InFusion: Inpainting 3D Gaussians via Learning Depth Completion from Diffusion Prior

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Liu, Zhiheng, Ouyang, Hao, Wang, Qiuyu, Cheng, Ka Leong, Xiao, Jie, Zhu, Kai, Xue, Nan, Liu, Yu, Shen, Yujun, Cao, Yang
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917642681450496
author Liu, Zhiheng
Ouyang, Hao
Wang, Qiuyu
Cheng, Ka Leong
Xiao, Jie
Zhu, Kai
Xue, Nan
Liu, Yu
Shen, Yujun
Cao, Yang
author_facet Liu, Zhiheng
Ouyang, Hao
Wang, Qiuyu
Cheng, Ka Leong
Xiao, Jie
Zhu, Kai
Xue, Nan
Liu, Yu
Shen, Yujun
Cao, Yang
contents 3D Gaussians have recently emerged as an efficient representation for novel view synthesis. This work studies its editability with a particular focus on the inpainting task, which aims to supplement an incomplete set of 3D Gaussians with additional points for visually harmonious rendering. Compared to 2D inpainting, the crux of inpainting 3D Gaussians is to figure out the rendering-relevant properties of the introduced points, whose optimization largely benefits from their initial 3D positions. To this end, we propose to guide the point initialization with an image-conditioned depth completion model, which learns to directly restore the depth map based on the observed image. Such a design allows our model to fill in depth values at an aligned scale with the original depth, and also to harness strong generalizability from largescale diffusion prior. Thanks to the more accurate depth completion, our approach, dubbed InFusion, surpasses existing alternatives with sufficiently better fidelity and efficiency under various complex scenarios. We further demonstrate the effectiveness of InFusion with several practical applications, such as inpainting with user-specific texture or with novel object insertion.
format Preprint
id arxiv_https___arxiv_org_abs_2404_11613
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle InFusion: Inpainting 3D Gaussians via Learning Depth Completion from Diffusion Prior
Liu, Zhiheng
Ouyang, Hao
Wang, Qiuyu
Cheng, Ka Leong
Xiao, Jie
Zhu, Kai
Xue, Nan
Liu, Yu
Shen, Yujun
Cao, Yang
Computer Vision and Pattern Recognition
3D Gaussians have recently emerged as an efficient representation for novel view synthesis. This work studies its editability with a particular focus on the inpainting task, which aims to supplement an incomplete set of 3D Gaussians with additional points for visually harmonious rendering. Compared to 2D inpainting, the crux of inpainting 3D Gaussians is to figure out the rendering-relevant properties of the introduced points, whose optimization largely benefits from their initial 3D positions. To this end, we propose to guide the point initialization with an image-conditioned depth completion model, which learns to directly restore the depth map based on the observed image. Such a design allows our model to fill in depth values at an aligned scale with the original depth, and also to harness strong generalizability from largescale diffusion prior. Thanks to the more accurate depth completion, our approach, dubbed InFusion, surpasses existing alternatives with sufficiently better fidelity and efficiency under various complex scenarios. We further demonstrate the effectiveness of InFusion with several practical applications, such as inpainting with user-specific texture or with novel object insertion.
title InFusion: Inpainting 3D Gaussians via Learning Depth Completion from Diffusion Prior
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2404.11613