Design of Algorithm for the 3D Object Representation Based on the Web3D Using X3D

Reconstruction 3D objects from a single view is an important research direction in computer vision. With the development of deep learning technology, a single-view 3D reconstruction based on deep learning has made remarkable progress in recent years. In reviewing 3D object reconstruction based on deep learning, firstly, the research progress of 3D reconstruction methods based on deep learning is analyzed in detail according to different representations of 3D objects, i.e., point cloud, voxel, and mesh. Secondly, the standard 3D reconstruction datasets and evaluation criteria are summarized. Finally, the challenges of 3D object reconstruction based on deep learning are summarized.

Computer graphics systems and CAD/CAM systems are widely used and an abundance of 3D-Data in various fields exists. However, based on the VRML technique, it is difficult to send such 3D-Data through the Internet, because of the large data size. Transmission of practical and highly detailed 3D-Data through the Internet becomes a primary requirement.Therefore, a compact and qualified 3D-Data representation method is greatly required. This paper describes XVL (eXtended VRML with Lattice), a new framework for compact 3D-Data representation with high quality surface shape. By utilizing a free-form surface technique, qualified surfaces are transferred with a limited amount of data size and rendered. Free-form surfaces transferred by an efficient data structure are called lattice structure. This data structure contains only vertices, topologies, and attributes, and can be converted to the original surface. Because the lattice structure is regarded as a polygon mesh, it can be easily integrated to a VRML file. These surfaces and lattice have the same topology and are thus interchangeable in the lattice structure. By using weighting attributes, a sophisticated surface shape can be represented. Some practical XVL applications, such as an intuitive surface design system, are also introduced.

Technologies for 3D mesh compression: A survey

Visual working memory (VWM) is typically measured using arrays of two-dimensional isolated stimuli with simple visual identities (e.g., color or shape), and these studies typically find strong capacity limits. Science, technology, engineering and mathematics (STEM) experts are tasked with reasoning with representations of three-dimensional (3D) connected objects, raising questions about whether those stimuli would be subject to the same limits. Here, we use a color change detection task to examine working memory capacity for 3D objects made up of differently colored cubes. Experiment 1a shows that increasing the number of parts of an object leads to less sensitivity to color changes, while change-irrelevant structural dimensionality (the number of dimensions into which parts of the structure extend) does not. Experiment 1b shows that sensitivity to color changes decreases similarly with increased complexity for multipart 3D connected objects and disconnected 2D squares, while sensitivity is slightly higher with 3D objects. Experiments 2a and 2b find that when other stimulus characteristics, such as size and visual angle, are controlled, change-irrelevant dimensionality and connectivity have no effect on performance. These results suggest that detecting color changes on 3D connected objects and on displays of isolated 2D stimuli are subject to similar set size effects and are not affected by dimensionality and connectivity when these properties are change-irrelevant, ruling out one possible explanation for scientists’ advantages in storing and manipulating representations of complex 3D objects.

We present new interaction and visualization paradigms relying on free form surfaces for studying and exploring human anatomy. We propose an interface for building three-dimensional anatomical scenes incorporating 3D anatomical organ models, freely orientated slices and free form surfaces extracted from the Visible Human dataset. Compared with planar slices, free form surfaces allow to follow curved anatomic structures such as the aorta tree or the pelvis. In the present paper, we describe in detail 3D interaction techniques for creating free form surfaces. The interactive placement of surface boundary curves relies on the combination of an interactive slice navigator and of a 3D visualization interface integrated within a single java applet. Surface boundary curve control points are placed with the mouse at the desired locations within the selected slices. The corresponding boundary curves are displayed in the 3D visualization interface as fat 3D cubic spline curves, which provide immediate feedback. Boundary curves may be easily duplicated, translated and modified. The specified boundary curves are interpolated by Coons patches, yielding a perfectly smooth surface. That surface may be visualized in combination with semi-transparent organ models. It may also be flattened and shown in a separate window. The presented application is available online as a Java applet (http://visiblehuman.epfl.ch).

The problem of determination of friction factor (or average fluid velocity) for a steady gravitational flow of fluid in an inclined open triangular groove is addressed. It is assumed that the angle of slope of the groove is known and the gravity is the only driving force of flow. The solution comprises of two parts. First, for a given groove’s basal angle, liquid-solid contact angle, and the Bond number, the shape of free surface is determined starting from the Young-Laplace equation. The optimization technique as developed formerly is applied. Then, having determined the shape of the free surface and slope of the groove, the distribution of fluid flow is determined. The boundary value problem is solved using boundary collocation method in least square sense. Given the distribution of fluid velocity the friction factor as a function of the other parameters of the model is analyzed.

When a fluid liquid interface joins a solid boundary a common line is formed, this line will be referre to as the contact line. The main objective of this paper is to obtain explicit solutions of the governing equations of fluid mechanics for viscous flow of a liquid, in the vicinity of a stationary contact line. In each case we have indicated the procedure for developing locally consistent asymptotic expansions for the flow field and the shape of the free surface, with explicit solutions mostly for the first two terms only. Apart from arbitrary constants of the type f'(0; Aj+1) y which are to be determined by the distant driving mechanisms, the flow field and the shape of free surface are completely determined by the contact angle 0(1 and the local arbitrary parameter pQ. However, mainly due to difficulties in labelling the eigen-values and partly because of the size of the paper, in this part we present results for wetting liquids only.

With the advent of techniques devised for the mass production of optical components made with surfaces of arbitrary form (also known as free form surfaces) in the last years, a parallel development of measuring systems adapted for these new kind of surfaces constitutes a real necessity for the industry. Profilometry is one of the preferred methods for the assessment of the quality of a surface, and is widely employed in the optical fabrication industry for the quality control of its products. In this work, we present the design, development and assembly of a new profilometer with five axis of movement, specifically suited to the measurement of medium size (up to 150 mm of diameter) free-form optical surfaces with sub-micrometer accuracy and low measuring times. The apparatus is formed by three X, Y, Z linear motorized positioners plus and additional angular and a tilt positioner employed to locate accurately the surface to be measured and the probe which can be a mechanical or an optical one, being optical one a confocal sensor based on chromatic aberration. Both optical and mechanical probes guarantee an accuracy lower than the micrometer in the determination of the surface height, thus ensuring an accuracy in the surface curvatures of the order of 0.01 D or better. An original calibration procedure based on the measurement of a precision sphere has been developed in order to correct the perpendicularity error between the axes of the linear positioners. To reduce the measuring time of the profilometer, a custom electronics, based on an Arduino™ controller, have been designed and produced in order to synchronize the five motorized positioners and the optical and mechanical probes so that a medium size surface (around 10 cm of diameter) with a dynamic range in curvatures of around 10 D, can be measured in less than 300 seconds (using three axes) keeping the resolution in height and curvature in the figures mentioned above.

An approach to automatically extract three dimensional (3D) models (that is, geometries and topologies) of physical objects in a facility is described. The rationale for this work is its repeated use in efficiently developing databases of 3D objects for applying virtual reality (VR) tools in detailed layout decision support. Obtaining 3D object models can be a challenging task. Sometimes they are available, for example, in a Computer-Aided Design (CAD) database and these can be readily imported into a VR database. But on many occasions one is not so fortunate and these object models have to be created in correlation to an existing or proposed facility, which can be an extremely tedious and time consuming task. A time efficient and economical alternative is to use video camera images, but quickly and accurately capturing the depth information from 2D camera images has so far remained elusive because the existing methodologies are too general purpose and operate at a lower level of abstraction, namely digitized images. We have developed a method for directly inputting 3D objects into VR-aided facility layout models, by integrating the strengths of previously tried and tested technological components: (i) camera calibration; (ii) image processing; (iii) stereo vision; and (iv) Delaunay triangulation. The techniques described here are embedded in a prototype architecture and toolkit called MIRRORS (Methodology for Inputting Raw Recordings into 3D Object Renderings for Stereo). The primary contribution of this paper is that it has been able to design an integrated system to build 3D object models from 2D images. The MIRRORS system has been primarily designed for objects without free-form surfaces and whose shape can be recovered from a relatively nondense set of points.

Multiview 3D reconstruction techniques enable digital reconstruction of 3D objects from the real world by fusing different viewpoints of the same object into a single 3D representation. This process is by no means trivial and the acquisition of high quality point cloud representations of dynamic 3D objects is still an open problem. In this paper, an approach for high fidelity 3D point cloud generation using low cost 3D sensing hardware is presented. The proposed approach runs in an efficient low-cost hardware setting based on several Kinect v2 scanners connected to a single PC. It performs autocalibration and runs in real-time exploiting an efficient composition of several filtering methods including Radius Outlier Removal (ROR), Weighted Median filter (WM) and Weighted Inter-Frame Average filtering (WIFA). The performance of the proposed method has been demonstrated through efficient acquisition of dense 3D point clouds of moving objects.

Three-dimensional object recognition is difficult because an object looks different when viewed from different viewpoints. One solution to this problem is to represent the 3D object as a set of 2D models, one for each of a set of view classes. A view class is a set of viewpoints that all produce images with the same or similar features. View-class matching consists of determining the correspondence between the features extracted from an image of an unknown object and the features of a particular view class of a particular object model. View-class matching is used in object recognition, pose estimation, and inspection systems.

A finite-element algorithm for computing free-surface flows driven by arbitrary body forces is presented. The algorithm is primarily designed for the microfluidic parameter range where (i) the Reynolds number is small and (ii) force-driven pressure and flow fields compete with the surface tension for the shape of a stationary free surface. The free surface shape is represented by the boundaries of finite elements that move according to the stress applied by the adjacent fluid. Additionally, the surface tends to minimize its free energy and by that adapts its curvature to balance the normal stress at the surface. The numerical approach consists of the iteration of two alternating steps: The solution of a fluidic problem in a prescribed domain with slip boundary conditions at the free surface and a consecutive update of the domain driven by the previously determined pressure and velocity fields. For a Stokes problem the first step is linear, whereas the second step involves the nonlinear free-surface boundary condition. This algorithm is justified both by physical and mathematical arguments. It is tested in two dimensions for two cases that can be solved analytically. The magnitude of the errors is discussed in dependence on the approximation order of the finite elements and on a step-width parameter of the algorithm. Moreover, the algorithm is shown to be robust in the sense that convergence is reached also from initial forms that strongly deviate from the final shape. The presented algorithm does not require a remeshing of the used grid at the boundary. This advantage is achieved by a built-in mechanism that causes a smooth change from the behavior of a free surface to that of a rubber blanket if the boundary mesh becomes irregular. As a side effect, the element sides building up the free surface in two dimensions all approach equal lengths. The presented variational derivation of the boundary condition corroborates the numerical finding that a second-order approximation of the velocity also necessitates a second-order approximation for the free surface discretization.

We explore representation of 3D objects in which several distinct 2D views are stored for each object. We demonstrate the ability of a two-layer network of thresholded summation units to support such representations. Using unsupervised Hebbian relaxation, the network learned to recognize ten objects from different viewpoints. The training process led to the emergence of compact representations of the specific input views. When tested on novel views of the same objects, the network exhibited a substantial generalization capability. In simulated psychophysical experiments, the network's behavior was qualitatively similar to that of human subjects.

Developing a three-dimensional (3D) data model for Geographic Information Systems (GIS) is an essential and complex issue. 3D modelling in GIS is becoming ever more important for the development of cyber cities and digital earth, which have recently become feasible. A competent 3D model forms an efficient foundation for 3D visualization, query and spatial analysis. As a development of the existing 3D models, this study proposes particular improvements in handling complex 3D objects. We present an object-oriented data model for handling complex 3D objects in GIS. First, the conceptual data model is developed based on the principle of object-oriented (OO) data modelling. This model is designed based on the following three basic geometric elements: node, segment and triangle. Accordingly, the abstract geometric objects are defined: including points, lines, surfaces and volumes. Second, the corresponding 3D logical model is designed based on the defined abstract objects and the relationships between them. Third, a formal representation of the 3D spatial objects is described in detail. Fourth, a prototype 3D GIS is developed based on the proposed 3D data model. Finally, we describe the results of an experimental study to reconstruct 3D objects using this 3D GIS and a comparison with the performance of other 3D data models. The proposed model is able to handle complex objects, such as complex buildings and TV towers, which is an essential functionality for building large-scale cyber cities, such as for Hong Kong. The proposed data model proves to be very efficient, particularly in visualization and rendering. The experimental results show which the data volume of the proposed model is compacted and the visualization speed for 3D objects is improved, compared with the existing models.

When machining a free-form surface automatically and digitally, especially in the case of sophisticated surface shapes, it is very difficult to control the surface quality, and thus sophisticated surfaces are usually polished using manual labor. Over the past few years, there has been little attention to the calculation of material removal depth models and construction of theoretical roughness models considering the influence of the curvature radius. Bonnet polishing can be automatically adapted to polish complex free-form surfaces. This paper explores key problems related to forecasting surface quality with respect to bonnet polishing of free-form surfaces. First, for the convex and planar sub-regions, this paper deduces the relationship expressing the maximum pressure distribution and the curvature radius, and presents a computational expression for material removal depth, taking into consideration the influence of the curvature radius. An expression is also deduced for the dwell time relationship between the adjacent processing points according to the experimental results pertaining to the material removal depth. From this, a theoretical roughness model is constructed that relates the bonnet curvature radius and the workpiece curvature radius. The validity of the experiments is summarized in the conclusion. The research findings provide a basic theory for the prediction of surface quality that can be automatically adapted to a free-form surface shape in bonnet polishing.