A Hierarchical Bounding Box Construction Method Based on Dynamic Task Scheduling
A technology of hierarchical bounding boxes and construction methods, which is applied in the directions of program startup/switching, multi-program installation, and inter-program communication, etc., which can solve problems such as high computing costs, and achieve high utilization rate, construction speed, and high frame rate. Effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0034] The technical solution adopted in the present invention is: a hierarchical bounding box construction algorithm based on dynamic task scheduling, which is characterized by the following steps:
[0035] Step 1: Initialize the global variable g_optimizecounter in the GPU, which indicates the number of tasks currently completed and running, and allocate shared memory;
[0036] Step 2: Assign tasks to threads according to the g_optimizecounter value and the number of scene triangles;
[0037] Step 3: Combining with the task ID, traverse the BVH from top to top, and read the data required for young tree reconstruction to local variables;
[0038] Step 4: When there is at least one thread in the warp, and there are 9 leaf nodes under the internal node in the thread, use warp-level programming to reconstruct the young tree with the clustering method;
[0039] Step 5: Repeat step 2 until the set end condition is reached, and the GPU outputs the calculation result;
[0040] The...
Embodiment 2
[0055] The principle of the present invention will be further described below in conjunction with Embodiment 2. please see figure 1 , the technical solution adopted in the present invention is: the i bounding box construction algorithm based on dynamic task scheduling, comprising the following steps:
[0056]Step 1: Initialize the global variables in the GPU, g_optimizecounter indicates the number of currently completed and running tasks, and allocate shared memory; among them, the allocation of shared memory is to determine the internal nodes of the young tree and the leaf nodes of the young tree according to the number of leaf nodes of the young tree , the surface area of the leaf node bounding box of the sapling and the shared memory size occupied by the distance matrix respectively.
[0057] Step 2: Assign tasks to threads according to the g_optimizecounter value and the number of scene triangles; use the built-in numerical function of the GPU to count the number of thr...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 
