Neural Slicer for Multi-Axis 3D Printing
Published:
Neural Slicer is a novel computational pipeline that employs neural networks to establish a deformation mapping, defining a scalar field in the space surrounding an input model. Isosurfaces are subsequently extracted from this field to generate curved layers for 3D printing.
Key Features
- Representation-agnostic: Works with diverse model representations (mesh, point cloud, implicit functions)
- Topology-independent: Handles intricate topology seamlessly without preprocessing
- Differentiable pipeline: Enables optimization through loss functions for better layer quality
- Support-free printing: Generates slicing results with improved performance and reduced material waste
Links
- GitHub Repository: Neural Slicer
- Publication: ACM TOG 2024