| Title: |
Polygonal Building Segmentation by Frame Field Learning |
| Authors: |
Girard, Nicolas; Smirnov, Dmitriy; Solomon, Justin; Tarabalka, Yuliya |
| Contributors: |
COMUE Université Côte d'Azur (2015-2019) (COMUE UCA); Geometric Modeling of 3D Environments (TITANE); Inria Sophia Antipolis - Méditerranée (CRISAM); Institut National de Recherche en Informatique et en Automatique (Inria)-Institut National de Recherche en Informatique et en Automatique (Inria); Computer Science and Artificial Intelligence Laboratory Cambridge (CSAIL); Massachusetts Institute of Technology (MIT); Luxcarta technology Mouans-Sartoux (LCT); Thanks to ANR for funding the project EPITOME ANR-17-CE23-0009 and to Inria Sophia Antipolis - Méditerranée “Nef” computation cluster for providing resources and support. The MIT Geometric Data Processing group acknowledges the generous support of Army Research Office grantW911NF2010168, of Air Force Office of Scientific Research award FA9550-19-1-031, of National Science Foundation grant IIS-1838071, from the CSAIL Systems that Learn program, from the MIT–IBM Watson AI Laboratory, from the Toyota–CSAIL Joint Research Center, from a gift from Adobe Systems, from an MIT.nano Immersion Lab/NCSOFT Gaming Program seed grant, and from the Skoltech–MIT Next Generation Program. This work was also supported by the National Science Foundation Graduate Research Fellowship under Grant No. 1122374.; ANR-19-P3IA-0002 - 3IA Côte d'Azur - Nice - Interdisciplinary Institute for Artificial Intelligence; ANR-17-CE23-0009,EPITOME,Représentation efficace pour des images satellites à grande échelle(2017); ANR-19-P3IA-0002,3IA@cote d'azur,3IA Côte d'Azur(2019) |
| Source: |
CVPR 2021 - IEEE Conference on Computer Vision and Pattern Recognition ; https://inria.hal.science/hal-02548545 ; CVPR 2021 - IEEE Conference on Computer Vision and Pattern Recognition, Jun 2021, Pittsburg / Virtual, United States |
| Publisher Information: |
HAL CCSD |
| Publication Year: |
2021 |
| Collection: |
Université de Rennes 1: Publications scientifiques (HAL) |
| Subject Terms: |
polygonization; PolyVector field; frame field; remote sensing; regularization; segmentation; [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV]; [INFO.INFO-LG]Computer Science [cs]/Machine Learning [cs.LG]; [INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV] |
| Subject Geographic: |
Pittsburg / Virtual; United States |
| Description: |
International audience ; While state of the art image segmentation models typically output segmentations in raster format, applications in geographic information systems often require vector polygons. To help bridge the gap between deep network output and the format used in downstream tasks, we add a frame field output to a deep segmentation model for extracting buildings from remote sensing images. We train a deep neural network that aligns a predicted frame field to ground truth contours. This additional objective improves segmentation quality by leveraging multi-task learning and provides structural information that later facilitates polygonization; we also introduce a polygonization algorithm that utilizes the frame field along with the raster segmentation. Our code is available at https://github.com/Lydorn/Polygonization-by-Frame-Field-Learning. |
| Document Type: |
conference object |
| Language: |
English |
| Availability: |
https://inria.hal.science/hal-02548545; https://inria.hal.science/hal-02548545v2/document; https://inria.hal.science/hal-02548545v2/file/archive.pdf |
| Rights: |
info:eu-repo/semantics/OpenAccess |
| Accession Number: |
edsbas.E1E2BCF9 |
| Database: |
BASE |