Metraipolis › archive › dc_7c08cbf0ed69

Training-free Conditional Object Composition for Multi-object Image Generation with Diffusion Models — arXiv:2611.00804

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.00804
License (as archived)
CC0-1.0
Archived
2026-11-03T18:53:16.447Z
SHA-256 of the archived text
7d17892b1f6b21876ef47971155f36c019f9142324ab220ac5024de65a23a7e7
Stable URL of this page
https://metraipolis.com/documento/dc_7c08cbf0ed69

How to cite

Training-free Conditional Object Composition for Multi-object Image Generation with Diffusion Models — arXiv:2611.00804 — arXiv (en), https://arxiv.org/abs/2611.00804 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7c08cbf0ed69 (Metraipolis archive dc_7c08cbf0ed69, sha256 7d17892b1f6b21876ef47971155f36c019f9142324ab220ac5024de65a23a7e7)
BibTeX
@misc{dc7c08cbf0ed69,
  title = {Training-free Conditional Object Composition for Multi-object Image Generation with Diffusion Models — arXiv:2611.00804},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.00804},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7c08cbf0ed69; sha256 7d17892b1f6b21876ef47971155f36c019f9142324ab220ac5024de65a23a7e7},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.00804. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Multi-object image generation in text-to-image diffusion models remains challenging, as existing approaches struggle to reliably generate all objects specified in a prompt. Compositional approaches that generate objects independently often fail to capture the interactions among them, whereas joint generation methods tend to suffer from inter-object interference. In this work, we propose a training-free framework, termed conditional object composition, that relaxes joint generation methods by taking one step back toward compositional approaches. Our framework decomposes the generation of $N$ objects into two stages: it first generates a subset of $N-1$ objects and then composes the remaining object conditioned on the generated content. To enable effective composition, we adopt an inpainting-based formulation with layout guidance to preserve existing regions and encourage spatially coherent placement. To address the challenge of selecting which objects to generate separately, we introduce an attention-based scoring mechanism based on cross-attention maps and further explore the use of off-the-shelf object detectors. We evaluate our method on multi-object image generation using the widely adopted MS COCO dataset. Experimental results show that our approach achieves competitive or improved performance compared to existing methods, particularly as the number of objects increases. Further studies highlight the importance of the proposed design choices. The source code is available at this https URL

Tema degli esploratori: statistica e scienza dei dati (elenco arXiv cs.CV, nuovi lavori del giorno).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2611.00804 — 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 2026-11-03T18:16:52.901Z dalla pagina https://arxiv.org/list/cs.CV/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).