ObjectMover: Generative Object Movement with Video Prior

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Yu, Xin, Wang, Tianyu, Kim, Soo Ye, Guerrero, Paul, Chen, Xi, Liu, Qing, Lin, Zhe, Qi, Xiaojuan
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866912269795852288
author Yu, Xin
Wang, Tianyu
Kim, Soo Ye
Guerrero, Paul
Chen, Xi
Liu, Qing
Lin, Zhe
Qi, Xiaojuan
author_facet Yu, Xin
Wang, Tianyu
Kim, Soo Ye
Guerrero, Paul
Chen, Xi
Liu, Qing
Lin, Zhe
Qi, Xiaojuan
contents Simple as it seems, moving an object to another location within an image is, in fact, a challenging image-editing task that requires re-harmonizing the lighting, adjusting the pose based on perspective, accurately filling occluded regions, and ensuring coherent synchronization of shadows and reflections while maintaining the object identity. In this paper, we present ObjectMover, a generative model that can perform object movement in highly challenging scenes. Our key insight is that we model this task as a sequence-to-sequence problem and fine-tune a video generation model to leverage its knowledge of consistent object generation across video frames. We show that with this approach, our model is able to adjust to complex real-world scenarios, handling extreme lighting harmonization and object effect movement. As large-scale data for object movement are unavailable, we construct a data generation pipeline using a modern game engine to synthesize high-quality data pairs. We further propose a multi-task learning strategy that enables training on real-world video data to improve the model generalization. Through extensive experiments, we demonstrate that ObjectMover achieves outstanding results and adapts well to real-world scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2503_08037
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ObjectMover: Generative Object Movement with Video Prior
Yu, Xin
Wang, Tianyu
Kim, Soo Ye
Guerrero, Paul
Chen, Xi
Liu, Qing
Lin, Zhe
Qi, Xiaojuan
Graphics
Artificial Intelligence
Computer Vision and Pattern Recognition
Simple as it seems, moving an object to another location within an image is, in fact, a challenging image-editing task that requires re-harmonizing the lighting, adjusting the pose based on perspective, accurately filling occluded regions, and ensuring coherent synchronization of shadows and reflections while maintaining the object identity. In this paper, we present ObjectMover, a generative model that can perform object movement in highly challenging scenes. Our key insight is that we model this task as a sequence-to-sequence problem and fine-tune a video generation model to leverage its knowledge of consistent object generation across video frames. We show that with this approach, our model is able to adjust to complex real-world scenarios, handling extreme lighting harmonization and object effect movement. As large-scale data for object movement are unavailable, we construct a data generation pipeline using a modern game engine to synthesize high-quality data pairs. We further propose a multi-task learning strategy that enables training on real-world video data to improve the model generalization. Through extensive experiments, we demonstrate that ObjectMover achieves outstanding results and adapts well to real-world scenarios.
title ObjectMover: Generative Object Movement with Video Prior
topic Graphics
Artificial Intelligence
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.08037