Automatic floor plan analysis and recognition
Abstrak
Over the last few decades, floor plan analysis and recognition has been an open research topic in computer science, aiming to generate the building’s model by automatically extracting meaningful information from diverse sources. Among these, the architectural drawings are one of the most common, typically composed of non-uniform notations, together with their relationship and constraints, defining the structure’s layout and usage. Usually, floor plans encompass a high variability in style and semantics, as there is no standard notation to describe each element. Thus, numerous methodologies have been proposed to recognize, vectorize, and model different objects such as walls, doors, and rooms. In this work, we review different procedures from rule-based and learning-based approaches between the years 1995 to 2021, restricting only those considering the plan data as a rasterized image format. Datasets, scopes, and performed tasks were summarized to guide future development within the construction and design industries.
Penulis (4)
Pablo N. Pizarro
N. Hitschfeld
I. Sipiran
J. M. Saavedra
Akses Cepat
PDF tidak tersedia langsung
Cek di sumber asli →- Tahun Terbit
- 2022
- Bahasa
- en
- Total Sitasi
- 94×
- Sumber Database
- Semantic Scholar
- DOI
- 10.1016/j.autcon.2022.104348
- Akses
- Open Access ✓