.png){kind=icon}
Return
Matrix representation and GPU-optimized parallel B-spline computing
.png){kind=icon}
PRE
AI
DOI：10.1016/j.cad.2025.103948.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A novel matrix representation method that can reduce all B-spline computations to matrix addition and multiplication, which is well-suitable for GPU computing. • A GPU-optimized task scheduling and memory access method that ensures efficient GPU resource utilization and maximizing parallelism. • A combined use of matrix representation and GPU optimization to resolve the structural mismatch between B-splines’ recursive nature and the parallelized architecture of GPUs.
Keywords:
B-spline
matrix representation
GPU computing
task scheduling
parallelism
Journal
C
IF:
3.1
Papers:
3.2K
Citations:
6.4K
Organization
Cited Papers
No cited papers available