{
  "abstract": "Fractal structures exhibit recurring patterns across multiple spatial scales, providing a useful foundation for organizing visual structures during image generation. Conventional fractal generation methods commonly rely on predefined mathematical functions, fixed recursive rules, or parameterized geometric transformations. Although these approaches can produce complex self-similar patterns, they do not necessarily provide a unified mechanism for adaptively controlling where structural refinement occurs, how deeply recursion proceeds, and how local self-similarity interacts with global image coherence and computational cost. This paper proposes Recursive Self-Similar Image Generation (RSSIG), a conceptual framework that organizes image synthesis through hierarchical recursive decomposition, scale-dependent structural transformations, controlled variation, and adaptive termination. The framework represents an image as a collection of recursively related visual structures whose positions, scales, appearances, and refinement states can evolve throughout generation. A multiscale objective is formulated to balance self-similarity, target consistency, structural coherence, diversity, and computational expenditure. An optional adaptive controller selects among splitting, rendering, refining, merging, and terminating local structures according to their estimated benefits and costs. The framework is designed to support both procedural image synthesis and extensions involving learned generative models or reinforcement learning. This paper establishes the conceptual architecture, mathematical formulation, algorithmic procedure, evaluation methodology, and experimental hypotheses of RSSIG. Because the proposed framework has not yet been empirically validated, its effectiveness and novelty relative to existing recursive, fractal, and multiscale generation methods remain open research questions.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": null,
  "author_total": 1,
  "authors": [
    {
      "name": "Jincheng Zhang",
      "orcid": "0009-0005-1860-0009"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/af64r9bc75ajt.bib",
  "cited_by": 0,
  "doi": "10.5281/zenodo.23556132",
  "fulltext_url": null,
  "id": "af64r9bc75ajt",
  "issns": [],
  "landing_url": "https://zenodo.org/records/23556132",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": null,
  "native_subjects": [],
  "openalex": null,
  "pdf_url": "https://zenodo.org/api/records/23556132/files/Recursive%20Self-Similar%20Image%20Generation%20An%20Adaptive%20Framework%20for%20Multiscale%20Visual%20Synthesis.pdf/content",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-22",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/af64r9bc75ajt.ris",
  "sources": [
    {
      "first_seen": "2026-11-05T12:20:44Z",
      "native_id": "23556132",
      "source": "zenodo",
      "url": "https://zenodo.org/records/23556132"
    }
  ],
  "subjects": [],
  "title": "Recursive Self-Similar Image Generation: An Adaptive Framework for Multiscale Visual Synthesis",
  "title_text": "Recursive Self-Similar Image Generation: An Adaptive Framework for Multiscale Visual Synthesis",
  "type": "preprint",
  "updated_at": "2026-11-05T12:20:44Z",
  "url": "https://research.draconitic.com/paper/af64r9bc75ajt",
  "venue": "Zenodo",
  "versions": []
}