Canvas-to-Image: Compositional Image Generation with Multimodal Controls
Abstract
Canvas-to-Image is a unified framework that encodes diverse control signals into a composite canvas image for high-fidelity multimodal image generation, outperforming existing methods in various benchmarks.
While modern diffusion models excel at generating high-quality and diverse images, they still struggle with high-fidelity compositional and multimodal control, particularly when users simultaneously specify text prompts, subject references, spatial arrangements, pose constraints, and layout annotations. We introduce Canvas-to-Image, a unified framework that consolidates these heterogeneous controls into a single canvas interface, enabling users to generate images that faithfully reflect their intent. Our key idea is to encode diverse control signals into a single composite canvas image that the model can directly interpret for integrated visual-spatial reasoning. We further curate a suite of multi-task datasets and propose a Multi-Task Canvas Training strategy that optimizes the diffusion model to jointly understand and integrate heterogeneous controls into text-to-image generation within a unified learning paradigm. This joint training enables Canvas-to-Image to reason across multiple control modalities rather than relying on task-specific heuristics, and it generalizes well to multi-control scenarios during inference. Extensive experiments show that Canvas-to-Image significantly outperforms state-of-the-art methods in identity preservation and control adherence across challenging benchmarks, including multi-person composition, pose-controlled composition, layout-constrained generation, and multi-control generation.
Community
While modern diffusion models excel at generating high-quality and diverse images, they still struggle with high-fidelity compositional and multimodal control. We introduce Canvas-to-Image, a unified framework that consolidates these heterogeneous controls into a single canvas interface. Our key idea is to encode diverse control signals, including subject references, bounding boxes, and pose skeletons, into a single composite canvas image that the model can directly interpret for integrated visual-spatial reasoning.
where can we try the demo??
How is the image consistency, e.g. with successive images (video)?
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- LayerComposer: Multi-Human Personalized Generation via Layered Canvas (2025)
- ConsistCompose: Unified Multimodal Layout Control for Image Composition (2025)
- SIGMA-GEN: Structure and Identity Guided Multi-subject Assembly for Image Generation (2025)
- ContextGen: Contextual Layout Anchoring for Identity-Consistent Multi-Instance Generation (2025)
- Query-Kontext: An Unified Multimodal Model for Image Generation and Editing (2025)
- UniVideo: Unified Understanding, Generation, and Editing for Videos (2025)
- iMontage: Unified, Versatile, Highly Dynamic Many-to-many Image Generation (2025)
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment:
@librarian-bot
recommend
This needs added to ComfyUI asap!
Where is download code for Comfy node?
Models citing this paper 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper