Proceedings. 2nd International Symposium on 3D Data Processing, Visualization and Transmission, 2004. 3DPVT 2004.最新文献

Multiple view reconstruction of people 人的多重视图重构

Proceedings. 2nd International Symposium on 3D Data Processing, Visualization and Transmission, 2004. 3DPVT 2004. Pub Date : 2004-09-06 DOI: 10.1109/TDPVT.2004.1335229

A. Hilton, J. Starck

{"title":"Multiple view reconstruction of people","authors":"A. Hilton, J. Starck","doi":"10.1109/TDPVT.2004.1335229","DOIUrl":"https://doi.org/10.1109/TDPVT.2004.1335229","url":null,"abstract":"This work presents a unified framework for model-based and model-free reconstruction of people from multiple camera views in a studio environment. Shape and appearance of the reconstructed model are optimised simultaneously based on multiple view silhouette, stereo and feature correspondence. A priori knowledge of surface structure is introduced as regularisation constraints. Model-based reconstruction assumes a known generic lutmanoid model a priori, which is fitted to the multi-view observations to produce a structured representation for animation. Model-free reconstruction makes no priori assumptions on scene geometry allowing the reconstruction of complex dynamic scenes. Results are presented for reconstruction of sequences of people from multiple views. The model-based approach produces a consistent structured representation, which is robust in the presence of visual ambiguities. This overcomes limitations of existing visual-hull and stereo techniques. Model-free reconstruction allows high-quality novel view-synthesis with accurate reproduction of the detailed dynamics for hair and loose clothing. Multiple view optimisation achieves a visual quality comparable to the captured video without visual artefacts due to misalignment of images.","PeriodicalId":191172,"journal":{"name":"Proceedings. 2nd International Symposium on 3D Data Processing, Visualization and Transmission, 2004. 3DPVT 2004.","volume":"22 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127216447","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 30

3D stereoscopic image pairs by depth-map generation 通过深度图生成三维立体图像对

Proceedings. 2nd International Symposium on 3D Data Processing, Visualization and Transmission, 2004. 3DPVT 2004. Pub Date : 2004-09-06 DOI: 10.1109/3DPVT.2004.10

S. Battiato, A. Capra, S. Curti, M. Cascia

引用次数: 98

Segmenting correlation stereo range images using surface elements 利用曲面元分割相关立体距离图像

Proceedings. 2nd International Symposium on 3D Data Processing, Visualization and Transmission, 2004. 3DPVT 2004. Pub Date : 2004-09-06 DOI: 10.1109/TDPVT.2004.1335301

D. Murray, J. Little

引用次数: 7

Hierarchical 3D surface reconstruction based on radial basis functions 基于径向基函数的分层三维曲面重建

Proceedings. 2nd International Symposium on 3D Data Processing, Visualization and Transmission, 2004. 3DPVT 2004. Pub Date : 2004-09-06 DOI: 10.1109/TDPVT.2004.1335290

P. Dalmasso, R. Nerino

{"title":"Hierarchical 3D surface reconstruction based on radial basis functions","authors":"P. Dalmasso, R. Nerino","doi":"10.1109/TDPVT.2004.1335290","DOIUrl":"https://doi.org/10.1109/TDPVT.2004.1335290","url":null,"abstract":"Volumetric methods based on implicit surfaces are commonly used in surface reconstruction from uniformly distributed sparse 3D data. The case of nonuniform distributed data has recently deserved more attention, because it occurs frequently in practice. This work describes a volumetric approach to surface reconstruction from nonuniform data which is suitable for the reconstruction of surfaces from images, in particular from multiple views. Differently from volumetric methods which use both 3D surface points and surface normals, the approach does not use the surface normals because they are often unreliable when estimated from image data. The method is based on a hierarchical partitioning of the volume data set. The working volume is split and classified at different scales of spatial resolution into surface, internal and external voxels and this hierarchy is described by an octree structure in a multiscale framework. The octree structure is used to build a multiresolution description of the surface by means of compact support radial basis functions (RBF). A hierarchy of surface approximations at different levels of details is built by representing the voxels at the same octree level as RBF of similar spatial support. At each scale, information related to the reconstruction error drives the reconstruction process at the following finer scale. Preliminary results on synthetic data and future perspectives are presented.","PeriodicalId":191172,"journal":{"name":"Proceedings. 2nd International Symposium on 3D Data Processing, Visualization and Transmission, 2004. 3DPVT 2004.","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122347147","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 7

Automated texture mapping of 3D city models with oblique aerial imagery 倾斜航空图像的3D城市模型的自动纹理映射

Proceedings. 2nd International Symposium on 3D Data Processing, Visualization and Transmission, 2004. 3DPVT 2004. Pub Date : 2004-09-06 DOI: 10.1109/TDPVT.2004.1335266

Christian Früh, R. Sammon, A. Zakhor

引用次数: 174

Robust 3D segmentation for underwater acoustic images 水声图像的鲁棒三维分割

Proceedings. 2nd International Symposium on 3D Data Processing, Visualization and Transmission, 2004. 3DPVT 2004. Pub Date : 2004-09-06 DOI: 10.1109/TDPVT.2004.1335399

L. Tao, U. Castellani, Vittorio Murino

{"title":"Robust 3D segmentation for underwater acoustic images","authors":"L. Tao, U. Castellani, Vittorio Murino","doi":"10.1109/TDPVT.2004.1335399","DOIUrl":"https://doi.org/10.1109/TDPVT.2004.1335399","url":null,"abstract":"A technique for 3D acoustic image segmentation and modelling is proposed. Especially, in the underwater environment, in which optical sensors suffer from visibility problems, the acoustical devices may provide efficient solutions, but, on the other hand, acoustic image interpretation is surely more difficult for a human operator. The proposed application involves the use of an acoustic camera which directly acquires images structured as a set of 3D points. Due to the noisy nature of this type of data, the segmentation problem becomes more challenging and the standard algorithms for range image segmentation are likely to fail. The proposed method is based on a simplified version of the so called recover and select paradigm in which the seed areas, from which the segmentation starts, are generated by adopting a robust approach based on the RANSAC (Random Sample And Consensus) algorithm. Superquadric primitives are directly recovered from raw data without any pre-segmentation processing. Experimental trials using both synthetic and real acoustical images confirm the goodness of the method, and a large robustness of the resulting segmented images, associated to a relatively low computational load.","PeriodicalId":191172,"journal":{"name":"Proceedings. 2nd International Symposium on 3D Data Processing, Visualization and Transmission, 2004. 3DPVT 2004.","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129936538","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 6

A closed-form solution for a two-view self-calibration problem under fixation 固定下双视图自校准问题的封闭解

Proceedings. 2nd International Symposium on 3D Data Processing, Visualization and Transmission, 2004. 3DPVT 2004. Pub Date : 2004-09-06 DOI: 10.1109/TDPVT.2004.1335300

T. Ueshiba, F. Tomita

{"title":"A closed-form solution for a two-view self-calibration problem under fixation","authors":"T. Ueshiba, F. Tomita","doi":"10.1109/TDPVT.2004.1335300","DOIUrl":"https://doi.org/10.1109/TDPVT.2004.1335300","url":null,"abstract":"It is well known that the epipolar geometry between two uncalibrated perspective views is completely encapsulated in the fundamental matrix. Since the fundamental matrix has seven degrees of freedom (DOF), self-calibration is possible if at most seven of the intrinsic or extrinsic camera parameters are unknown by extracting them from the fundamental matrix. This work presents a linear algorithm for self-calibrating a perspective camera which undergoes fixation, that is, a special motion in which the camera's optical axis is confined in a plane. Since this fixation has four degrees of freedom, which is one smaller than that of general motion, we can extract at most three intrinsic parameters from the fundamental matrix. We here assume that the focal length (1 DOF) and the principal point (2 DOF) are unknown but fixed for two views. It will be shown that these three parameters are obtained from the fundamental matrix in an analytical fashion and a closed-form solution is derived. We also characterize all the degenerate motions under which there exists an infinite set of solutions.","PeriodicalId":191172,"journal":{"name":"Proceedings. 2nd International Symposium on 3D Data Processing, Visualization and Transmission, 2004. 3DPVT 2004.","volume":"67 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128661528","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 2

A Bayesian framework for 3D models retrieval based on characteristic views 基于特征视图的三维模型检索贝叶斯框架

Proceedings. 2nd International Symposium on 3D Data Processing, Visualization and Transmission, 2004. 3DPVT 2004. Pub Date : 2004-09-06 DOI: 10.1109/TDPVT.2004.1335187

Tarik Filali Ansary, Jean-Philippe Vandeborre, S. Mahmoudi, M. Daoudi

引用次数: 29

Fast and robust bore detection in range image data for industrial automation 用于工业自动化的距离图像数据的快速鲁棒孔检测

Proceedings. 2nd International Symposium on 3D Data Processing, Visualization and Transmission, 2004. 3DPVT 2004. Pub Date : 2004-09-06 DOI: 10.1109/TDPVT.2004.1335282

G. Biegelbauer, M. Vincze

引用次数: 6

Enhanced vector quantization for data reduction and filtering 增强矢量量化的数据减少和过滤

Proceedings. 2nd International Symposium on 3D Data Processing, Visualization and Transmission, 2004. 3DPVT 2004. Pub Date : 2004-09-06 DOI: 10.1109/TDPVT.2004.1335275

S. Ferrari, I. Frosio, V. Piuri, N. A. Borghese

{"title":"Enhanced vector quantization for data reduction and filtering","authors":"S. Ferrari, I. Frosio, V. Piuri, N. A. Borghese","doi":"10.1109/TDPVT.2004.1335275","DOIUrl":"https://doi.org/10.1109/TDPVT.2004.1335275","url":null,"abstract":"Modern automatic digitizers can sample huge amounts of 3D data points on the object surface in a short time. Point based graphics is becoming a popular framework to reduce the cardinality of these data sets and to filter measurement noise, without having to store in memory and process mesh connectivity. Main contribution of this paper is the introduction of soft clustering techniques in the field of point clouds processing. In this approach data points are not assigned to a single cluster, but they contribute in the determination of the position of several cluster centres. As a result a better representation of the data is achieved. In soft clustering techniques, a data set is represented with a reduced number of points called reference vectors (RV), which minimize an adequate error measure. As the position of the RVs is determined by \"learning\", which can be viewed as an iterative optimization procedure, they are inherently slow. We show here how partitioning the data domain into disjointed regions called hyperboxes (HB), the computation can be localized and the computational time reduced to linear in the number of data points (O(N)), saving more than 75% on real applications with respect to classical soft-VQ solutions, making therefore VQ suitable to the task. The procedure is suitable for a parallel HW implementation, which would lead to a complexity sublinear in N. An automatic procedure for setting the voxel side and the other parameters can be derived from the data-set analysis. Results obtained in the reconstruction of faces of both humans and puppets as well as on models from clouds of points made available on the Web are reported and discussed in comparison with other available methods.","PeriodicalId":191172,"journal":{"name":"Proceedings. 2nd International Symposium on 3D Data Processing, Visualization and Transmission, 2004. 3DPVT 2004.","volume":"71 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2004-09-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128202756","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 5