Sketch-based Methods for Researching Building Layouts through the Semantic Fingerprint of Architecture

The paper focuses on the early stages of the design process where the architect needs assistance in finding reference projects and describes different aspects of a concept for retrieving previous design solutions with similar layout characteristics. Such references are typically used to see how others have solved a similar architectural problem or simply for inspiration. Current electronic search methods use textual information rather than graphical information. The configuration of space and the relations between rooms are hard to represent using keywords, in fact transforming these spatial configurations into verbally expressed typologies tends to result in unclear and often imprecise descriptions of architecture. Nowadays, modern IT-technologies lead to fundamental changes during the process of designing buildings. Digital representations of architecture require suitable approaches to the storage, indexing and management of information as well as adequate retrieval methods. Traditionally planning information is represented in the form of floor plans, elevations, sections and textual descriptions. State of the art digital representations include renderings, computer aided design (CAD) and semantic information like Building Information Modelling (BIM) including 2D and 3D file formats such as Industry Foundation Classes (IFC) (IAI, 2010). In the paper, we examine the development of IT-technologies in the area of case-based reasoning (Richter et al., 2007) to provide a sketch-based submission and retrieval system for publishing and researching building layouts including their manipulation and subsequent use. The user interface focuses on specifying space and their relations by drawing them. This query style supports the spatial thinking approach that architects use, who often have a visual representation in mind without being able to provide an accurate description of the spatial configuration. The semantic fingerprint proposed by (Langenhan, 2008) is a description and query language for creating an index of floor plans to store meta-data about architecture, which can be used as signature for retrieving reference projects. The functional spaces, such as living room or kitchen and the relation among on another, are used to create a fingerprint. Furthermore, we propose a visual sketch-based interface (Weber et al., 2010) based on the Touch&Write paradigm (Liwicki et al., 2010) for the submission and the retrieval phase. During the submission process the architect is sketching the space-boundaries, space relations and functional coherence's. Using state of the art document analysis techniques, the architects are supported offering an automatic detection of room boundaries and their physical relations. During the retrieval the application will interpret the sketches of the architect and find reference projects based on a similarity based search utilizing the semantic fingerprint. By recommending reference projects, architects will be able to reuse collective experience which match the current requirements. The way of performing a search using a sketch as a query is a new way of thinking and working. The retrieval of 3D models based on a sketched shape are already realized in several domains. We already propose a step further, using the semantics of a spatial configuration. Observing the design process of buildings reveals that the initial design phase serves as the foundation for the quality of the later outcome. The sketch-based approach to access valuable information using the semantic fingerprint enables the user to digitally capture knowledge about architecture, to recover and reuse it in common-sense. Furthermore, automatically analysed fingerprints can put forward both commonly used as well as best practice projects. It will be possible to rate architecture according to the fingerprint of a building