Authentication email has already been sent, please check your email box: and activate it as soon as possible.
You can login to My Profile and manage your email alerts.
If you haven’t received the email, please:
|
|
There are 9 papers published in subject: > since this site started. |
Results per page: |
Select Subject |
Select/Unselect all | For Selected Papers |
Saved Papers
Please enter a name for this paper to be shown in your personalized Saved Papers list
|
1. Optimal View Detection for Automatic Photographing on a Wireless-controlled Drone | |||
Xiong Xiaoliang,Feng Jie,Zhou Bingfeng | |||
Computer Science and Technology 11 December 2015 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:Automatic photographing is a challenge problem because it is difficult to measure the illumination, the color and the composition during taking photos. In this work, we focus on the image composition optimization and make use of the currently popular aerocraft "Drone" to create a flying automatic photographer. Some commonly used composition guidelines are adopted to calculate an aesthetic score of an image, based on which the photographer make some adjustments to a better view, until an optimal view is detected and a photo is token. Experiments demonstrate that our optimal view detection strategy is effective. | |||
TO cite this article:Xiong Xiaoliang,Feng Jie,Zhou Bingfeng. Optimal View Detection for Automatic Photographing on a Wireless-controlled Drone[OL].[11 December 2015] http://en.paper.edu.cn/en_releasepaper/content/4666739 |
2. Sparse Sampling and Completion for Light Transport in VPL-based Rendering | |||
Yuchi Huo, Rui Wang, Xinguo Liu, Hujun Bao | |||
Computer Science and Technology 26 November 2015 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:The many-light formulation provides a general framework for rendering various illumination effects using hundreds of thousands of virtual point lights (VPLs). To efficiently gather the contributions of the VPLs, lightcuts and its extensions cluster the VPLs, which implicitly approximates the lighting matrix with some representative blocks similar to vector quantization. In this paper, we propose a new approximation method based on the previous lightcut method and a low-rank matrix factorization model. As many researchers pointed out, the lighting matrix is low rank, which implies that it can be completed from a small set of known entries.We first generate a conservative global light cut with bounded error and partition the lighting matrix into slices by the coordinate and normal of the surface points using the method of lightslice. Then we perform two passes of randomly sampling on each matrix slice. In the first pass, uniformly distributed random entries are sampled to coarsen the global light cut, further clustering the similar light for the spatially localized surface points of the slices. In the second pass, more entries are sampled according to the possibility distribution function estimated from the first sampling result. Then each matrix slice is factorized into a product of two smaller low-rank matrices constrained by the sampled entries, which delivers a completion of the lighting matrix. The factorized form provides an additional speedup for adding up the matrix columns which is more GPU friendly. Compared with the previous lightcut based methods, we approximate the lighting matrix with some signal specialized bases via factorization. The experimental results shows that we can achieve significant acceleration than the state of the art many-light methods. | |||
TO cite this article:Yuchi Huo, Rui Wang, Xinguo Liu, et al. Sparse Sampling and Completion for Light Transport in VPL-based Rendering[OL].[26 November 2015] http://en.paper.edu.cn/en_releasepaper/content/4663747 |
3. The Applications of 3D Model's Multi-instances Generation Algorithm | |||
Wan Lili | |||
Computer Science and Technology 25 December 2011 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:In shape matching, retrieval and classification, a 3D model is always taken as a whole to do feature extraction and similarity measure, but in fact many 3D models from the internet are composed of multiple 3D objects. 3D model's multi-instances generation algorithm can solve this problem, whose key idea is to divide 3D model consisted of several objects into some instances. In this paper, we discuss two applications which can highly benefit from this alogrithm. These applications include 3D model retrieval and semantic annotation. In the first application, the key is the shape similarity between the query model and the model having multiple objects. In the second application, the semantic annotation of each instance is preferentially solved. Experiments show the two applications achieve good results. | |||
TO cite this article:Wan Lili. The Applications of 3D Model's Multi-instances Generation Algorithm[OL].[25 December 2011] http://en.paper.edu.cn/en_releasepaper/content/4457565 |
4. Structsurfel: Physical Structure Parameterized Surfel for Animation | |||
Yang Qi,Li Sheng | |||
Computer Science and Technology 24 January 2011 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:Animation methods of point based model are always lack of efficiency since using point based resampling and relaxation algorithms will have to access every animated surface point of the model. This paper presents a new method for accelerated animation of point based models by introducing new parameters to surface splatting. Differs to the classical method, surface splatting of the model are divided into soft parts and hard parts, hard parts are grouped into blocks and animated while soft parts are generated on the fly using GPU subdivision during the animation process instead of using low efficiency relaxation or resampling methods. The animation process includes two lines: assemble line which constructs the structsurfels and pseudo topology information which is needed by the soft part generation by physical parameters; animation line generates the animation of hard parts and soft parts of the structsurfel model by combining the point based methods and subdivision methods. We achieved a high performance and a same quality of animation compared to the traditional methods. | |||
TO cite this article:Yang Qi,Li Sheng. Structsurfel: Physical Structure Parameterized Surfel for Animation[OL].[24 January 2011] http://en.paper.edu.cn/en_releasepaper/content/4408906 |
5. Adjacency Sampling: A Scalable Line Drawing Kernel with Artifact Reduced | |||
Tang Chen,Li Sheng | |||
Computer Science and Technology 17 January 2011 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:In this paper we exploit sampling topology information in image space directly for visibility of line drawing and silhouette extraction. We propose a new line drawing kernel that depends on image-space adjacency test between primitives in GPU without any preprocessing step or extra adjacent information prestored. By this kernel, our visibility test acquires high accuracy in wireframe rendering and performs fairly well also in sketch and stylized line drawing, and our silhouette extraction method extracts visible portion of silhouette edges in image-space with clear and regular outlook. Our methods can be easily implemented and be controlled. The experiments show the privileges of our method in line drawing. | |||
TO cite this article:Tang Chen,Li Sheng. Adjacency Sampling: A Scalable Line Drawing Kernel with Artifact Reduced[OL].[17 January 2011] http://en.paper.edu.cn/en_releasepaper/content/4406859 |
6. Automatic Human Body Reconstruction Based on Orthogonal Image Contour | |||
Zhong Yueqi | |||
Computer Science and Technology 02 August 2010 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:With the development of online apparel retailing, the shape of the human body becomes the keystone when evaluating the dressing result in the cyberspace. In this paper, an automatic method has been presented to generate an avatar by deforming a candidate model against two orthogonal images of the online shopper. The candidate model is prepared from range scanned data. Mean value deformation is initiated to make it coincide with the input image contours. The correspondence between control points on the candidate model and points on the image contour are built automatically by applying positional and directional constraints respectively. Using the proposed method, we can approximate the body shape of the online shopper effectively by preserving his/her features. | |||
TO cite this article:Zhong Yueqi. Automatic Human Body Reconstruction Based on Orthogonal Image Contour[OL].[ 2 August 2010] http://en.paper.edu.cn/en_releasepaper/content/4380531 |
7. Mesh Topology in Scanned Garment Reconstruction | |||
Yueqi Zhong,Hongyan Liu | |||
Computer Science and Technology 26 February 2010 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:3D garment with high fidelity can be obtained via range data scanning. The original output of body scanner is an unorganized points cloud. In this paper, the geometrical surface of 3D garment is reconstructed through a series of treatments. The primary target of this work is to investigate the dynamic behavior of the corresponding physical model transferred from different mesh topologies. A mass-spring model is constructed for both regular meshes and irregular meshes. The performance under various integration methods is evaluated. Experimental results reveal the procedure of regularization is suitable for the integrators that are sensitive to the physically-based simulation of scanned garments. | |||
TO cite this article:Yueqi Zhong,Hongyan Liu. Mesh Topology in Scanned Garment Reconstruction[OL].[26 February 2010] http://en.paper.edu.cn/en_releasepaper/content/40256 |
8. Implicit Surface Fitting by Regularized Regression and Compactly Supported Radial Basis Functions | |||
Pan Rongjiang | |||
Computer Science and Technology 27 November 2007 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:We describe a direct way of making use of surface normal vectors at sample points in the problem of implicit surface fitting with compactly supported radial basis functions. The normal vectors are incorporated in a regularized regression problem that leads to a n by n positive definite linear system given n surface point/normal pairs. Compared with the widely used heuristic methods, our method avoids of introducing manufactured off-surface points and can fit much larger datasets effectively. We demonstrate its robust performance on several datasets. | |||
TO cite this article:Pan Rongjiang . Implicit Surface Fitting by Regularized Regression and Compactly Supported Radial Basis Functions[J]. |
9. Three Dimensional Strain Field Morphing | |||
Han-Bing Yan,Shi-Min Hu,Ralph Martin | |||
Computer Science and Technology 12 December 2005 | |||
Show/Hide Abstract | Cite this paper︱Full-text: PDF (0 B) | |||
Abstract:In this paper, we present a new technique for 3D shape morphing using strain fields. Strain is an important geometric quantity used in mechanics to describe the deformation extent of objects. We use it to analyze and control deformation in morphing. Using the position vector field, the strain field between source and target shapes can be obtain. This strain field is then interpolated between zero and its final value to give the position field for intermediate shapes. This method ensures that the 3D morphing process is smooth and no shape jittering or wobbling happens, and that locally, volumes suffer minimal distortion. We also show how to modify the method so that changes of shape (in particular, lengths) vary linearly with time. | |||
TO cite this article:Han-Bing Yan,Shi-Min Hu,Ralph Martin. Three Dimensional Strain Field Morphing[OL].[12 December 2005] http://en.paper.edu.cn/en_releasepaper/content/4233 |
Select/Unselect all | For Selected Papers |
Saved Papers
Please enter a name for this paper to be shown in your personalized Saved Papers list
|
Results per page: |
About Sciencepaper Online | Privacy Policy | Terms & Conditions | Contact Us
© 2003-2012 Sciencepaper Online. unless otherwise stated