Enregistré dans:
Détails bibliographiques
Auteurs principaux: Zhou, Hang, Zuo, Xinxin, Wang, Sen, Cheng, Li
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:https://arxiv.org/abs/2603.06873
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866912950722232320
author Zhou, Hang
Zuo, Xinxin
Wang, Sen
Cheng, Li
author_facet Zhou, Hang
Zuo, Xinxin
Wang, Sen
Cheng, Li
contents Despite strong single-turn performance, diffusion-based image compositing often struggles to preserve coherent spatial relations in pairwise or sequential edits, where subsequent insertions may overwrite previously generated content and disrupt physical consistency. We introduce PICS, a self-supervised composition-by-decomposition paradigm that composes objects in parallel while explicitly modeling the compositional interactions among (fully-/partially-)visible objects and background. At its core, an Interaction Transformer employs mask-guided Mixture-of-Experts to route background, exclusive, and overlap regions to dedicated experts, with an adaptive α-blending strategy that infers a compatibility-aware fusion of overlapping objects while preserving boundary fidelity. To further enhance robustness to geometric variations, we incorporate geometry-aware augmentations covering both out-of-plane and in-plane pose changes of objects. Our method delivers superior pairwise compositing quality and substantially improved stability, with extensive evaluations across virtual try-on, indoor, and street scene settings showing consistent gains over state-of-the-art baselines. Code and data are available at https://github.com/RyanHangZhou/PICS
format Preprint
id arxiv_https___arxiv_org_abs_2603_06873
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle PICS: Pairwise Image Compositing with Spatial Interactions
Zhou, Hang
Zuo, Xinxin
Wang, Sen
Cheng, Li
Computer Vision and Pattern Recognition
Despite strong single-turn performance, diffusion-based image compositing often struggles to preserve coherent spatial relations in pairwise or sequential edits, where subsequent insertions may overwrite previously generated content and disrupt physical consistency. We introduce PICS, a self-supervised composition-by-decomposition paradigm that composes objects in parallel while explicitly modeling the compositional interactions among (fully-/partially-)visible objects and background. At its core, an Interaction Transformer employs mask-guided Mixture-of-Experts to route background, exclusive, and overlap regions to dedicated experts, with an adaptive α-blending strategy that infers a compatibility-aware fusion of overlapping objects while preserving boundary fidelity. To further enhance robustness to geometric variations, we incorporate geometry-aware augmentations covering both out-of-plane and in-plane pose changes of objects. Our method delivers superior pairwise compositing quality and substantially improved stability, with extensive evaluations across virtual try-on, indoor, and street scene settings showing consistent gains over state-of-the-art baselines. Code and data are available at https://github.com/RyanHangZhou/PICS
title PICS: Pairwise Image Compositing with Spatial Interactions
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.06873