Orienting 3D-printing toolpaths to create new reflections.

ANR JCJC 2024–2028 ANR-24-CE10-6403 Computer graphics & additive manufacturing
Three photographs of an AtomSlicer 3D-printed house showing anisotropic reflections from different viewpoints
Anisotropic appearance fabricated with AtomSlicer. The local orientation of deposition toolpaths controls the reflections visible on the printed surface. The three views show how the appearance changes with viewpoint.

Project goal

ANISO develops digital tools and manufacturing processes to control the local orientation of material deposition in 3D printing. This control makes it possible to create customized anisotropic appearances—direction-dependent reflections similar to brushed metal—on planar surfaces and 3D objects using a single material, without paint or added chemical treatments.

The scientific challenge is to go from a desired spatially varying orientation to a dense, regular, continuous and manufacturable trajectory. The project combines computational geometry, computer graphics and digital fabrication. It targets fused filament fabrication (FFF) and surface brushing; current results primarily focus on non-planar and multi-axis FFF.

Résumé en français

ANISO développe des outils numériques et des procédés de fabrication permettant de contrôler localement l’orientation du dépôt de matière en impression 3D. Ce contrôle permet de créer, avec un matériau unique, des apparences anisotropes personnalisées — comparables aux reflets d’un métal brossé — sur des surfaces planes ou des objets 3D, sans peinture ni traitement chimique ajouté.

Le défi consiste à transformer une orientation souhaitée, variable dans l’espace, en trajectoires denses, régulières, continues et effectivement fabricables. Le projet combine géométrie numérique, informatique graphique et fabrication numérique.

From orientation fields to fabricated objects

1. Design orientations

Represent the desired local anisotropy as direction fields on surfaces or inside volumes.

2. Generate filling curves

Construct field-aligned curves with controlled spacing, robust stitching and scalability to complex geometries.

3. Make them manufacturable

Generate non-planar, continuous toolpaths while controlling bead geometry, nozzle accessibility and collisions.

Publications and results

The project has produced a sequence of complementary results, from real-time rendering of anisotropic materials to geometric algorithms, machine kinematics and experimentally validated toolpaths.

Wave-Guided Field-Aligned Volume-Filling Curves
Giovanni Cocco, Xavier Chermain
Computer Graphics Forum (Proceedings of the Symposium on Geometry Processing), 2026
Constructs well-spaced field-aligned volume-filling curves at scale, including outputs with more than 10 million vertices and validation on 4,398 solids.
AtomSlicer - Constant-Thickness Field-Aligned Non-Planar Slicing and Continuous Toolpaths for FFF
Giovanni Cocco, Vincent Belle, Eric Garner, Sylvain Lefebvre, Xavier Chermain
ACM Transactions on Graphics (Proceedings of SIGGRAPH), 2026
Generates near-constant-thickness field-aligned non-planar layers and continuous collision-free toolpaths, validated on 13 printed models.
Field-Aligned Surface-Filling Curve via Implicit Stitching
Giovanni Cocco, Xavier Chermain
Computer Graphics Forum (Proceedings of Eurographics), 2026
Generates robust and scalable field-aligned surface-filling curves through implicit stitching, with validation on Thingi10K.
Towards Accessible Non-Planar FFF Using Triple Z-Axis Kinematics
Giovanni Cocco, Eric Garner, Vincent Belle, Cédric Zanni, Xavier Chermain
Proceedings of the ACM Symposium on Computational Fabrication, 2025
Makes non-planar FFF more accessible through a desktop-printer architecture with three independently actuated Z axes and open-source kinematics.
Atomizer - Beyond Non-Planar Slicing for Fused Filament Fabrication
Computer Graphics Forum (Proceedings of the Symposium on Geometry Processing), 2025
SGP Honorable Mention
Moves beyond layer-based slicing using oriented volumetric “atoms” to generate flexible, collision-free FFF toolpaths, including anisotropic appearance fabrication on curved surfaces.
Anisotropic Specular Image-Based Lighting Based on BRDF Major Axis Sampling
Giovanni Cocco, Cédric Zanni, and Xavier Chermain
Computer Graphics Forum (Proceedings of Pacific Graphics), 2024
Provides real-time image-based lighting for anisotropic specular materials, improving reflection realism over bent normals through BRDF major-axis sampling.

Scientific foundation

ANISO builds on our 2023 SIGGRAPH work, Orientable Dense Cyclic Infill for Anisotropic Appearance Fabrication, which demonstrated that orienting continuous FFF deposition paths can directly control anisotropic reflections on printed surfaces. ANISO extends this idea from planar patterns toward general 3D surfaces, volumes, non-planar toolpaths and fabrication-aware algorithms.

Open-source software and prototypes

A dedicated triple-Z-axis FFF prototype was also developed from a desktop printer. By independently actuating the three Z axes, the print bed can tilt to execute non-planar trajectories with limited hardware modifications, providing an accessible platform for experimental validation.

Videos and demonstrations

The following presentation gives a broad introduction to Atomizer, why conventional slicing is restrictive, and how the project approaches non-planar toolpath generation.

People

Xavier Chermain — Principal investigator and project coordinator
Giovanni Cocco — PhD student
Eric Garner — Postdoctoral researcher
Vincent Belle — Research engineer
Sylvain Lefebvre — Advisor

The project is carried out in the MFX team at LORIA, with Université de Lorraine, CNRS and Inria, Nancy, France.

Funding

ANISO is funded by the French National Research Agency (ANR) through the JCJC — Young Researchers instrument.

  • Project: Anisotropic Appearance Fabrication With High-Resolution and Spatially Varying Orientations
  • ANR reference: ANR-24-CE10-6403
  • Period: October 2024 – October 2028
  • Call: Industry 4.0: People, Organization, Technologies

The PhD work is also co-funded by the Région Grand Est.

Agence nationale de la recherche logo

Contact

Interested in the project, its software, or potential collaborations? Contact Xavier Chermain at xavier.chermain@inria.fr.