Yuanjun He
Shanghai Jiao Tong University
Network
Latest external collaboration on country level. Dive into details by clicking on the dots.
Publication
Featured researches published by Yuanjun He.
Advanced Engineering Informatics | 2008
Jiale Wang; Yuanjun He; Haishan Tian; Hongming Cai
Reuse of existing designs can help companies save up a significant amount of time and money in large-scale product development. Searching out similar CAD models of products is a key to facilitating design reuse. This paper proposes a method for retrieving CAD models of mechanical parts by freehand sketches. Users can draw three 2D outline sketches and a skeleton sketch to express their query intent. The 2D outline sketches describe the geometrical information of models and the skeleton sketch contains the topological information. A relevance feedback mechanism is introduced to combine the two similarity degrees measured by 2D outlines or skeleton into a unified one. Experiments are conducted to evaluate the performance of the proposed method.
international conference on computational science and its applications | 2005
Yong Wu; Yuanjun He; Haishan Tian
An algorithm based on barycentric coordinates is presented to solve the point location problem in spherical triangulation meshes. During the preprocessing stage, a subdivision connectivity mesh is constructed to partition the spherical domain into some subdivision regions. Then we find a representation triangle from the triangle set of original spherical mesh for each subdivision region. During the locating stage, we firstly find the subdivision region containing the query point p and select the corresponding representation triangle as the start one to walk. Then the barycentric coordinates are used to extract local heuristic information about the location of p, so as to find the shortest path from the start triangle to the target one. In comparison with traditional algorithms, our approach has better time-space performance.
international conference on artificial reality and telexistence | 2006
Haishan Tian; Yuanjun He; Hongming Cai; Lirong Feng
We present an automated implemented and computationally efficient method for smoothly morphing of point set surface. In the issue, corresponding between the surfaces is difficult for point set surface without any topological information. To overcome this, the method is guided by registration and warp functions, then the intermediate surfaces are constructed depend on only the local geometric information expressed by the point locations in 3D space. To reduce the distortion of the intermediate shapes, the distance metric between points is defined, and the algorithm optimally matches the points in the two models. The automated method provides the animator with a technique that can be used to create a set of models forming a smooth transition between pairs of a given sequence of keyframe models. The advantage of the new approach is that the capability of morphing between objects having a different topological without any intervene of animators.
The Visual Computer | 2005
Yong Wu; Yuanjun He; Haishan Tian
For closed two-manifold genus-0 meshes, the sphere is the most natural parameterization domain. Like planar parameterization, spherical parameterization without foldover and with low distortion consitutes two basic challenges. However, in order to prevent foldovers, many current studies cannot validly reduce distortion, especially area distortion. In this paper, we present a two-stage relaxation method to reduce the distortion of spherical meshes parameterized by traditional methods. The first stage is an authalic (area-preserving) relaxation algorithm that iteratively relaxes the mesh surface by growing patches around seed triangles. The second one is a relaxation procedure based on SCDM (spherical-domain compositive distortion metric).
ieee international conference on progress in informatics and computing | 2010
Qianwei Bian; Weimin Luo; Yuanjun He; Hongming Cai
Due to the recent significant improvements in game entertainment, CAD, CAE and some other 3D fields, the number of 3D models is increasing rapidly. As a result, there is an increasing need for procedures supporting the automatic search for 3D objects in databases. In this paper, we present a novel framework to retrieve 3D models in 3 steps. Firstly, users input multiple ordered images of an object as input. Certain photo environment is needed. A voxel model will be generated automatically according to these inputs. Secondly, we provide a tool to modify voxel model. This procedure can help eliminate defects caused by uncertain photo environments. If the model produced by the first step is reasonable, this step will be optional. Finally, the modified model will be retrieved by a method called Solid-D2, which provides a fast and discriminating descriptor for 3D shapes. This algorithm is chosen to err on the side of retrieval speed with certain precision. Related models from the database will be listed. Experimental results indicate that the proposed method is easy-to-use, efficient and applicable for 3D model retrieval.
international conference on artificial reality and telexistence | 2006
Zhi-ping Hu; Yuanjun He; Zong-ying Ou
A general IBR (image based rendering) approach schema for synthesis image of new viewpoint is presented in this paper. Giving a series of images of an environment on closely spaced two circles viewpoints the proposed algorithm can generate perspective intermediate images of the environment in any viewpoint between two circles. In this case the same environment is displayed from the view of a virtual camera controlled by the user. The synthesis processing consists of three steps: First, the ring-like region between outer and inner circles can be divided as a set of fan-like sub-regions. Find the fan-like region to encircle the specified viewpoint; then an IBR processing between two images taken by moving camera along optical axis is implemented to generate two intermediate image; finally, an IBR processing between two images taken by moving camera around the scene is implemented to generate the destination image corresponding the specified viewpoint. Some experiment result images are also shown in this paper.
computer-aided design and computer graphics | 2005
Haishan Tian; Yuanjun He; Yong Wu
Parameterization of 3D mesh plays an important role in computer graphics application society, such as geometry compression, texture mapping, morphing and so on. As far as closed two-manifold genus-0 meshes are concerned, natural choice of parameterization domain is sphere, and the basic challenges are no-foldover and low-distortion respectively. To solve the spherical parameterization problem, we present spherical-domain hybrid stretch metric (SHSM). The concept of global area is introduced into SHSM to further reduce the area distortion of parameterized triangles. We demonstrate the ability of our method to both parameterize fast and avoid high distortion, realizing parameterization from the spatial genus-0 meshes onto the unit sphere.
international conference on information sciences and interaction sciences | 2010
Qianwei Bian; Zhiping Hu; Yuanjun He; Hongming Cai
As the number of 3D models available is increasing rapidly, searching out models from such databases is becoming an eager demand in many fields. In this paper, we present a new search engine called Bling to retrieve 3D models by sketch. Aiming at developing a kind of user interface with which even novice users can draw to search, we employed freehand sketches to create voxel models as query. A novel matching algorithm called volume D2 (VD2) is also adopted to compare generated models with database models. Experiments show that Bling provides a feasible solution for easy-to-use 3D model retrieval.
Data Science Journal | 2008
Jiale Wang; Hongming Cai; Yuanjun He
With the development of computer graphics and digitalizing technologies, 3D model databases are becoming ubiquitous. This paper presents a method for content-based searching for similar 3D models in databases. To assess the similarity between 3D models, shape feature information of models must be extracted and compared. We propose a new 3D shape feature extraction algorithm. Experimental results show that the proposed method achieves good retrieval performance with short computation time.
international symposium on communications and information technologies | 2005
Haishan Tian; Yuanjun He; Hongming Cai
This paper summarizes the progress achieved through the build of active virtual scene Web service; an active VR three-tier Web service that is designed to acquire virtual scene requests from users connected to a Web browser, perform the virtual scene schemas process and provide them with the virtual scene results. To make schema broker more intelligent, schema-oriented monitoring registry and schema miner are introduced. Dynamic modeling capability is achieved by X-VRML which is a high-level XML-based procedural language. Using X-VRML, a method of modeling active virtual world in database is present; the method provides means of parameterization of all elements of the dynamically generated virtual scenes. In this way, a virtual world model is composed of VR-Data, VR-Schema, VR-Templates, and VR-Scenes.