Multitwine: Multi-Object Compositing with Text and Layout Control

Gemma Canet Tarrés, Zhe Lin, Zhifei Zhang, He Zhang, Andrew Gilbert, John Collomosse, Soo Ye Kim; Proceedings of the Computer Vision and Pattern Recognition Conference (CVPR), 2025, pp. 8094-8104

We introduce the first generative model capable of simultaneous multi-object compositing, guided by both text and layout. Our model allows for the addition of multiple objects within a scene, capturing a range of interactions from simple positional relations (e.g., next to, in front of) to complex actions requiring reposing (e.g., hugging, playing guitar). When an interaction implies additional props, like 'taking a selfie', our model autonomously generates these supporting objects. By jointly training for compositing and subject-driven generation, also known as customization, we achieve a more balanced integration of textual and visual inputs for text-driven object compositing. As a result, we obtain a versatile model with state-of-the-art performance in both tasks. We further present a data generation pipeline leveraging visual and language models to effortlessly synthesize multimodal, aligned training data.