Millions Of Triangles Research Articles

Multi-Material Mesh-Based Surface Tracking with Implicit Topology Changes

We introduce a multi-material non-manifold mesh-based surface tracking algorithm that converts self-intersections into topological changes. Our algorithm generalizes prior work on manifold surface tracking with topological changes: it preserves surface features like mesh-based methods, and it robustly handles topological changes like level set methods. Our method also offers improved efficiency and robustness over the state of the art. We demonstrate the effectiveness of the approach on a range of examples, including complex soap film simulations with thousands of interacting bubbles, and boolean unions of non-manifold meshes consisting of millions of triangles.

High-performance Ellipsoidal Clipmaps

This paper presents performance improvements for Ellipsoid Clipmaps, an out-of-core planet-sized geodetically accurate terrain rendering algorithm. The performance improvements were achieved by eliminating unnecessarily dense levels, more accurate block culling in the geographic coordinate system, and more efficient rendering methods. The elimination of unnecessarily dense levels is the result of analyzing and determining the optimal relative height of the viewer with respect to the most detailed level, resulting in the most consistent size of triangles across all visible levels. The proposed method for estimating the visibility of blocks based on view orientation allows rapid block-level view frustum culling performed in data space before visualization and spatial transformation of blocks. The use of a modern geometry pipeline through task and mesh shaders forced the handling of extremely fine granularity of blocks, but also shifted a significant part of the block culling process from CPU to the GPU. The approach described achieves high throughput and enables geodetically accurate rendering of the terrain based on the WGS 84 reference ellipsoid at very high resolution and in real time, with tens of millions of triangles with an average area of about 0.5 pix2 on a 1080p screen on mid-range graphics cards.

Trim Regions for Online Computation of From-Region Potentially Visible Sets

Visibility computation is a key element in computer graphics applications. More specifically, a from-region potentially visible set (PVS) is an established tool in rendering acceleration, but its high computational cost means a from-region PVS is almost always precomputed. Precomputation restricts the use of PVS to static scenes and leads to high storage cost, in particular, if we need fine-grained regions. For dynamic applications, such as streaming content over a variable-bandwidth network, online PVS computation with configurable region size is required. We address this need with trim regions, a new method for generating from-region PVS for arbitrary scenes in real time. Trim regions perform controlled erosion of object silhouettes in image space, implicitly applying the shrinking theorem known from previous work. Our algorithm is the first that applies automatic shrinking to unconstrained 3D scenes, including non-manifold meshes, and does so in real time using an efficient GPU execution model. We demonstrate that our algorithm generates a tight PVS for complex scenes and outperforms previous online methods for from-viewpoint and from-region PVS. It runs at 60 Hz for realistic game scenes consisting of millions of triangles and computes PVS with a tightness matching or surpassing existing approaches.

B/Surf: Interactive Bézier Splines on Surface Meshes.

We present a practical framework to port Bzier curves to surfaces. We support the interactive drawing and editing of Bzier splines on manifold meshes with millions of triangles, by relying on just repeated manifold averages. We show that direct extensions of the de Casteljau and Bernstein evaluation algorithms to the manifold setting are fragile, and prone to discontinuities when control polygons become large. Conversely, approaches based on subdivision are robust and can be implemented efficiently. We implement manifold extensions of the recursive de Casteljau bisection, and an open-uniform Lane-Riesenfeld subdivision scheme. For both schemes, we present algorithms for curve tracing, point evaluation, and approximated point insertion. We run bulk experiments to test our algorithms for robustness and performance, and we compare them with other methods at the state of the art, always achieving correct results and superior performance. For interactive editing, we port all the basic user interface interactions found in 2D tools directly to the mesh. We also support mapping complex SVG drawings to the mesh and their interactive editing.

EMBER

Boolean operators are an essential tool in a wide range of geometry processing and CAD/CAM tasks. We present a novel method, EMBER, to compute Boolean operations on polygon meshes which is exact, reliable, and highly performant at the same time. Exactness is guaranteed by using a plane-based representation for the input meshes along with recently introduced homogeneous integer coordinates. Reliability and robustness emerge from a formulation of the algorithm via generalized winding numbers and mesh arrangements. High performance is achieved by avoiding the (pre-)construction of a global acceleration structure. Instead, our algorithm performs an adaptive recursive subdivision of the scene's bounding box while generating and tracking all required data on the fly. By leveraging a number of early-out termination criteria, we can avoid the generation and inspection of regions that do not contribute to the output. With a careful implementation and a work-stealing multi-threading architecture, we are able to compute Boolean operations between meshes with millions of triangles at interactive rates. We run an extensive evaluation on the Thingi10K dataset to demonstrate that our method outperforms state-of-the-art algorithms, even inexact ones like QuickCSG, by orders of magnitude.

Efficient Ray Tracing of Large 3D Scenes for Mobile Distributed Computing Environments.

Cluster computing has attracted much attention as an effective way of solving large-scale problems. However, only a few attempts have been made to explore mobile computing clusters that can be easily built using commodity smartphones and tablets. To investigate the possibility of mobile cluster-based rendering of large datasets, we developed a mobile GPU ray tracer that renders nontrivial 3D scenes with many millions of triangles at an interactive frame rate on a small-scale mobile cluster. To cope with the limited processing power and memory space, we first present an effective 3D scene representation scheme suitable for mobile GPU rendering. Then, to avoid performance impairment caused by the high latency and low bandwidth of mobile networks, we propose using a static load balancing strategy, which we found to be more appropriate for the vulnerable mobile clustering environment than a dynamic strategy. Our mobile distributed rendering system achieved a few frames per second when ray tracing 1024 × 1024 images, using only 16 low-end smartphones, for large 3D scenes, some with more than 10 million triangles. Through a conceptual demonstration, we also show that the presented rendering scheme can be effectively explored for augmenting real scene images, captured or perceived by augmented and mixed reality devices, with high quality ray-traced images.

You can find geodesic paths in triangle meshes by just flipping edges

This paper introduces a new approach to computing geodesics on polyhedral surfaces---the basic idea is to iteratively perform edge flips , in the same spirit as the classic Delaunay flip algorithm. This process also produces a triangulation conforming to the output geodesics, which is immediately useful for tasks in geometry processing and numerical simulation. More precisely, our FlipOut algorithm transforms a given sequence of edges into a locally shortest geodesic while avoiding self-crossings (formally: it finds a geodesic in the same isotopy class ). The algorithm is guaranteed to terminate in a finite number of operations; practical runtimes are on the order of a few milliseconds, even for meshes with millions of triangles. The same approach is easily applied to curves beyond simple paths, including closed loops, curve networks, and multiply-covered curves. We explore how the method facilitates tasks such as straightening cuts and segmentation boundaries, computing geodesic Bézier curves, extending the notion of constrained Delaunay triangulations (CDT) to curved surfaces, and providing accurate boundary conditions for partial differential equations (PDEs). Evaluation on challenging datasets such as Thingi10k indicates that the method is both robust and efficient, even for low-quality triangulations.

Wavefront tracing for undersea acoustic propagation

An acoustic wavefront tracing algorithm and corresponding computational implementation is presented. Similar to ray tracing and beam tracing, wavefront tracing casts independent rays that refract through the medium and reflect from boundaries. However, with wavefront tracing these rays step congruously and are connected as a triangulated surface. As this triangulated surface moves through space and time it contacts boundaries, receivers, or other points of interest, thereby providing all possible paths from the source. As each triangle steps forward, three space-filling tetrahedra are formed and acoustic properties are then linearly interpolated via barycentric coordinates. This interpolation also applies at a boundary surface, allowing for both specular reflection and diffuse scattering. The algorithm is derived from the wavefront construction method, in which the triangulated surface is adaptively modified to maintain a desired level of detail. However, by keeping the wavefront topology fixed, along with an efficient encoding of the algorithm, highly detailed simulations with millions of triangles are achievable. Borrowing from the computer graphics community, a bounding volume hierarchy is used to accelerate the computation of ray–boundary intersections. An example simulation of a synthetic aperture sonar ping is provided, in which the wavefront comprises 21 million triangles and a rippled seafloor comprises five hundred thousand triangles.

A Combinatorial Approach for Constructing Lattice Structures

Abstract Lattice structures exhibit unique properties including a large surface area and a highly distributed load-path. This makes them very effective in engineering applications where weight reduction, thermal dissipation, and energy absorption are critical. Furthermore, with the advent of additive manufacturing (AM), lattice structures are now easier to fabricate. However, due to inherent surface complexity, their geometric construction can pose significant challenges. A classic strategy for constructing lattice structures exploits analytic surface–surface intersection; this, however, lacks robustness and scalability. An alternate strategy is voxel mesh-based isosurface extraction. While this is robust and scalable, the surface quality is mesh-dependent, and the triangulation will require significant postdecimation. A third strategy relies on explicit geometric stitching where tessellated open cylinders are stitched together through a series of geometric operations. This was demonstrated to be efficient and scalable, requiring no postprocessing. However, it was limited to lattice structures with uniform beam radii. Furthermore, existing algorithms rely on explicit convex-hull construction which is known to be numerically unstable. In this paper, a combinatorial stitching strategy is proposed where tessellated open cylinders of arbitrary radii are stitched together using topological operations. The convex hull construction is handled through a simple and robust projection method, avoiding expensive exact-arithmetic calculations and improving the computational efficiency. This is demonstrated through several examples involving millions of triangles. On a typical eight-core desktop, the proposed algorithm can construct approximately up to a million cylinders per second.

Automatic Shadow Detection in Urban Very-High-Resolution Images Using Existing 3D Models for Free Training

Up-to-date 3D city models are needed for many applications. Very-high-resolution (VHR) images with rich geometric and spectral information and a high update rate are increasingly applied for the purpose of updating 3D models. Shadow detection is the primary step for image interpretation, as shadow causes radiometric distortions. In addition, shadow itself is valuable geometric information. However, shadows are often complicated and environment-dependent. Supervised learning is considered to perform well in detecting shadows when training samples selected from these images are available. Unfortunately, manual labeling of images is expensive. Existing 3D models have been used to reconstruct shadows to provide free, computer-generated training samples, i.e., samples free from intensive manual labeling. However, accurate shadow reconstruction for large 3D models consisting of millions of triangles is either difficult or time-consuming. In addition, due to inaccuracy and incompleteness of the model, and different acquisition time between 3D models and images, mislabeling refers to training samples that are shadows but labeled as non-shadows and vice versa. We propose a ray-tracing approach with an effective KD tree construction to feasibly reconstruct accurate shadows for a large 3D model. An adaptive erosion approach is first provided to remove mislabeling effects near shadow boundaries. Next, a comparative study considering four classification methods, quadratic discriminant analysis (QDA) fusion, support vector machine (SVM), K nearest neighbors (KNN) and Random forest (RF), is performed to select the best classification method with respect to capturing the complicated properties of shadows and robustness to mislabeling. The experiments are performed on Dutch Amersfoort data with around 20% mislabels and the Toronto benchmark by simulating mislabels from inverting shadows to non-shadows. RF is tested to give robust and best results with 95.38% overall accuracy (OA) and a value of 0.9 for kappa coefficient (KC) for Amersfoort and around 96% OA and 0.92 KC for Toronto benchmarks when no more than 50% of shadows are inverted. QDA fusion and KNN are tested to be robust to mislabels but their capability to capture complicated properties of shadows is worse than RF. SVM is tested to have a good capability to separate shadow and non-shadows but is largely affected by mislabeled samples. It is shown that RF with free-training samples from existing 3D models is an automatic, effective, and robust approach for shadow detection from VHR images.

Scalable non-rigid registration for multi-view stereo data

We propose a new non-rigid registration method for large 3D meshes from Multi-View Stereo (MVS) reconstruction characterized by low-frequency shape deformations induced by several factors, such as low sensor quality and irregular sampling object coverage. Starting from a reference model to which we want to align a new 3D mesh, our method starts by decomposing it in patches using a Lloyd clustering before running an ICP local registration for each patch. Then, we improve the alignment using few geometric constraints and finally, we build a global deformation function that blends the estimated per-patch transformations. This function is structured on top of a deformation graph derived from the dual graph of the clustering. Our algorithm is iterated until convergence, increasing progressively the number of patches in the clustering to capture smaller deformations. The method comes with a scalable multicore implementation that enables, for the first time, the alignment of meshes made of tens of millions of triangles in a few minutes. We report extensive experiments of our algorithm on several dense Multi-View Stereo models, using a 3D scan or another MVS reconstruction as reference. Beyond MVS data, we also applied our algorithm to different scenarios, exhibiting more complex and larger deformations, such as 3D motion capture dataset or 3D scans of dynamic objects. The good alignment results obtained for both datasets highlights the efficiency and the flexibility of our approach.

ARTMe: a new array-based algorithm for Adaptive Refinement of Triangle Meshes

This work presents a new efficient array-based algorithm for adaptive mesh refinement capable of interactively generating millions of triangles. The new refinement algorithm satisfies important topological mesh properties, e.g., vertex valence control and a good mesh gradation. Furthermore, all local topological modifications of the triangle mesh are based on Stellar operators implemented on top of the Corner-Table topological data structure. This paper also shows that the proposed implementation provides a good balance in the trade-off between memory and processing time.

Multilevel mesh workflows towards CONUS scale watersheds: How small should triangles be to capture stream curvature for hydrological modeling?

Generating quality meshes for hydrological modeling is challenging. This article demonstrates using mesh workflows to incorporate national stream networks into very large dynamic meshes for distributed High Performance Computing (HPC). A multilevel quadtree is used to partition watersheds and merge stream networks ranging from hill-slope to four level-4 Hydrological Unit Code (HUC) scales, generating mesh sizes from hundreds to tens of millions of triangles. By using a mesh workflow, it is demonstrated how users control mesh quality, including triangle sizes, and the removal of small triangles and slivers. Four watersheds are studied, ranging in scale from Shale Hills (hill-slope) to the Chesapeake Bay (177,968 sq. km.) and are used to compare mesh characteristics with user parameters and different elevation sources to generate a spectrum of multilevel quadtree meshes. By using triangles to represent land surface and stream networks, stream curvature is lost and the mesh workflow is used to demonstrate what parameters are important to maintain stream characteristics. To meet HPC constraints, the multilevel quadtree is demonstrated by picking quadtree tiles along a storm path to show how users can combine quadtree tiles at multiple levels to concentrate HPC resources for hydrological science research.

Deferred Warping.

We introduce deferred warping, a novel approach for real-time deformation of 3D objects attached to an animated or manipulated surface. Our target application is virtual prototyping of garments where 2D pattern modeling is combined with 3D garment simulation which allows an immediate validation of the design. The technique works in two steps: First, the surface deformation of the target object is determined and the resulting transformation field is stored as a matrix texture. Then the matrix texture is used as look-up table to transform a given geometry onto a deformed surface. Splitting the process in two steps yields a large flexibility since different attachment types can be realized by simply defining specific mapping functions. Our technique can directly handle complex topology changes within the surface. We demonstrate a fast implementation in the vertex shading stage allowing the use of highly decorated surfaces with millions of triangles in real-time.

Efficient GPU out-of-core visualization of large-scale CAD models with voxel representations

Visualizing large-scale CAD models has been recognized as one of the most challenging tasks in engineering software development. Due to the constraints of limited GPU memory size and computation capacity, the CAD model of a complex product with hundreds of millions triangles cannot be loaded and rendered in real-time using most of modern GPUs. In this paper, an efficient voxel assisted GPU out-of-core framework is proposed for visualizing massive CAD models interactively. In order to reduce memory cost and improve efficiency of data streaming, a parallel off-line geometry attributes compression scheme is introduced to minimize the storage cost of each primitive by quantifying the LOD (levels of detail) geometries into a highly compact format. At the rendering stage, voxel representation is utilized to query visible objects by efficient ray casting algorithms, which is distinguishable from primitive or bounding box based visibility culling methods. The voxel representation is also utilized for shadow ray intersection test to generate soft shadow effect which results in enhancement of rendering realism. A prototype software system is developed to preprocess and render massive models with the proposed framework. Experimental results show that users can interactively visualize CAD models with hundreds of millions of triangles at high frame rates using our framework.

An efficient GPU out‐of‐core framework for interactive rendering of large‐scale CAD models

AbstractReal‐time rendering of large‐scale engineering computer‐aided design (CAD) models has been recognized as a challenging task. Because of the constraints of limited graphics processing unit (GPU) memory size and computation capacity, a massive model with hundreds of millions of triangles cannot be loaded and rendered in real‐time using most of modern GPUs. In this paper, an efficient GPU out‐of‐core framework is proposed for interactively visualizing large‐scale CAD models. To improve efficiency of data fetching from CPU host memory to GPU device memory, a parallel offline geometry compression scheme is introduced to minimize the storage cost of each primitive by compressing the levels of detail (LOD) geometries into a highly compact format. At the rendering stage, occlusion culling and LOD processing algorithms are integrated and implemented with an efficient GPU‐based approach to determine a minimal scale of primitives to be transferred for each frame. A prototype software system is developed to preprocess and render massive CAD models with the proposed framework. Experimental results show that users can walkthrough massive CAD models with hundreds of millions of triangles at high frame rates using our framework. Copyright © 2016 John Wiley & Sons, Ltd.

Next viewing directions for the scanning of dental impressions

This paper proposes a scanning procedure for a structured light system (SLS) to measure dental impressions. Although increasing the number of scanning orientations may improve the quality of a scanned model, it is desirable to minimize the number of scanning operations to time and data storage. We attempt to reduce the number of scanning operations to the least number that will still acquire a complete model. The proposed procedure must resolve two sub-problems: (1) identification of missing areas from given range images, and (2) determination of the next viewing directions to fill those missing areas. If we consider range images as triangular meshes, the sub-problems can be solved by using well-known geometric algorithms. The triangular meshes, however, may consist of tens of millions of triangles, which require an unacceptably long time to compute. To cope with this problem, we propose two key ideas: (1) utilizing an inherent attribute of a range image, the map structure; (2) utilizing a graphics board to accelerate the evaluation of visibility. Our demonstration proves that the proposed approach improves the quality of scanned models and reduces the number of scanning operations.

An improved slicing algorithm with efficient contour construction using STL files

Slicing is an important step for all layer-based additive manufacturing (AM) processes. This paper proposes an improved and robust slicing algorithm with efficient contour construction to reduce the slicing time and be able to slice a complex STereoLithography (STL) model with millions of triangles (resulting from high-accuracy STL files). Also, another important feature of the proposed method is it can identify outer and inner contours automatically. Test results to demonstrate the improvements in execution time and comparisons with results from published papers have been given to illustrate the algorithm efficiency. The robustness of this slicing algorithm is demonstrated by several complex models with large numbers of nested contours on each slice.

YASARA View - molecular graphics for all devices - from smartphones to workstations.

Summary: Today's graphics processing units (GPUs) compose the scene from individual triangles. As about 320 triangles are needed to approximate a single sphere—an atom—in a convincing way, visualizing larger proteins with atomic details requires tens of millions of triangles, far too many for smooth interactive frame rates. We describe a new approach to solve this ‘molecular graphics problem’, which shares the work between GPU and multiple CPU cores, generates high-quality results with perfectly round spheres, shadows and ambient lighting and requires only OpenGL 1.0 functionality, without any pixel shader Z-buffer access (a feature which is missing in most mobile devices).Availability and implementation: YASARA View, a molecular modeling program built around the visualization algorithm described here, is freely available (including commercial use) for Linux, MacOS, Windows and Android (Intel) from www.YASARA.org.Contact: elmar@yasara.orgSupplementary information: Supplementary data are available at Bioinformatics online.

Optimizing redundant-data clustering for interactive walkthrough applications

In modern walkthrough applications, storing massive datasets has become easy and inexpensive due to the availability of gigantic disk-based storage devices including hard drives, DVDs, and Blu-ray discs. However, fetching data from these devices for processing and rendering in interactive environments remains a bottleneck as data transfer speed has not kept pace with the sizes of both the secondary storage and main memory.Out-of-core algorithms are commonly used as a solution to transfer data efficiently from the secondary storage to main memory. Existing algorithms strongly rely on suitable data layout algorithms to reduce the data fetch time. However, in spite of all commonly used techniques, the total time required to seek and transfer data can still easily exceed the budget for total data fetch time. In this work, we propose an orthogonal approach to aggregate data and store them redundantly in multiple places in the storage device to ensure consistent data fetching performance. We pose this as a linear integer programming problem to minimize the amount of redundancy subject to the fetch time budget constraint. We provide an implementation on datasets with hundreds of millions of triangles to demonstrate how this data clustering can be created in practice and how the optimal solution is found.