Drying-Induced Fracture Generation (DIFG): A Stress-Accumulation Framework for Three-Dimensional Shape Synthesis Through Moisture Loss, Shrinkage, and Structural Separation
Abstract
Three-dimensional generative models have achieved substantial progress in synthesizing geometric structures, yet the mechanisms underlying the formation of many natural and material surfaces remain distinct from conventional geometry-first generation. Drying, shrinkage, and fracture provide an alternative perspective in which three-dimensional shapes emerge through the progressive loss of moisture, spatially nonuniform contraction, stress accumulation, threshold-triggered cracking, crack propagation, and structural separation. This paper proposes Drying-Induced Fracture Generation (DIFG), a conceptual framework for three-dimensional generative computing that treats drying-induced mechanical evolution as an explicit shape-generation process rather than merely as a source of surface texture.
DIFG represents an evolving object through a moisture field, a shrinkage field, a mechanical stress state, a damage field, and a dynamically updated geometric representation. Its central computational mechanism consists of five coupled operations: moisture-dependent contraction, stress accumulation, fracture-threshold evaluation, crack propagation, and geometry reconstruction. Local material heterogeneity, boundary constraints, drying gradients, and fracture-history dependence jointly determine the resulting morphology. The framework additionally incorporates warping, fragment formation, and multiscale surface deformation.
DIFG is designed to operate with mesh-based geometry, point clouds, signed distance fields, neural implicit representations, and differentiable rendering representations. It can also be integrated with three-dimensional diffusion models or other learned generative systems. The proposed formulation establishes a research direction in which physical transformation becomes a computational mechanism for synthesizing geometry. This paper defines the conceptual architecture, mathematical formulation, generation algorithm, learning objectives, evaluation protocol, and potential applications. No experimental validation is claimed; the proposed method requires implementation and empirical comparison to establish its effectiveness and novelty.
Related
- Phase Redistribution and Separation Generation: A Component-Transport Framework for Controllable Three-Dimensional Microstructure Synthesis
- Splitting-and-Branching 3D Generation: A Recursive Structural Growth Framework for Three-Dimensional Shape Synthesis
- Phase-Transition-Driven 3D Generative Computing: A State-Evolution Framework for Constructing Three-Dimensional Geometry through Nucleation, Crystallization, Solidification, and Phase Separation
- Thermally Induced Structural Transformation in a Two-Dimensional Conductive Metal-Organic Framework
- The High-Dimensional Geometric Probability System (HDGPS): A Deductive Framework Unifying Curvature Concentration Duality, Statistical–Geometric Separation, and Statistical-Dimension Phase Transitions within the ZFC Axiomatic Bridge
BibTeX
@misc{zhang2026dryinginduced,
title = {{Drying-Induced Fracture Generation (DIFG): A Stress-Accumulation Framework for Three-Dimensional Shape Synthesis Through Moisture Loss, Shrinkage, and Structural Separation}},
author = {Jincheng Zhang},
howpublished = {Zenodo},
year = {2026},
month = oct,
doi = {10.5281/zenodo.23636001},
url = {https://zenodo.org/records/23636001},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Drying-Induced Fracture Generation (DIFG): A Stress-Accumulation Framework for Three-Dimensional Shape Synthesis Through Moisture Loss, Shrinkage, and Structural Separation
AU - Jincheng Zhang
PY - 2026
DA - 2026/10/25
PB - Zenodo
DO - 10.5281/zenodo.23636001
UR - https://zenodo.org/records/23636001
AB - Three-dimensional generative models have achieved substantial progress in synthesizing geometric structures, yet the mechanisms underlying the formation of many natural and material surfaces remain distinct from conventional geometry-first generation. Drying, shrinkage, and fracture provide an alternative perspective in which three-dimensional shapes emerge through the progressive loss of moisture, spatially nonuniform contraction, stress accumulation, threshold-triggered cracking, crack propagation, and structural separation. This paper proposes Drying-Induced Fracture Generation (DIFG), a conceptual framework for three-dimensional generative computing that treats drying-induced mechanical evolution as an explicit shape-generation process rather than merely as a source of surface texture. DIFG represents an evolving object through a moisture field, a shrinkage field, a mechanical stress state, a damage field, and a dynamically updated geometric representation. Its central computational mechanism consists of five coupled operations: moisture-dependent contraction, stress accumulation, fracture-threshold evaluation, crack propagation, and geometry reconstruction. Local material heterogeneity, boundary constraints, drying gradients, and fracture-history dependence jointly determine the resulting morphology. The framework additionally incorporates warping, fragment formation, and multiscale surface deformation. DIFG is designed to operate with mesh-based geometry, point clouds, signed distance fields, neural implicit representations, and differentiable rendering representations. It can also be integrated with three-dimensional diffusion models or other learned generative systems. The proposed formulation establishes a research direction in which physical transformation becomes a computational mechanism for synthesizing geometry. This paper defines the conceptual architecture, mathematical formulation, generation algorithm, learning objectives, evaluation protocol, and potential applications. No experimental validation is claimed; the proposed method requires implementation and empirical comparison to establish its effectiveness and novelty.
ER -