Learning Image Fractals Using Chaotic Differentiable Point Splatting
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909508386684928 |
|---|---|
| author | Djeacoumar, Adarsh Mujkanovic, Felix Seidel, Hans-Peter Leimkühler, Thomas |
| author_facet | Djeacoumar, Adarsh Mujkanovic, Felix Seidel, Hans-Peter Leimkühler, Thomas |
| contents | Fractal geometry, defined by self-similar patterns across scales, is crucial for understanding natural structures. This work addresses the fractal inverse problem, which involves extracting fractal codes from images to explain these patterns and synthesize them at arbitrary finer scales. We introduce a novel algorithm that optimizes Iterated Function System parameters using a custom fractal generator combined with differentiable point splatting. By integrating both stochastic and gradient-based optimization techniques, our approach effectively navigates the complex energy landscapes typical of fractal inversion, ensuring robust performance and the ability to escape local minima. We demonstrate the method's effectiveness through comparisons with various fractal inversion techniques, highlighting its ability to recover high-quality fractal codes and perform extensive zoom-ins to reveal intricate patterns from just a single image. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_17230 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Learning Image Fractals Using Chaotic Differentiable Point Splatting Djeacoumar, Adarsh Mujkanovic, Felix Seidel, Hans-Peter Leimkühler, Thomas Graphics Fractal geometry, defined by self-similar patterns across scales, is crucial for understanding natural structures. This work addresses the fractal inverse problem, which involves extracting fractal codes from images to explain these patterns and synthesize them at arbitrary finer scales. We introduce a novel algorithm that optimizes Iterated Function System parameters using a custom fractal generator combined with differentiable point splatting. By integrating both stochastic and gradient-based optimization techniques, our approach effectively navigates the complex energy landscapes typical of fractal inversion, ensuring robust performance and the ability to escape local minima. We demonstrate the method's effectiveness through comparisons with various fractal inversion techniques, highlighting its ability to recover high-quality fractal codes and perform extensive zoom-ins to reveal intricate patterns from just a single image. |
| title | Learning Image Fractals Using Chaotic Differentiable Point Splatting |
| topic | Graphics |
| url | https://arxiv.org/abs/2502.17230 |