{
  "abstract": "Contemporary image generation methods commonly construct visual content through learned transformations of latent variables, sequential token prediction, or iterative denoising. This paper proposes Wave Propagation Image Generation (WPIG), a framework that investigates an alternative organizational principle: generating images through the propagation and interaction of spatially distributed wave states. In WPIG, image formation begins with one or more wave sources that carry local appearance, structural, or semantic information. These sources propagate through a spatial representation, modify their states according to local conditions, interact with other waves, and contribute to the progressive formation of an image. Unlike a purely visual ripple effect, the proposed framework treats propagation as a computational mechanism for constructing image content. Its principal components include adaptive wave-source initialization, learned propagation operators, multi-wave interaction, spatially conditioned modulation, localized correction, and adaptive termination. The framework is formulated for both pixel-space and latent-space implementations and can be trained using reconstruction objectives, perceptual losses, conditional generation objectives, and optional reinforcement learning. A computational formulation and experimental protocol are provided to support systematic evaluation. The proposed method is compared conceptually with diffusion models, reaction-diffusion systems, neural cellular automata, and autoregressive image generation. No experimental superiority is assumed. Instead, the paper establishes testable hypotheses concerning generation quality, structural coherence, controllability, computational efficiency, and the contribution of individual propagation mechanisms. WPIG is intended as a research framework for investigating whether wavefront-driven information organization can provide a useful alternative or complement to existing image synthesis architectures.",
  "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/fgj8xwe0mmxcd.bib",
  "cited_by": 0,
  "doi": "10.5281/zenodo.23555966",
  "fulltext_url": null,
  "id": "fgj8xwe0mmxcd",
  "issns": [],
  "landing_url": "https://zenodo.org/records/23555966",
  "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/23555966/files/Wave%20Propagation%20Image%20Generation.pdf/content",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-22",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/fgj8xwe0mmxcd.ris",
  "sources": [
    {
      "first_seen": "2026-11-05T12:20:46Z",
      "native_id": "23555966",
      "source": "zenodo",
      "url": "https://zenodo.org/records/23555966"
    }
  ],
  "subjects": [],
  "title": "Wave Propagation Image Generation: A Framework for Wavefront-Driven Visual Synthesis Through Multi-Source Interaction, Adaptive Propagation, and Local Correction",
  "title_text": "Wave Propagation Image Generation: A Framework for Wavefront-Driven Visual Synthesis Through Multi-Source Interaction, Adaptive Propagation, and Local Correction",
  "type": "preprint",
  "updated_at": "2026-11-05T12:20:46Z",
  "url": "https://research.draconitic.com/paper/fgj8xwe0mmxcd",
  "venue": "Zenodo",
  "versions": []
}