Prompt-Driven Building Footprint Extraction in Aerial Images with Offset-Building Model

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Kai, Deng, Yupeng, Kong, Yunlong, Liu, Diyou, Chen, Jingbo, Meng, Yu, Ma, Junxian, Wang, Chenhao
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912240603496448
author Li, Kai
Deng, Yupeng
Kong, Yunlong
Liu, Diyou
Chen, Jingbo
Meng, Yu
Ma, Junxian
Wang, Chenhao
author_facet Li, Kai
Deng, Yupeng
Kong, Yunlong
Liu, Diyou
Chen, Jingbo
Meng, Yu
Ma, Junxian
Wang, Chenhao
contents More accurate extraction of invisible building footprints from very-high-resolution (VHR) aerial images relies on roof segmentation and roof-to-footprint offset extraction. Existing methods based on instance segmentation suffer from poor generalization when extended to large-scale data production and fail to achieve low-cost human interaction. This prompt paradigm inspires us to design a promptable framework for roof and offset extraction, and transforms end-to-end algorithms into promptable methods. Within this framework, we propose a novel Offset-Building Model (OBM). Based on prompt prediction, we first discover a common pattern of predicting offsets and tailored Distance-NMS (DNMS) algorithms for offset optimization. To rigorously evaluate the algorithm's capabilities, we introduce a prompt-based evaluation method, where our model reduces offset errors by 16.6\% and improves roof Intersection over Union (IoU) by 10.8\% compared to other models. Leveraging the common patterns in predicting offsets, DNMS algorithms enable models to further reduce offset vector loss by 6.5\%. To further validate the generalization of models, we tested them using a newly proposed test set, Huizhou test set, with over 7,000 manually annotated instance samples. Our algorithms and dataset will be available at https://github.com/likaiucas/OBM.
format Preprint
id arxiv_https___arxiv_org_abs_2310_16717
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Prompt-Driven Building Footprint Extraction in Aerial Images with Offset-Building Model
Li, Kai
Deng, Yupeng
Kong, Yunlong
Liu, Diyou
Chen, Jingbo
Meng, Yu
Ma, Junxian
Wang, Chenhao
Computer Vision and Pattern Recognition
I.4.6; I.4.7; I.3.5; I.5.1
More accurate extraction of invisible building footprints from very-high-resolution (VHR) aerial images relies on roof segmentation and roof-to-footprint offset extraction. Existing methods based on instance segmentation suffer from poor generalization when extended to large-scale data production and fail to achieve low-cost human interaction. This prompt paradigm inspires us to design a promptable framework for roof and offset extraction, and transforms end-to-end algorithms into promptable methods. Within this framework, we propose a novel Offset-Building Model (OBM). Based on prompt prediction, we first discover a common pattern of predicting offsets and tailored Distance-NMS (DNMS) algorithms for offset optimization. To rigorously evaluate the algorithm's capabilities, we introduce a prompt-based evaluation method, where our model reduces offset errors by 16.6\% and improves roof Intersection over Union (IoU) by 10.8\% compared to other models. Leveraging the common patterns in predicting offsets, DNMS algorithms enable models to further reduce offset vector loss by 6.5\%. To further validate the generalization of models, we tested them using a newly proposed test set, Huizhou test set, with over 7,000 manually annotated instance samples. Our algorithms and dataset will be available at https://github.com/likaiucas/OBM.
title Prompt-Driven Building Footprint Extraction in Aerial Images with Offset-Building Model
topic Computer Vision and Pattern Recognition
I.4.6; I.4.7; I.3.5; I.5.1
url https://arxiv.org/abs/2310.16717