LayerComposer: Multi-Human Personalized Generation via Layered Canvas
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911284246609920 |
|---|---|
| author | Qian, Guocheng Gordon Zhang, Ruihang Chen, Tsai-Shien Dalva, Yusuf Goyal, Anujraaj Argo Menapace, Willi Skorokhodov, Ivan Dong, Meng Sahni, Arpit Ostashev, Daniil Hu, Ju Tulyakov, Sergey Wang, Kuan-Chieh Jackson |
| author_facet | Qian, Guocheng Gordon Zhang, Ruihang Chen, Tsai-Shien Dalva, Yusuf Goyal, Anujraaj Argo Menapace, Willi Skorokhodov, Ivan Dong, Meng Sahni, Arpit Ostashev, Daniil Hu, Ju Tulyakov, Sergey Wang, Kuan-Chieh Jackson |
| contents | Despite their impressive visual fidelity, existing personalized image generators lack interactive control over spatial composition and scale poorly to multiple humans. To address these limitations, we present LayerComposer, an interactive and scalable framework for multi-human personalized generation. Inspired by professional image-editing software, LayerComposer provides intuitive reference-based human injection, allowing users to place and resize multiple subjects directly on a layered digital canvas to guide personalized generation. The core of our approach is the layered canvas, a novel representation where each subject is placed on a distinct layer, enabling interactive and occlusion-free composition. We further introduce a transparent latent pruning mechanism that improves scalability by decoupling computational cost from the number of subjects, and a layerwise cross-reference training strategy that mitigates copy-paste artifacts. Extensive experiments demonstrate that LayerComposer achieves superior spatial control, coherent composition, and identity preservation compared to state-of-the-art methods in multi-human personalized image generation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_20820 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | LayerComposer: Multi-Human Personalized Generation via Layered Canvas Qian, Guocheng Gordon Zhang, Ruihang Chen, Tsai-Shien Dalva, Yusuf Goyal, Anujraaj Argo Menapace, Willi Skorokhodov, Ivan Dong, Meng Sahni, Arpit Ostashev, Daniil Hu, Ju Tulyakov, Sergey Wang, Kuan-Chieh Jackson Computer Vision and Pattern Recognition Despite their impressive visual fidelity, existing personalized image generators lack interactive control over spatial composition and scale poorly to multiple humans. To address these limitations, we present LayerComposer, an interactive and scalable framework for multi-human personalized generation. Inspired by professional image-editing software, LayerComposer provides intuitive reference-based human injection, allowing users to place and resize multiple subjects directly on a layered digital canvas to guide personalized generation. The core of our approach is the layered canvas, a novel representation where each subject is placed on a distinct layer, enabling interactive and occlusion-free composition. We further introduce a transparent latent pruning mechanism that improves scalability by decoupling computational cost from the number of subjects, and a layerwise cross-reference training strategy that mitigates copy-paste artifacts. Extensive experiments demonstrate that LayerComposer achieves superior spatial control, coherent composition, and identity preservation compared to state-of-the-art methods in multi-human personalized image generation. |
| title | LayerComposer: Multi-Human Personalized Generation via Layered Canvas |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2510.20820 |