DGTR: Distributed Gaussian Turbo-Reconstruction for Sparse-View Vast Scenes

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Hao, Gao, Yuanyuan, Peng, Haosong, Wu, Chenming, Ye, Weicai, Zhan, Yufeng, Zhao, Chen, Zhang, Dingwen, Wang, Jingdong, Han, Junwei
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913581744783360
author Li, Hao
Gao, Yuanyuan
Peng, Haosong
Wu, Chenming
Ye, Weicai
Zhan, Yufeng
Zhao, Chen
Zhang, Dingwen
Wang, Jingdong
Han, Junwei
author_facet Li, Hao
Gao, Yuanyuan
Peng, Haosong
Wu, Chenming
Ye, Weicai
Zhan, Yufeng
Zhao, Chen
Zhang, Dingwen
Wang, Jingdong
Han, Junwei
contents Novel-view synthesis (NVS) approaches play a critical role in vast scene reconstruction. However, these methods rely heavily on dense image inputs and prolonged training times, making them unsuitable where computational resources are limited. Additionally, few-shot methods often struggle with poor reconstruction quality in vast environments. This paper presents DGTR, a novel distributed framework for efficient Gaussian reconstruction for sparse-view vast scenes. Our approach divides the scene into regions, processed independently by drones with sparse image inputs. Using a feed-forward Gaussian model, we predict high-quality Gaussian primitives, followed by a global alignment algorithm to ensure geometric consistency. Synthetic views and depth priors are incorporated to further enhance training, while a distillation-based model aggregation mechanism enables efficient reconstruction. Our method achieves high-quality large-scale scene reconstruction and novel-view synthesis in significantly reduced training times, outperforming existing approaches in both speed and scalability. We demonstrate the effectiveness of our framework on vast aerial scenes, achieving high-quality results within minutes. Code will released on our [https://3d-aigc.github.io/DGTR].
format Preprint
id arxiv_https___arxiv_org_abs_2411_12309
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle DGTR: Distributed Gaussian Turbo-Reconstruction for Sparse-View Vast Scenes
Li, Hao
Gao, Yuanyuan
Peng, Haosong
Wu, Chenming
Ye, Weicai
Zhan, Yufeng
Zhao, Chen
Zhang, Dingwen
Wang, Jingdong
Han, Junwei
Computer Vision and Pattern Recognition
Novel-view synthesis (NVS) approaches play a critical role in vast scene reconstruction. However, these methods rely heavily on dense image inputs and prolonged training times, making them unsuitable where computational resources are limited. Additionally, few-shot methods often struggle with poor reconstruction quality in vast environments. This paper presents DGTR, a novel distributed framework for efficient Gaussian reconstruction for sparse-view vast scenes. Our approach divides the scene into regions, processed independently by drones with sparse image inputs. Using a feed-forward Gaussian model, we predict high-quality Gaussian primitives, followed by a global alignment algorithm to ensure geometric consistency. Synthetic views and depth priors are incorporated to further enhance training, while a distillation-based model aggregation mechanism enables efficient reconstruction. Our method achieves high-quality large-scale scene reconstruction and novel-view synthesis in significantly reduced training times, outperforming existing approaches in both speed and scalability. We demonstrate the effectiveness of our framework on vast aerial scenes, achieving high-quality results within minutes. Code will released on our [https://3d-aigc.github.io/DGTR].
title DGTR: Distributed Gaussian Turbo-Reconstruction for Sparse-View Vast Scenes
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2411.12309