VisACD: Visibility-Based GPU-Accelerated Approximate Convex Decomposition

2026-04-05

AI Summary

This work proposes a visibility-based approximate convex decomposition method to address limitations in existing approaches, such as excessive numbers of convex parts, sensitivity to input mesh orientation, and low computational efficiency. By incorporating a visibility mechanism, the method achieves rotation equivariance, thereby significantly reducing the number of convex components and eliminating dependence on input orientation. Furthermore, it leverages GPU parallelization and efficient geometric processing techniques to substantially accelerate computation. Experimental results demonstrate that the proposed approach yields high-quality decompositions with fewer convex parts, faster runtime, and robustness to shape rotation.

Abstract

Physics-based simulation involves trade-offs between performance and accuracy. In collision detection, one trade-off is the granularity of collider geometry. Primitive-based colliders such as bounding boxes are efficient, while using the original mesh is more accurate but often computationally expensive. Approximate Convex Decomposition (ACD) methods strive for a balance of efficiency and accuracy. Prior works can produce high-quality decompositions but require large numbers of convex parts and are sensitive to the orientation of the input mesh. We address these weaknesses with VisACD, a visibility-based, rotation-equivariant, and intersection-free ACD algorithm with GPU acceleration. Our approach produces high-quality decompositions with fewer convex parts, is not sensitive to shape orientation, and is more efficient than prior work.

Problem

• Approximate Convex Decomposition
• collision detection
• convex decomposition
• mesh orientation
• physics-based simulation

Innovation

• visibility-based
• rotation-equivariant
• intersection-free
• GPU-accelerated
• Approximate Convex Decomposition