Placing Any Object at Any 3D Position
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (48) · 41742-41744
DOI: 10.1609/aaai.v40i48.42400 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Junhao Zhang und 6 weiteren Autor:innen.
Abstract (Originalsprache)
In this work, we propose a diffusion-based method for 3D-aware image composition. Previous approaches have focused on 2D-view image composition, which limits their handling of complex 3D spatial relationships. Consequently, they are not well-suited for applications requiring precise 3D object control and iterative refinement, including interior design visualization, visual effects prototyping, and virtual reality scene construction. In contrast, our method extracts 3D bounding boxes for all objects in the scene image. Users can then specify a new 3D bounding box based on existing spatial context and provide an image of the target object. Leveraging a fine-tuned diffusion model, our approach enables high-fidelity image composition while preserving the underlying 3D structure of the scene.
Zitieren
APA
Zhang, J., Kong, M., Hu, Z., Qin, H., Xu, Z., Zhu, X., & Zhu, Q. (2026). Placing Any Object at Any 3D Position. Proceedings of the AAAI Conference on Artificial Intelligence, 40(48), 41742-41744. https://doi.org/10.1609/aaai.v40i48.42400
Vancouver
Zhang J, Kong M, Hu Z, Qin H, Xu Z, Zhu X, et al. Placing Any Object at Any 3D Position. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(48):41742-41744. doi:10.1609/aaai.v40i48.42400
BibTeX
@article{Zhang2026,
author = {Zhang, Junhao and Kong, Ming and Hu, Zhanbin and Qin, Hao and Xu, Zhijie and Zhu, Xiaojun and Zhu, Qiang},
title = {Placing Any Object at Any 3D Position},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {48},
pages = {41742--41744},
doi = {10.1609/aaai.v40i48.42400}
}
EUVIMED