From Density to Geometry: YOLOv8 Instance Segmentation for Reverse Engineering of Optimized Structures

arXiv:2404.18763v1 Announce Type: new
Abstract: This paper introduces YOLOv8-TO, a novel approach for reverse engineering of topology-optimized structures into interpretable geometric parameters using the YOLOv8 instance segmentation model. Density-based topology optimization methods require post-processing to convert the optimal density distribution into a parametric representation for design exploration and integration with CAD tools. Traditional methods such as skeletonization struggle with complex geometries and require manual intervention. YOLOv8-TO addresses these challenges by training a custom YOLOv8 model to automatically detect and reconstruct …

abstract arxiv cs.ce cs.cv design distribution engineering exploration geometry instance novel optimization paper parameters parametric post-processing processing representation segmentation topology type yolov8