Using Hardware Ray Transforms to Accelerate Ray/Primitive Intersections for Long, Thin Primitive Types

2020 ◽  
Vol 3 (2) ◽  
pp. 1-16
Author(s):  
Ingo Wald ◽  
Nate Morrical ◽  
Stefan Zellmann ◽  
Lei Ma ◽  
Will Usher ◽  
...  
Keyword(s):  
Ray Tracing ◽  
Data Sets ◽  
Bounding Box ◽  
Bounding Volume ◽  
Recent Addition ◽  
Bounding Boxes ◽  
Ray Transforms ◽  
Bounding Volume Hierarchies

With the recent addition of hardware ray tracing capabilities, GPUs have become incredibly efficient at ray tracing both triangular geometry, and instances thereof. However, the bounding volume hierarchies that current ray tracing hardware relies on are known to struggle with long, thin primitives like cylinders and curves, because the axis-aligned bounding boxes that these hierarchies rely on cannot tightly bound such primitives. In this paper, we evaluate the use of RTX ray tracing capabilities to accelerate these primitives by tricking the GPU's instancing units into executing a hardware-accelerated oriented bounding box (OBB) rejection test before calling the user's intersection program. We show that this can be done with minimal changes to the intersection programs and demonstrate speedups of up to 5.9× on a variety of data sets.

Hardware-Accelerated Dual-Split Trees

2020 ◽  
Vol 3 (2) ◽  
pp. 1-21
Author(s):  
Daqi Lin ◽  
Elena Vasiou ◽  
Cem Yuksel ◽  
Daniel Kopta ◽  
Erik Brunvand
Keyword(s):  
Ray Tracing ◽  
Hardware Acceleration ◽  
Memory Storage ◽  
Compact Representation ◽  
Space Partitioning ◽  
Data Movement ◽  
Bounding Volume ◽  
Bounding Boxes ◽  
Split Trees ◽  
Bounding Volume Hierarchies

Bounding volume hierarchies (BVH) are the most widely used acceleration structures for ray tracing due to their high construction and traversal performance. However, the bounding planes shared between parent and children bounding boxes is an inherent storage redundancy that limits further improvement in performance due to the memory cost of reading these redundant planes. Dual-split trees can create identical space partitioning as BVHs, but in a compact form using less memory by eliminating the redundancies of the BVH structure representation. This reduction in memory storage and data movement translates to faster ray traversal and better energy efficiency. Yet, the performance benefits of dual-split trees are undermined by the processing required to extract the necessary information from their compact representation. This involves bit manipulations and branching instructions which are inefficient in software. We introduce hardware acceleration for dual-split trees and show that the performance advantages over BVHs are emphasized in a hardware ray tracing context that can take advantage of such acceleration. We provide details on how the operations needed for decoding dual-split tree nodes can be implemented in hardware and present experiments in a number of scenes with different sizes using path tracing. In our experiments, we have observed up to 31% reduction in render time and 38% energy saving using dual-split trees as compared to binary BVHs representing identical space partitioning.

A Survey on Bounding Volume Hierarchies for Ray Tracing

2021 ◽  
Vol 40 (2) ◽  
pp. 683-712
Author(s):  
Daniel Meister ◽  
Shinji Ogaki ◽  
Carsten Benthin ◽  
Michael J. Doyle ◽  
Michael Guthe ◽  
...  
Keyword(s):  
Ray Tracing ◽  
Bounding Volume ◽  
Bounding Volume Hierarchies

Efficient construction of bounding volume hierarchies into a complete octree for ray tracing

2016 ◽  
Vol 27 (3-4) ◽  
pp. 358-368 ◽  
Author(s):  
Ulises Olivares ◽  
Héctor G. Rodríguez ◽  
Arturo García ◽  
Félix F. Ramos
Keyword(s):  
Ray Tracing ◽  
Bounding Volume ◽  
Efficient Construction ◽  
Bounding Volume Hierarchies

scholarly journals EFN: Field-Based Object Detection for Aerial Images

2020 ◽  
Vol 12 (21) ◽  
pp. 3630
Author(s):  
Jin Liu ◽  
Haokun Zheng
Keyword(s):  
Object Detection ◽  
Semantic Segmentation ◽  
Natural Images ◽  
Aerial Images ◽  
Aerial Image ◽  
Data Sets ◽  
Data Set ◽  
Bounding Box ◽  
Good Score ◽  
Bounding Boxes

Object detection and recognition in aerial and remote sensing images has become a hot topic in the field of computer vision in recent years. As these images are usually taken from a bird’s-eye view, the targets often have different shapes and are densely arranged. Therefore, using an oriented bounding box to mark the target is a mainstream choice. However, this general method is designed based on horizontal box annotation, while the improved method for detecting an oriented bounding box has a high computational complexity. In this paper, we propose a method called ellipse field network (EFN) to organically integrate semantic segmentation and object detection. It predicts the probability distribution of the target and obtains accurate oriented bounding boxes through a post-processing step. We tested our method on the HRSC2016 and DOTA data sets, achieving mAP values of 0.863 and 0.701, respectively. At the same time, we also tested the performance of EFN on natural images and obtained a mAP of 84.7 in the VOC2012 data set. These extensive experiments demonstrate that EFN can achieve state-of-the-art results in aerial image tests and can obtain a good score when considering natural images.

Efficient Hair Rendering with a GPU Cone Tracing Approach

2017 ◽  
Vol 8 (1) ◽  
pp. 1-19
Author(s):  
Jorge R. Martins ◽  
Vasco S. Costa ◽  
João M. Pereira
Keyword(s):  
Human Hair ◽  
State Of The Art ◽  
Sampling Rate ◽  
Current State ◽  
Bounding Volume ◽  
Bounding Boxes ◽  
Bounding Volume Hierarchies ◽  
Bounding Volume Hierarchy

Rendering human hair can be a hard task because of the required high super-sampling rate to render thin hair fibers without noticeable aliasing. Additionally, the current state-of-the-art bounding volume hierarchies (BVHs) are not suitable to hair rendering. In fact, the axis-aligned bounding boxes (AABBs) do not tightly bind hair primitives which impacts negatively the intersection tests activity. Both limitations can degrade severely the rendering performance so described in this article, a cone tracing GPU approach coupled with a hybrid bounding volume hierarchy to tackle these problems. The hybrid BVH makes use of both oriented and axis aligned bounding boxes. It is shown that the experiment is able to drastically reduce the super-sampling required to produce aliasing free images while minimizing the number of intersection tests and achieving speedups of up to 4, depending on the scene.

scholarly journals Dynamic Garment Simulation based on Hybrid Bounding Volume Hierarchy

2016 ◽  
Vol 16 (4) ◽  
pp. 241-249 ◽  
Author(s):  
Dongyong Zhu ◽  
Zhong Li ◽  
Feng Xia ◽  
Yong Xu
Keyword(s):  
Human Body ◽  
Least Squares Method ◽  
Elliptic Cylinder ◽  
Human Body Model ◽  
Mass Model ◽  
Bounding Box ◽  
Bounding Volume ◽  
Simulation Based ◽  
Bounding Boxes ◽  
Bounding Volume Hierarchy

Abstract In order to solve the computing speed and efficiency problem of existing dynamic clothing simulation, this paper presents a dynamic garment simulation based on a hybrid bounding volume hierarchy. It firstly uses MCASG graph theory to do the primary segmentation for a given three-dimensional human body model. And then it applies K-means cluster to do the secondary segmentation to collect the human body’s upper arms, lower arms, upper legs, lower legs, trunk, hip and woman’s chest as the elementary units of dynamic clothing simulation. According to different shapes of these elementary units, it chooses the closest and most efficient hybrid bounding box to specify these units, such as cylinder bounding box and elliptic cylinder bounding box. During the process of constructing these bounding boxes, it uses the least squares method and slices of the human body to get the related parameters. This approach makes it possible to use the least amount of bounding boxes to create close collision detection regions for the appearance of the human body. A spring-mass model based on a triangular mesh of the clothing model is finally constructed for dynamic simulation. The simulation result shows the feasibility and superiority of the method described.

Ray tracing deformable scenes using dynamic bounding volume hierarchies

2007 ◽  
Vol 26 (1) ◽  
pp. 6 ◽  
Author(s):  
Ingo Wald ◽  
Solomon Boulos ◽  
Peter Shirley
Keyword(s):  
Ray Tracing ◽  
Bounding Volume ◽  
Bounding Volume Hierarchies

scholarly journals Constructing K-d Tree on GPU through Treelet Restructuring

2020 ◽  
pp. short56-1-short56-8
Author(s):  
Vadim Bulavintsev ◽  
Dmitry Zhdanov
Keyword(s):  
Ray Tracing ◽  
Graphics Processing Units ◽  
Global Illumination ◽  
Mapping Method ◽  
Bounding Volume ◽  
Illumination Algorithms ◽  
Graphics Processing ◽  
New Generation ◽  
Bounding Volume Hierarchies ◽  
Acceleration Structure

With every new generation of graphics processing units (GPUs), offloading ray-tracing algorithms to GPUs becomes more feasible. Software-hardware solutions for ray-tracing focus on implementing its basic components, such as building and traversing bounding volume hierarchies (BVH). However, global illumination algorithms, such as photon mapping method, depend on another kind of acceleration structure, namely k-d trees. In this work, we adapt state-ofthe-art GPU-based BVH-building algorithm of treelet restructuring to k-d trees. By evaluating the performance of the resulting k-d tree, we show that treelet optimisation heuristic suitable for BVHs of triangles is inadequate for k-d trees of points.

Sign in / Sign up

Export Citation Format

Share Document