Segmentation of Floorplans and Heritage Sites: An Approach to Unbalanced Dataset
Segmentation of Floorplans and Heritage Sites: An Approach to Unbalanced Dataset
Author(s): Sujith Gunjur Umapathy, Alexander I. IlievSubject(s): Library and Information Science, Information Architecture, Library operations and management, Education and training
Published by: Институт по математика и информатика - Българска академия на науките
Keywords: Deep Learning; Unbalanced Dataset; Semantic Segmentation; Focal Loss; Jaccard Loss
Summary/Abstract: To identify structural deficiencies, floorplan digitization is essential for legacy buildings, and historically significant and culturally rich sites. However, these technologies heavily depend on vector graphics. For an efficient and economical solution, we present an approach with multi-loss functions to handle unbalanced classes in floorplans using U-Net architecture.
Journal: Digital Presentation and Preservation of Cultural and Scientific Heritage
- Issue Year: 2022
- Issue No: XII
- Page Range: 205-216
- Page Count: 12
- Language: English