ShapeShift: Text-to-Mosaic Synthesis via Semantic Phase-Field Guidance
Vihaan Misra, Peter Schaldenbrand, Jean Oh · 2025
Read ShapeShift: Text-to-Mosaic Synthesis via Semantic Phase-Field Guidance by Vihaan Misra, Peter Schaldenbrand, Jean Oh (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-14720.txt.
Description
Vihaan Misra, Peter Schaldenbrand, and Jean Oh presented ShapeShift in 2025, a method that arranges rigid physical objects so the arrangement itself pictures an idea named in words. A mosaic of blocks or tiles, seen from above, should read as the requested concept. Pretrained diffusion models can judge how well a layout matches a concept, but the objects must also obey physics and not overlap.
The hard part is fixing overlaps without wrecking the picture, since pushing shapes apart in the geometrically obvious direction often destroys the recognizable arrangement. ShapeShift represents the boundary as a deformable phase field that grows unevenly, steered by features inside the diffusion model that encode direction and shape. Experiments show layouts that stay physically valid while keeping the intended meaning clear.
Catalog record
| CREATOR | Vihaan Misra, Peter Schaldenbrand, Jean Oh |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 34 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-14720