Hybrid quantum classical GAN generates CAD vector shapes efficiently
HMB-GAN: Hybrid Multi-Bézier GAN for Vector Shape Synthesis
Machine Learning
Summary
Creating precise digital shapes for design and manufacturing is challenging. This work uses a new kind of AI model called HMB-GAN to generate smooth, closed shapes made of connected curves. The authors also test a version that uses both quantum and classical computing to improve training speed and reduce complexity. Although the quantum approach shows promise, current hardware limits prevent full benefits, and classical simulation is slow. Overall, their method shows that combining quantum and classical techniques can work for designing computer graphics shapes but needs better hardware.
What this means in practice
- •For cad software developers: Design AI tools that generate smooth vector shapes ready for computer-aided design workflows using multi-segment Bézier curves.
- •For quantum computing engineers: Optimize hybrid quantum-classical models for geometric data generation to explore advantages in model size reduction and faster training.
Authors
Elian Hugh Thiele-Evans, Binh Duong Pham, Hani Omar M Alharbi, Liibaan Aaden, Syed Umer Hasnain Zaidi, Prem Prakash Jayaraman, Muhammad Saeed, Boris Eisenbart
Abstract
We explore the use of hybrid quantum-classical generative adversarial networks for synthesising CAD-ready vector geometries. Unlike prior work that operates in rasterised or single-Bézier domains, we introduce HMB-GAN (Hybrid Multi-Bézier GAN), an end-to-end differentiable generative framework that constructs closed shapes through stitched multi-segment Bézier representations with geometric continuity enforced by construction. We compare a quantum-enhanced generator with a classical generator within this architecture and evaluate them across point cloud distribution metrics and geometric shape statistics. Results show that despite faster convergence, a reduction in model parameter count, and slightly improved performance on point cloud metrics, the quantum generator suffers from excessive simulator overhead and thus classically-simulated evaluation suffers from hardware constraints. These results demonstrate the feasibility of modelling structured geometries through hybrid quantum architectures whilst highlighting contemporary hardware limitations.