Ambient Occlusion Research Articles

UV‐free Texturing using Sparse Voxel DAGs

AbstractAn application may have to load an unknown 3D model and, for enhanced realistic rendering, precompute values over the surface domain, such as light maps, ambient occlusion, or other global‐illumination parameters. High‐quality uv‐unwrapping has several problems, such as seams, distortions, and wasted texture space. Additionally, procedurally generated scene content, perhaps on the fly, can make manual uv unwrapping impossible. Even when artist manipulation is feasible, good uv layouts can require expertise and be highly labor intensive.This paper investigates how to use Sparse Voxel DAGs (or DAGs for short) as one alternative to avoid uv mapping. The result is an algorithm enabling high compression ratios of both voxel structure and colors, which can be important for a baked scene to fit in GPU memory. Specifically, we enable practical usage for an automatic system by targeting efficient real‐time mipmap filtering using compressed textures and adding support for individual mesh voxelizations and resolutions in the same DAG. Furthermore, the latter increases the texture‐compression ratios by up to 32% compared to using one global voxelization, DAG compression by 10 – 15% compared to using a DAG per mesh, and reduces color‐bleeding problems for large mipmap filter sizes.The voxel‐filtering is more costly than standard hardware 2D‐texture filtering. However, for full HD with deferred shading, it is optimized down to 2.5 ± 0.5 ms for a custom multisampling filtering (e.g., targeted for minification of low‐frequency textures) and 5 ± 2 ms for quad‐linear mipmap filtering (e.g., for high‐frequency textures). Multiple textures sharing voxelization can amortize the majority of this cost. Hence, these numbers involve 1–3 textures per pixel (Fig. 1c).

RGB2AO: Ambient Occlusion Generation from RGB Images

AbstractWe present RGB2AO, a novel task to generate ambient occlusion (AO) from a single RGB image instead of screen space buffers such as depth and normal. RGB2AO produces a new image filter that creates a non‐directional shading effect that darkens enclosed and sheltered areas. RGB2AO aims to enhance two 2D image editing applications: image composition and geometry‐aware contrast enhancement. We first collect a synthetic dataset consisting of pairs of RGB images and AO maps. Subsequently, we propose a model for RGB2AO by supervised learning of a convolutional neural network (CNN), considering 3D geometry of the input image. Experimental results quantitatively and qualitatively demonstrate the effectiveness of our model.

Adapting spherical-harmonics-based geometric morphometrics (SPHARM) for 3D images containing large cavity openings using ambient occlusion: a study with hermit crab claw shape variability

One of the advantages of mesh-based geometric morphometrics (GM) over landmark-based approaches, is that it affords the possibility of the precise examination of highly irregular shapes and complex topographic surfaces. In the case of spherical-harmonic-based GM, the main prerequisite is a completely closed mesh surface, which is often lacking, particularly when dealing with natural objects. Here we present a methodology to prepare 3D segmentations with large cavity openings for spherical-harmonics-based GM analysis. This will be exemplified by a case study of hermit crab (Paguroidea, Decapoda, and Crustacea) claws. The joint openings of the claws—between manus and ‘movable finger’ (dactylus)—typify the problem of the large cavity opening. We described a methodology, including an ambient-occlusion-based segmentation algorithm, which leads to precise results and thereby enables the study of inter- and intraspecific differences in the shape of hermit crab claws. Analysis showed a statistically significant separation between all examined diogenid and pagurid claws, whereas separation observed between all left and right claws was not significant. Additionally, the procedure offers other benefits: it is easy to reproduce and causes little variance in the data, closures integrate smoothly into the overall structures and the algorithm saves a significant amount of time.

The Fusion of Unmatched Infrared and Visible Images Based on Generative Adversarial Networks

Visible images contain clear texture information and high spatial resolution but are unreliable under nighttime or ambient occlusion conditions. Infrared images can display target thermal radiation information under day, night, alternative weather, and ambient occlusion conditions. However, infrared images often lack good contour and texture information. Therefore, an increasing number of researchers are fusing visible and infrared images to obtain more information from them, which requires two completely matched images. However, it is difficult to obtain perfectly matched visible and infrared images in practice. In view of the above issues, we propose a new network model based on generative adversarial networks (GANs) to fuse unmatched infrared and visible images. Our method generates the corresponding infrared image from a visible image and fuses the two images together to obtain more information. The effectiveness of the proposed method is verified qualitatively and quantitatively through experimentation on public datasets. In addition, the generated fused images of the proposed method contain more abundant texture and thermal radiation information than other methods.

DeepAO: Efficient Screen Space Ambient Occlusion Generation via Deep Network

Ambient occlusion ( abbr . AO) plays an important role in realistic rendering applications because AO produces more realistic ambient lighting, which is achieved by calculating the brightness of certain screen parts based on objects’ geometry. However, the baseline computation of AO algorithm is time-consuming, which limits its application for real-time rendering. Currently, most AO algorithms are based on screen space to reduce the computational consumption, which leads to unrealistic results due to the usage of artificial features. To overcome these challenges, in this paper, we first create a well-crafted dataset with the pair of deferred shading buffer data and ground-truth AO shaded images. Then, we design an efficient deep neural network for the screen space AO image generation, based on which we further design a Compute Shader Library to compute the shaded AO images. Our extensive experimental results show that our method achieves competent performance than existing screen space ambient or volumetric ambient based AO methods both in visual quality and efficiency.

물리 기반 쉐이딩을 바탕으로 하는 텍스처링에 관한 연구

급속히 변화를 거듭하는 미디어의 환경 속에서 컴퓨터그래픽은 자신의 영역을 빠르게 확장해 나아가고 있다. 이러한 미디어의 환경 속에서 텍스처링은 컴퓨터그래픽의 핵심적인 역할을 담당한다. 지금까지 컴퓨터그래픽의 주된 목적이 실재 세계의 재현이라고 보았을 때 컴퓨터그래픽 제작의 한 과정인 텍스처링 또한 기술적 발전의 도움으로 컴퓨터그래픽과 함께 발맞춰 나아가고 있다. 텍스처링은 모델링 된 오브젝트의 표면에 대해 육안으로 보이는 모습의 특성을 결정해주는 단계이다. 본 연구는 텍스처링이 컴퓨터그래픽의 기존 텍스처링의 개념을 넘어서는 물리 기반 쉐이딩을 바탕으로 하는 텍스처링에 관하여 연구하였다.BR 물리 기반 쉐이딩(Physically Based Shading)은 컴퓨터그래픽을 좀 더 사실적으로 표현하는데 중요한 역할을 담당하고 있다. 이러한 물리 기반 쉐이딩을 바탕으로 텍스처링은 아티스트가 일일이 직감적으로 텍스처를 만들어가는 기존의 방식에서 새로운 방식의 다양한 맵을 바탕으로 물리적으로 좀 더 정확한 텍스처를 만들어가는 방식으로 변화하고 있다. 물리 기반 렌더링을 바탕으로 하는 텍스처링에서 금속성(Metallic)와 거칠기(Roughness) 채널은 중요한 요소이기에 물리 기반 쉐이더에서 핵심적 위치를 차지하고 있다. 텍스처링을 위해서는 다양한 맵들이 필요한데 대표적으로 노말 맵(Normal Map), 앰비언트 오크루전 맵(Ambient Occlusion), 커버처 맵(Curvature Map), 시크니스 맵(Thickness Map) 그리고 포지션 맵(Position Map) 등이 있다. 이렇게 다양한 맵들을 활용하여 물리 기반 방식을 바탕으로 좀 더 정확하고 쉽게 텍스처링을 할 수 있다.

Volumetric Isosurface Rendering with Deep Learning-Based Super-Resolution.

Rendering an accurate image of an isosurface in a volumetric field typically requires large numbers of data samples. Reducing this number lies at the core of research in volume rendering. With the advent of deep learning networks, a number of architectures have been proposed recently to infer missing samples in multidimensional fields, for applications such as image super-resolution. In this article, we investigate the use of such architectures for learning the upscaling of a low resolution sampling of an isosurface to a higher resolution, with reconstruction of spatial detail and shading. We introduce a fully convolutional neural network, to learn a latent representation generating smooth, edge-aware depth and normal fields as well as ambient occlusions from a low resolution depth and normal field. By adding a frame-to-frame motion loss into the learning stage, upscaling can consider temporal variations and achieves improved frame-to-frame coherence. We assess the quality of inferred results and compare it to bi-linear and cubic upscaling. We do this for isosurfaces which were never seen during training, and investigate the improvements when the network can train on the same or similar isosurfaces. We discuss remote visualization and foveated rendering as potential applications.

Optimization of screen-space directional occlusion algorithms

Abstract Developers of video games and simulations from the day one have been trying to improve visuals of their products. The appearance of the scenes depends to a large extent on the approximation to the physical basis of light behaviour in the environments presented. The best effects in this regard are global illumination. However, it is too computationally expensive. One of the methods to simulate global illumination without a lot of processing is Screen-Space Ambient Occlusion. Many implementations of this technique were created, though few take into account direction and colour of the incoming light. An exception is a technique named SSDO – Screen-Space Directional Occlusion. Unfortunately, it suffers from the same drawbacks as its less realistic cousins, such as noise and banding while also remaining moderately expensive for computation. The main purpose of this paper is to optimize basic SSDO method using technique called Statistical Volumetric Obscurance, enhancing its performance while retaining plausible visual effect.

Tracking Registration Algorithm for Augmented Reality Based on Template Tracking

Tracking registration is a key issue in augmented reality applications, particularly where there are no artificial identifier placed manually. In this paper, an efficient markerless tracking registration algorithm which combines the detector and the tracker is presented for the augmented reality system. We capture the target images in real scenes as template images, use the random ferns classi- fier for target detection and solve the problem of reinitialization after tracking registration failures due to changes in ambient lighting or occlusion of targets. Once the target has been successfully detected, the pyramid Lucas-Kanade (LK) optical flow tracker is used to track the detected target in real time to solve the problem of slow speed. The least median of squares (LMedS) method is used to adaptively calculate the homography matrix, and then the three-dimensional pose is estimated and the virtual object is rendered and registered. Experimental results demonstrate that the algorithm is more accurate, faster and more robust.

Interactive directional ambient occlusion and shadow computations for volume ray casting

Volume rendering is a widely used technique to visualize 3D scalar data. To enhance visual shape and depth perception, distinct illumination techniques have been proposed, adding different types of lighting effects. In this paper, we explore a new strategy to compute directional ambient occlusion and shadows for volume ray casting to render structured datasets. Our algorithm computes occlusion of traced cones by evaluating Gaussian integrals at discrete samples along the cone axis. The computed occlusion is then used to add directional ambient occlusion effects and to generate shadows. Given the extinction coefficient data volume, we create one extra volume computing representative amplitudes of Gaussian functions. Mipmapping is then used to effectively evaluate Gaussian integrals with different sizes placed along the cone axis, adapting a circle packing approach. We demonstrate that the proposed method delivers a better balance between quality results and performance when compared to previous specialized procedures, with the advantage of combining directional ambient occlusion and shadow generation under the same framework.

RayCaching: Amortized Isosurface Rendering for Virtual Reality

AbstractReal‐time virtual reality requires efficient rendering methods to deal with high‐ resolution stereoscopic displays and low latency head‐tracking. Our proposed RayCaching method renders isosurfaces of large volume datasets by amortizing raycasting over several frames and caching primary rays as small bricks that can be efficiently rasterized. An occupancy map in form of a clipmap provides level of detail and ensures that only bricks corresponding to visible points on the isosurface are being cached and rendered. Hard shadows and ambient occlusion from secondary rays are also accumulated and stored in the cache. Our method supports real‐time isosurface rendering with dynamic isovalue and allows stereoscopic visualization and exploration of large volume datasets at framerates suitable for virtual reality applications.

Effects of cropping, smoothing, triangle count, and mesh resolution on 6 dental topographic metrics.

Dental topography is a widely used method for quantifying dental morphology and inferring dietary ecology in animals. Differences in methodology have brought into question the comparability of different studies. Using primate mandibular second molars, we investigated the effects of mesh preparation parameters smoothing, cropping, and triangle count/mesh resolution (herein, resolution) on six topographic variables (Dirichlet normal energy, DNE; orientation patch count rotated, OPCR; relief index, RFI; ambient occlusion, portion de ciel visible, PCV; enamel surface area, SA; tooth size) to determine the effects of smoothing, cropping, and triangle count/resolution on topographic values and the relationship between these values and diet. All topographic metrics are sensitive to smoothing, cropping method, and triangle count/resolution. In general, smoothing decreased DNE, OPCR, RFI, and SA, increased PCV, and had no predictable effect on tooth size. Relative to the basin cut off (BCO) cropping method, the entire enamel cap (EEC) method increased RFI, SA, and size, and had no predictable effect on DNE and OPCR. Smoothing and cropping affected DNE/OPCR and surfaces with low triangle counts more than other metrics and surfaces with high triangle counts. There was a positive correlation between DNE/OPCR and triangle count/resolution, and the rate of increase was weakly correlated to diet. PCV tended to converge or decrease with increases in triangle count/resolution, and RFI, SA, and size converged. Finally, there appears to be no optimal triangle count or resolution for predicting diet from this sample, and constant triangle count appeared to perform better than constant resolution for predicting diet.

Ambient occlusion and PCV (portion de ciel visible): A new dental topographic metric and proxy of morphological wear resistance.

Recently, ambient occlusion, quantified through portion de ciel visible (PCV) was introduced as a method for quantifying dental morphological wear resistance and reconstructing diet in mammals. Despite being used to reconstruct diet and investigate the relationship between dental form and function, no rigorous analysis has investigated the correlation between PCV and diet. Using a sample of platyrrhine and prosimians M2s, we show average PCV was significantly different between most dietary groups. In prosimian, insectivores had the lowest PCV, followed by folivores, omnivores, frugivores, and finally hard-object feeders. In platyrrhines, omnivores had the lowest average PCV, followed by folivores, frugivores, and finally hard-object feeders. PCV was correlated to two topographic variables (Dirichlet normal energy, DNE, and relief index, RFI) but uncorrelated to three others (orientation patch count rotated, OPCR, tooth surface area, and tooth size). The OPCR values here differed greatly from previously published values using the same sample, showing how differences in data acquisition (i.e., using 2.5D vs. 3D surfaces) can lead to drastic differences in results. Compared to other popular topographic variables, PCV performed as well or better at predicting diet in these groups, and when combined with a metric for size, the percent of successful dietary classifications reached 90%. Further, using an ontogenetic series of hominin (Paranthropus robustus) M2s, we show that PCV correlates well with probability of wear, with PCV values being higher on the portions of the occlusal surface that experience more wear (e.g., cusps and crest tips, wear facets) than the portions of the tooth that experience less. This relationship is strongest once wear facets have begun to form on the occlusal surface. These results highlight the usefulness of PCV in quantifying morphological wear resistance and predicting diet in mammals.

Real-Time Computed Tomography Volume Visualization with Ambient Occlusion of Hand-Drawn Transfer Function Using Local Vicinity Statistic.

ObjectivesIn this paper, we present an efficient method to visualize computed tomography (CT) datasets using ambient occlusion, which is a global illumination technique that adds depth cues to the output image. We can change the transfer function (TF) for volume rendering and generate output images in real time.MethodsIn preprocessing, the mean and standard deviation of each local vicinity are calculated. During rendering, the ambient light intensity is calculated. The calculation is accelerated on the assumption that the CT value of the local vicinity of each point follows the normal distribution. We approximate complex TF forms with a smaller number of connected line segments to achieve additional acceleration. Ambient occlusion is combined with the existing local illumination technique to produce images with depth in real time.ResultsWe tested the proposed method on various CT datasets using hand-drawn TFs. The proposed method enabled real-time rendering that was approximately 40 times faster than the previous method. As a result of comparing the output image quality with that of the conventional method, the average signal-to-noise ratio was approximately 40 dB, and the image quality did not significantly deteriorate.ConclusionsWhen rendering CT images with various TFs, the proposed method generated depth-sensing images in real time.

Multi-Resolution Rendering for Computationally Expensive Lighting Effects

Many lighting methods used in computer graphics such as indirect illumination can have very high computational costs and need to be approximated for real-time applications. These costs can be reduced by means of upsampling techniques which tend to introduce artifacts and affect the visual quality of the rendered image. This paper suggests a versatile approach for accelerating the rendering of screen space methods while maintaining the visual quality. This is achieved by exploiting the low frequency nature of many of these illumination methods and the geometrical continuity of the scene. First the screen space is dynamically divided into separate sub-images, then the illumination is rendered for each sub-image in an adequate resolution and finally the sub-images are put together in order to compose the final image. Therefore we identify edges in the scene and generate masks precisely specifying which part of the image is included in which sub-image. The masks therefore determine which part of the image is rendered in which resolution. A step wise upsampling and merging process then allows optically soft transitions between the different resolution levels. For this paper, the introduced multi-resolution rendering method was implemented and tested on three commonly used lighting methods. These are screen space ambient occlusion, soft shadow mapping and screen space global illumination.

Computer-Assisted Orientation and Drawing of Archaeological Pottery

Archaeologists spend considerable time orienting and drawing ceramic fragments by hand for documentation, to infer their manufacture, the nature of the discovery site and its chronology, and to develop hypotheses about commercial and cultural exchanges, social organisation, resource exploitation, and taphonomic processes. This study presents a survey of existing solutions to the time-consuming problem of orienting and drawing pottery fragments. Orientation is based on the 3D geometry of pottery models, which can now be acquired in minutes with low-cost 3D scanners. Several methods are presented: they are based on normal vectors, or circle fittings, or profile fittings. All these methods seek to determine the optimal position of the rotation axis. We also present and discuss new approaches and improvements to existing methods. We have developed a suite of functions for the computer-assisted orientation and drawing of archaeological pottery. The profile and contours of the fragment, as well as any possible decoration, can be depicted in various ways: photorealistic rendering or dotted patterns, calculated by ambient occlusion, combined or not with artificial light. The general workflow, evaluated using both synthetic and real-world fragments, is rapid, accurate, and reproducible. It drastically reduces the amount of routine work required to document ceramic artefacts. The information produced, together with the 3D representation of the fragments, can easily be archived and/or exchanged within the archaeological community for further research. The source code (built in the R environment), together with an installation notice and examples, is freely downloadable.

Embedding QR codes onto B-spline surfaces for 3D printing

Recent advance of Additive Manufacturing technologies allows us to manufacture various parts used in real-world products. Consequently, product tracking of such 3D printed parts is an important issue. Quick Response (QR) code which is a two-dimensional matrix barcode invented by Denso, a Japanese automotive industry, in 1994, can be used for this purpose. It can store more data than the 1D barcode in a smaller space, and using a smartphone as a scanner, one can directly visit a website where all the information of the parts is stored. However, QR codes require secondary procedures to add them to products and are also vulnerable to wear and tear. Moreover, QR codes cannot be added to freeform surfaces, but only to developable surfaces. In this paper we propose a novel technique to embed QR codes onto CAD models consisting of freeform surfaces represented by B-spline surfaces, which produces 3D QR codes. 3D QR codes work similar to 2D QR codes and can be read by existing QR scanners, but are designed by grooving the surface to obtain light and dark regions caused by ambient occlusion. Unlike conventional QR codes, 3D QR codes do not fall off from the part and can even be painted if necessary. Furthermore, we do not need to prepare dark-colored and light-colored materials for 3D printing as the dark color is provided by the grooving. We demonstrate the effectiveness of our technique with various examples.

A Voxel-Based Rendering Pipeline for Large 3D Line Sets.

We present a voxel-based rendering pipeline for large 3D line sets that employs GPU ray-casting to achieve scalable rendering including transparency and global illumination effects. Even for opaque lines we demonstrate superior rendering performance compared to GPU rasterization of lines, and when transparency is used we can interactively render amounts of lines that are infeasible to be rendered via rasterization. We propose a direction-preserving encoding of lines into a regular voxel grid, along with the quantization of directions using face-to-face connectivity in this grid. On the regular grid structure, parallel GPU ray-casting is used to determine visible fragments in correct visibility order. To enable interactive rendering of global illumination effects like low-frequency shadows and ambient occlusions, illumination simulation is performed during ray-casting on a level-of-detail (LoD) line representation that considers the number of lines and their lengths per voxel. In this way we can render effects which are very difficult to render via GPU rasterization. A detailed performance and quality evaluation compares our approach to rasterization-based rendering of lines.

Matching modality in L2 pragmatics research design

• The ambient occlusion principle allows reproducible pore segmentations in 3D objects. • Ambient occlusion-derived data fields improve the separation of connected pores. • Objects can be spatially characterized based on ambient occlusion-derived data fields.

Ambient occlusion – A powerful algorithm to segment shell and skeletal intrapores in computed tomography data

During the last decades, X-ray (micro-)computed tomography has gained increasing attention for the description of porous skeletal and shell structures of various organism groups. However, their quantitative analysis is often hampered by the difficulty to discriminate cavities and pores within the object from the surrounding region.Herein, we test the ambient occlusion (AO) algorithm and newly implemented optimisations for the segmentation of cavities (implemented in the software Amira). The segmentation accuracy is evaluated as a function of (i) changes in the ray length input variable, and (ii) the usage of AO (scalar) field and other AO-derived (scalar) fields. The results clearly indicate that the AO field itself outperforms all other AO-derived fields in terms of segmentation accuracy and robustness against variations in the ray length input variable. The newly implemented optimisations improved the AO field-based segmentation only slightly, while the segmentations based on the AO-derived fields improved considerably.Additionally, we evaluated the potential of the AO field and AO-derived fields for the separation and classification of cavities as well as skeletal structures by comparing them with commonly used distance-map-based segmentations. For this, we tested the zooid separation within a bryozoan colony, the stereom classification of an ophiuroid tooth, the separation of bioerosion traces within a marble block and the calice (central cavity)-pore separation within a dendrophyllid coral. The obtained results clearly indicate that the ideal input field depends on the three-dimensional morphology of the object of interest. The segmentations based on the AO-derived fields often provided cavity separations and skeleton classifications that were superior to or impossible to obtain with commonly used distance-map-based segmentations. The combined usage of various AO-derived fields by supervised or unsupervised segmentation algorithms might provide a promising target for future research to further improve the results for this kind of high-end data segmentation and classification. Furthermore, the application of the developed segmentation algorithm is not restricted to X-ray (micro-)computed tomographic data but may potentially be useful for the segmentation of 3D volume data from other sources.