Saved in:
Bibliographic Details
Main Authors: Liu, Zhixuan, Zhu, Haokun, Chen, Rui, Francis, Jonathan, Hwang, Soonmin, Zhang, Ji, Oh, Jean
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.13816
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917066041196544
author Liu, Zhixuan
Zhu, Haokun
Chen, Rui
Francis, Jonathan
Hwang, Soonmin
Zhang, Ji
Oh, Jean
author_facet Liu, Zhixuan
Zhu, Haokun
Chen, Rui
Francis, Jonathan
Hwang, Soonmin
Zhang, Ji
Oh, Jean
contents We introduce a diffusion-based approach for generating privacy-preserving digital twins of multi-room indoor environments from depth images only. Central to our approach is a novel Multi-view Overlapped Scene Alignment with Implicit Consistency (MOSAIC) model that explicitly considers cross-view dependencies within the same scene in the probabilistic sense. MOSAIC operates through a multi-channel inference-time optimization that avoids error accumulation common in sequential or single-room constraints in panorama-based approaches. MOSAIC scales to complex scenes with zero extra training and provably reduces the variance during denoising process when more overlapping views are added, leading to improved generation quality. Experiments show that MOSAIC outperforms state-of-the-art baselines on image fidelity metrics in reconstructing complex multi-room environments. Resources and code are at https://mosaic-cmubig.github.io
format Preprint
id arxiv_https___arxiv_org_abs_2503_13816
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MOSAIC: Generating Consistent, Privacy-Preserving Scenes from Multiple Depth Views in Multi-Room Environments
Liu, Zhixuan
Zhu, Haokun
Chen, Rui
Francis, Jonathan
Hwang, Soonmin
Zhang, Ji
Oh, Jean
Computer Vision and Pattern Recognition
We introduce a diffusion-based approach for generating privacy-preserving digital twins of multi-room indoor environments from depth images only. Central to our approach is a novel Multi-view Overlapped Scene Alignment with Implicit Consistency (MOSAIC) model that explicitly considers cross-view dependencies within the same scene in the probabilistic sense. MOSAIC operates through a multi-channel inference-time optimization that avoids error accumulation common in sequential or single-room constraints in panorama-based approaches. MOSAIC scales to complex scenes with zero extra training and provably reduces the variance during denoising process when more overlapping views are added, leading to improved generation quality. Experiments show that MOSAIC outperforms state-of-the-art baselines on image fidelity metrics in reconstructing complex multi-room environments. Resources and code are at https://mosaic-cmubig.github.io
title MOSAIC: Generating Consistent, Privacy-Preserving Scenes from Multiple Depth Views in Multi-Room Environments
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.13816