face Image Using Canonical Correlation analysis Pattern Recognition. 2010, 43(7): 2532-2543 Hua Huang, Huiting He, Xin Fan, Junping Zhang
Super-resolution reconstruction of face image is the problem of reconstructing a high resolution face image from one or more low resolution face images. Assuming that high and low resolution images share similar intrinsic geometries, various recent super-resolution methods reconstruct high resolution images based on a weights determined from nearest neighbors in the local embedding of low resolution images. These methods suffer disadvantages from the finite number of samples and the nature of manifold learning techniques, and hence yield unrealistic reconstructed images. To address the problem, we apply canonical correlation analysis (CCA), which maximizes the correlation between the local neighbor relationships of high and low resolution images. We use it separately for reconstruction of global face appearance, and facial details. Experiments using a collection of frontal human faces show that the proposed algorithm improves reconstruction quality over existing state-of-the-art super-resolution algorithms, both visually, and using a quantitative peak signal-to-noise ratio assessment.
_____________________________________________________________________________________________________
A Simple Approach to Multiview Face Hallucination IEEE Signal Processing Letter. 2010, 17(6): 579-582 Xiang Ma, Hua Huang, Shaopeng Wang, and Chun Qi
Most face hallucination methods are usually limited to frontal face with small pose variations. This letter presents a simple and efficient multiview face hallucination (MFH) method to generate high-resolution (HR) multiview faces from a single given low-resolution (LR) one. The problem is addressed in two steps. A simple face transformation method is proposed by defining a constrained least square problem for LR multiview face transformation and a position-patch based face hallucination method is extended to incorporate HR multiview face details. Experimental results show that our approach has some advantages over existing MFH methods.
_____________________________________________________________________________________________________
Example-Based Painting Guided by Color Features The Visual Computer. 2010, 26(6-8): 933 - 942 Hua Huang, Yu Zang, Chen-Feng Li
In this paper, by analyzing and learning the color features of the reference painting with a novel set of measures, an example-based approach is developed to transfer some key color features from the template to the source image. First, color features of a given template painting is analyzed in terms of hue distribution and the overall color tone. These features are then extracted and learned by the algorithm through an optimization scheme. Next, to ensure the spatial coherence of the final result, a segmentation based post processing is performed. Finally, a new color blending model, which avoids the dependence of edge detection and adjustment of inconvenient tune parameters, is developed to provide a flexible control for the accuracy of painting. Experimental results show that the new example-based painting system can produce paintings with specific color features of the template, and it can also be applied to changing color themes of art pieces, designing color styles of paintings/real images, and specific color harmonization.
_____________________________________________________________________________________________________
Example-based contrast enhancement by gradient mapping The Visual Computer. 2010, 26(6-8): 731-738 Hua Huang, Xuezhong Xiao
Contrast enhancement is a very important problem in image processing. The key issue is how to assign correct enhancement levels for the local regions in an image, which makes previous methods incur much artifacts, e.g., over-enhancement, halo. In this paper, an example-based contrast enhancement algorithm is proposed, which works in the gradient domain. We utilize GMM model to describe the gradient distribution of an image. Then a GMM-based gradient mapping method is proposed to transfer the gradient of a reference image to the source image. The enhanced image is obtained by solving a Poisson equation defined by the altered gradient. Experimental results show the effectiveness and robustness of our method.
_____________________________________________________________________________________________________
Mesh Reconstruction by Meshless Denoising and Parameterization Computers & Graphics. 2010, 34(3): 198-208 Lei Zhang, Ligang Liu, Craig Gotsman, Hua Huang
We present a new approach to simultaneously denoise and parameterize unorganized point cloud data. This is achieved by minimizing an appropriate energy function defined on the denoised point cloud and its parameterization. An iterative algorithm to minimize the energy is described. The key ingredient of our approach is an ``as-rigid-as-possible" meshless parameterization to map a point cloud with disk topology to the plane without building the connectivity of the point cloud. Then 2D triangulation method can be applied to the planar parameterization to provide triangle connectivity for the 2D points, which can be transferred back to the 3D point cloud to form a triangle mesh surface. We also show how to generalize the approach to meshes with closed topology, of any genus. Experimental results have shown that our approach can effectively denoise the point cloud and our meshless parameterization can preserve local distances in the point cloud, resulting in a more regular 3D triangle mesh, compared to other methods.
_____________________________________________________________________________________________________
RBF Network-Based Temporal Color Morphing Journal of computer animation and virtual worlds. 2010, 21(3-4): 289-296, SCI UT ISI: 000280135400016 Xuezhong Xiao, Hua Huang, Lizhuang Ma
A method of RBF network-based temporal color morphing is proposed to simulate the natural phenomena characterized by temporal color alteration, e.g., turning green of foliage, resurgence of leaves. Such phenomena usually span a long time and it is very difficult to capture their whole process. Our system accepts a source image sequence and a reference image as input. The source sequence contains the desired scene except for color alteration, and the reference image has the color style which the source sequence is expected to advance into. First, an RBF network is employed to model the mapping between the colors of the source sequence and the reference image. Then, a simple interpolation algorithm is applied to render the resulting sequence. The effectiveness of the new method is verified by experiments.

_____________________________________________________________________________________________________

返回顶部