Skip to content
DraconiticResearch
Search Sources API
203,948 studies31 sourcesUpdated 41 min ago
PreprintCC BY 4.0Zenodo

Gas-Expansion and Lock-In Synthesis: A Process-Guided Framework for Three-Dimensional Generation of Cellular and Porous Structures

Abstract

Three-dimensional generative modeling has advanced through diffusion-based generation, neural radiance fields, 3D Gaussian Splatting, and point-cloud generation. However, these approaches primarily differ in their representations, optimization procedures, or sampling mechanisms, whereas a broad class of real-world structures emerges through gas generation, bubble nucleation, volumetric expansion, competition for space, and subsequent solidification. Examples include polymeric foams, volcanic pumice, cellular ceramics, and other porous materials. This paper proposes Gas-Expansion and Lock-In Synthesis (GELIS), a process-guided framework for generating three-dimensional cellular and porous structures through the simulated evolution of gas-filled cavities within a deformable matrix. Rather than directly predicting a final surface or point distribution, GELIS represents the evolving object as a coupled system of gas domains, matrix material, interfaces, and local solidification states. Its proposed stages include gas-source initialization, stochastic nucleation, pressure-driven expansion, inter-cell interaction, matrix deformation, solidification-front propagation, and final geometry extraction. A unified energy-inspired objective is introduced to balance target-shape agreement, cellular morphology, process consistency, structural integrity, and computational cost. The framework can use implicit fields, explicit meshes, voxel grids, or point-based representations as alternative output backends. This separation between formation dynamics and geometric representation provides a basis for investigating a distinct process-guided approach to three-dimensional generation. The paper formalizes the framework, presents a mathematical model and implementation algorithm, and outlines experiments required to evaluate its feasibility. Because bubble-growth simulation and porous-structure generation already have substantial prior literature, GELIS is presented as a research proposal rather than an empirically validated or demonstrably unprecedented method.

Related

  1. Freezing-Front-Driven 3D Generation: A Process-Based Framework for Constructing Three-Dimensional Ice Structures Through Phase-Transition Dynamics
    Jincheng Zhang
    ZenodoPreprintCC BY 4.0Full textZenodo
  2. Gelation-Driven 3D Generation: A Network-Formation Framework for Constructing Three-Dimensional Structures Through Crosslinking and Liquid Confinement
    Jincheng Zhang
    ZenodoPreprintCC BY 4.0Full textZenodo
  3. Splitting-and-Branching 3D Generation: A Recursive Structural Growth Framework for Three-Dimensional Shape Synthesis
    Jincheng Zhang
    ZenodoPreprintCC BY 4.0Full textZenodo
  4. Phase Redistribution and Separation Generation: A Component-Transport Framework for Controllable Three-Dimensional Microstructure Synthesis
    Jincheng Zhang
    ZenodoPreprintCC BY 4.0Full textZenodo
  5. Deletion-Driven 3D Generation: A Progressive Removal Framework for Shape, Topology, and Material-Aware Three-Dimensional Synthesis
    Jincheng Zhang
    ZenodoPreprintCC BY 4.0Full textZenodo
BibTeX
@misc{zhang2026gasexpansion,
  title = {{Gas-Expansion and Lock-In Synthesis: A Process-Guided Framework for Three-Dimensional Generation of Cellular and Porous Structures}},
  author = {Jincheng Zhang},
  howpublished = {Zenodo},
  year = {2026},
  month = oct,
  doi = {10.5281/zenodo.23636148},
  url = {https://zenodo.org/records/23636148},
  note = {CC BY 4.0},
}
RIS
TY  - UNPB
TI  - Gas-Expansion and Lock-In Synthesis: A Process-Guided Framework for Three-Dimensional Generation of Cellular and Porous Structures
AU  - Jincheng Zhang
PY  - 2026
DA  - 2026/10/25
PB  - Zenodo
DO  - 10.5281/zenodo.23636148
UR  - https://zenodo.org/records/23636148
AB  - Three-dimensional generative modeling has advanced through diffusion-based generation, neural radiance fields, 3D Gaussian Splatting, and point-cloud generation. However, these approaches primarily differ in their representations, optimization procedures, or sampling mechanisms, whereas a broad class of real-world structures emerges through gas generation, bubble nucleation, volumetric expansion, competition for space, and subsequent solidification. Examples include polymeric foams, volcanic pumice, cellular ceramics, and other porous materials. This paper proposes Gas-Expansion and Lock-In Synthesis (GELIS), a process-guided framework for generating three-dimensional cellular and porous structures through the simulated evolution of gas-filled cavities within a deformable matrix. Rather than directly predicting a final surface or point distribution, GELIS represents the evolving object as a coupled system of gas domains, matrix material, interfaces, and local solidification states. Its proposed stages include gas-source initialization, stochastic nucleation, pressure-driven expansion, inter-cell interaction, matrix deformation, solidification-front propagation, and final geometry extraction. A unified energy-inspired objective is introduced to balance target-shape agreement, cellular morphology, process consistency, structural integrity, and computational cost. The framework can use implicit fields, explicit meshes, voxel grids, or point-based representations as alternative output backends. This separation between formation dynamics and geometric representation provides a basis for investigating a distinct process-guided approach to three-dimensional generation. The paper formalizes the framework, presents a mathematical model and implementation algorithm, and outlines experiments required to evaluate its feasibility. Because bubble-growth simulation and porous-structure generation already have substantial prior literature, GELIS is presented as a research proposal rather than an empirically validated or demonstrably unprecedented method.
ER  -