Activities per year
Abstract
New computational methods provide means to deduce semantic information from
measurements, such as range scans and photographs of building interiors. In this paper, we
showcase a method that allows to estimate elements that are not directly observable – ducts
and power lines in walls. For this, we combine information, which is deducted by algorithms
from the raw data, with implicit information that is publicly available: technical standards that
restrict the placement of powerlines. These requirements define preferred installation zones,
which are represented by a rule‐based system in the proposed approach.
The approach is structured into the following steps: First, a coarse geometry is extracted from
input measurements; i.e. the unstructured, laser‐scanned point cloud is transformed into a
simplistic building model. Then, visible endpoints of electrical appliances (e.g. sockets,
switches) are detected from picture information using machine learning techniques and a pre-trained classifier. Afterwards, the positions of installation zones in walls are generated using the rule‐based system mentioned above. Finally, a hypothesis of non‐visible cable ducts is generated, under the assumption that (i) the real configuration obeys the rules of legal requirements and standards and (ii) the configuration connects all endpoints using as little as possible resources, i.e. cable length. Results of a first automatic pipeline are discussed.
measurements, such as range scans and photographs of building interiors. In this paper, we
showcase a method that allows to estimate elements that are not directly observable – ducts
and power lines in walls. For this, we combine information, which is deducted by algorithms
from the raw data, with implicit information that is publicly available: technical standards that
restrict the placement of powerlines. These requirements define preferred installation zones,
which are represented by a rule‐based system in the proposed approach.
The approach is structured into the following steps: First, a coarse geometry is extracted from
input measurements; i.e. the unstructured, laser‐scanned point cloud is transformed into a
simplistic building model. Then, visible endpoints of electrical appliances (e.g. sockets,
switches) are detected from picture information using machine learning techniques and a pre-trained classifier. Afterwards, the positions of installation zones in walls are generated using the rule‐based system mentioned above. Finally, a hypothesis of non‐visible cable ducts is generated, under the assumption that (i) the real configuration obeys the rules of legal requirements and standards and (ii) the configuration connects all endpoints using as little as possible resources, i.e. cable length. Results of a first automatic pipeline are discussed.
Original language | English |
---|---|
Pages | 77-84 |
Number of pages | 8 |
Publication status | Published - Apr 2016 |
Event | Places and Technologies 2016 - University of Belgrade - Faculty of Architecture, Belgrade, Serbia Duration: 14 Apr 2016 → 15 Apr 2016 |
Conference
Conference | Places and Technologies 2016 |
---|---|
Country/Territory | Serbia |
City | Belgrade |
Period | 14/04/16 → 15/04/16 |
Keywords
- builiding information modeling
- as-built BIM
- semantic enrichment
- geometric enrichment
ASJC Scopus subject areas
- Computer Graphics and Computer-Aided Design
- Computer Vision and Pattern Recognition
Fields of Expertise
- Information, Communication & Computing
Activities
- 1 Talk at conference or symposium
-
Places and Technologies 2016
Ulrich Krispel (Speaker)
15 Apr 2016Activity: Talk or presentation › Talk at conference or symposium › Science to science