Metraipolis › archive › dc_a2eebc37c100

Compositional SVG Generation via VLM-Driven Hierarchical Semantic Parsing — arXiv:2609.14657

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.14657
License (as archived): CC0-1.0
Archived: 2026-10-15T22:53:14.068Z
SHA-256 of the archived text: 5b0aa529b314b70ea647a965d739d9955ea081124a444b5777dea0ae9d2c092d
Stable URL of this page: https://metraipolis.com/documento/dc_a2eebc37c100

How to cite

Compositional SVG Generation via VLM-Driven Hierarchical Semantic Parsing — arXiv:2609.14657 — arXiv (en), https://arxiv.org/abs/2609.14657 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a2eebc37c100 (Metraipolis archive dc_a2eebc37c100, sha256 5b0aa529b314b70ea647a965d739d9955ea081124a444b5777dea0ae9d2c092d)

BibTeX

@misc{dca2eebc37c100,
  title = {Compositional SVG Generation via VLM-Driven Hierarchical Semantic Parsing — arXiv:2609.14657},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.14657},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a2eebc37c100; sha256 5b0aa529b314b70ea647a965d739d9955ea081124a444b5777dea0ae9d2c092d},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.14657. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

While Vision-Language Models (VLMs) excel at visual reasoning, generating structured, editable Scalable Vector Graphics (SVG) remains a fundamental challenge. Existing pipelines predominantly yield flat, semantically agnostic collections of paths, where editing a single object requires manually identifying its constituent paths. To address this, we propose a VLM-driven agentic framework for semantic compositional SVG generation. Our pipeline recursively parses visual scenes into semantic and geometric hierarchies via top-down decomposition, visual grounding, and prompt-driven amodal occlusion recovery, ensuring each component is geometrically complete. Furthermore, we introduce the Semantic SVG Benchmark with human-annotated semantic groups and novel sub-component metrics (Semantic Recall/Precision, PERE) to explicitly evaluate structural compositionality and functional editability. Experiments show that our natively predicted structures surpass the upper bounds of existing flat-generation methods in both grouping quality and editability, while maintaining state-of-the-art visual fidelity.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-15 in cs.AI).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.14657 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-15T22:29:03.707Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-15?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).