Floor Plan Images Research Articles

Symbols Detection and Classification using Graph Neural Networks

In this paper, we propose a method to both extract and classify symbols in floorplan images. This method relies on the very recent developments of Graph Neural Networks (GNN). In the proposed approach, floorplan images are first converted into Region Adjacency Graphs (RAGs). In order to achieve both classification and extraction, two different GNNs are used. The first one aims at classifying each node of the graph while the second targets the extraction of clusters corresponding to symbols. In both cases, the model is able to take into account edge features. Each model is firstly evaluated independently before combining both tasks simultaneously, increasing the quickness of the results.

AUTOMATIC 3D BUILDING MODEL GENERATION USING DEEP LEARNING METHODS BASED ON CITYJSON AND 2D FLOOR PLANS

Abstract. In the past decade, a lot of effort is put into applying digital innovations to building life cycles. 3D Models have been proven to be efficient for decision making, scenario simulation and 3D data analysis during this life cycle. Creating such digital representation of a building can be a labour-intensive task, depending on the desired scale and level of detail (LOD). This research aims at creating a new automatic deep learning based method for building model reconstruction. It combines exterior and interior data sources: 1) 3D BAG, 2) archived floor plan images. To reconstruct 3D building models from the two data sources, an innovative combination of methods is proposed. In order to obtain the information needed from the floor plan images (walls, openings and labels), deep learning techniques have been used. In addition, post-processing techniques are introduced to transform the data in the required format. In order to fuse the extracted 2D data and the 3D exterior, a data fusion process is introduced. From the literature review, no prior research on automatic integration of CityGML/JSON and floor plan images has been found. Therefore, this method is a first approach to this data integration.

How Many Bedrooms Do You Need? A Real-Estate Recommender System from Architectural Floor Plan Images

This paper introduces an automated image processing method to analyze an architectural floor plan database. The floor plan information, such as the measurement of the rooms, dimension lines, and even the location of each room, can be automatically produced. This assists the real-estate agents to maximise the chances of the closure of deals by providing explicit insights to the prospective purchasers. With a clear idea about the layout of the place, customers can quickly make an analytical decision. Besides, it reduces the specialized training cost and increases the efficiency in business actions by understanding the property types with the greatest demand. Succinctly, this paper utilizes both the traditional image processing and convolutional neural networks (CNNs) to detect the bedrooms by undergoing the segmentation and classification processes. A thorough experiment, analysis, and evaluation had been performed to verify the effectiveness of the proposed framework. As a result, a three-class bedroom classification accuracy of ∼ 90% was achieved when validating on more than 500 image samples that consist of the different room numbers. In addition, qualitative findings were presented to manifest visually the feasibility of the algorithm developed.

A novel CNN method for the accurate spatial data recovery from digital images

The parsing of floorplans has been an issue for a long time in automated document processing and with algorithmic methods until recent years. This problem has also improved output with the emergence of convolutionary neural networks (CNN). The job here is to obtain spatial and geometrical data from floorplans as accurately as possible. The aim of this project is to extract the most information from a floor plan image around instance segmentation models like Cascade Mask R-CNN. A new style of key point CNN is being implemented to supplement the segmentation to find correct corner positions. Then the resulting segmentation is combined in a post-processing stage. With a mean IoU of 72.7 percent versus 57.5 percent, the resulting segmentation scores surpass the existing baseline of the CubiCasa5k floorplan data base. Moreover, for almost every class, the mean IoU for each class is increased. Cascade Mask R-CNN has also been shown to be better suited to this role than Mask R-CNN.

Evaluation of Deep Learning-Based Automatic Floor Plan Analysis Technology: An AHP-Based Assessment

This study proposes a technology that allows automatic extraction of vectorized indoor spatial information from raster images of floor plans. Automatic reconstruction of indoor spaces from floor plans is based on a deep learning algorithm, which trains on scanned floor plan images and extracts critical indoor elements such as room structures, junctions, walls, and openings. The newly developed technology proposed herein can handle complicated floor plans which could not be automatically extracted by previous studies because of its complexity and difficulty in being trained in deep learning. Such complicated reconstruction solely from a floor plan image can be digitized and vectorized either through manual drawing or with the help of newly developed deep learning-based automatic extraction. This study proposes an evaluation framework for assessing this newly developed technology against manual digitization. Using the analytical hierarchy process, the hierarchical aspects of technology value and their relative importance are systematically quantified. The analysis suggested that the automatic technology using a deep learning algorithm had predominant criteria followed by, substitutability, completeness, and supply and demand. In this study, the technology value of automatic floor plan analysis compared with that of traditional manual edits is compared systemically and assessed qualitatively, which had not been done in existing studies. Consequently, this study determines the effectiveness and usefulness of automatic floor plan analysis as a reasonable technology for acquiring indoor spatial information.

Knowledge-driven description synthesis for floor plan interpretation

Image captioning is a widely known problem in the area of AI. Caption generation from floor plan images has applications in indoor path planning, real estate, and providing architectural solutions. Several methods have been explored in the literature for generating captions or semi-structured descriptions from floor plan images. Since only the caption is insufficient to capture fine-grained details, researchers also proposed descriptive paragraphs from images. However, these descriptions have a rigid structure and lack flexibility, making it difficult to use them in real-time scenarios. This paper offers two models, description synthesis from image cue (DSIC) and transformer-based description generation (TBDG), for text generation from floor plan images. These two models take advantage of modern deep neural networks for visual feature extraction and text generation. The difference between both models is in the way they take input from the floor plan image. The DSIC model takes only visual features automatically extracted by a deep neural network, while the TBDG model learns textual captions extracted from input floor plan images with paragraphs. The specific keywords generated in TBDG and understanding them with paragraphs make it more robust in a general floor plan image. Experiments were carried out on a large-scale publicly available dataset and compared with state-of-the-art techniques to show the proposed model’s superiority.

EyeLoc: Smartphone Vision-Enabled Plug-n-Play Indoor Localization in Large Shopping Malls

Indoor localization is becoming an emerging requirement in many large shopping malls. Existing indoor localization systems, however, require exhausted system bootstraps and calibration phases. The huge sunk cost usually hinders practical deployment of the indoor localization systems in large shopping malls. In contrast, we observe that floor-plan images of large shopping malls, which highlight the positions of many shops, are widely available in Google Maps, Gaode Maps, Baidu Maps, etc. According to several observed shops, people can localize themselves (called self-localization). However, due to the requirements of geometric sense and space transformation, not all people get used to this way. In this article, we propose EyeLoc, which uses smartphone vision to enable accurate self-localization on a floor-plan image. EyeLoc addresses several challenges, including developing a ubiquitous smartphone vision system, extracting efficient vision clues, and achieving robust measurement error mitigation. We implement EyeLoc in Android and evaluate its performance in emulated environment, two large shopping malls and a semioutdoor large Outlets. The results show that the 90-percentile errors of localization and heading direction are 5.97 m and 20° in 70 000 m <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> malls.

Data-driven floor plan understanding in rural residential buildings via deep recognition

Automatic understanding of floor plan images is a key component of various applications. Due to the style diversity of rural housing design, the latest learning-based approaches cannot achieve satisfactory recognition results. In this paper, we present a new framework for parsing floor plans of rural residence that combines semantic neural networks with a post-processed room segmentation. First, we take case studies from typical residential buildings in China’s rural areas and provide a novel image dataset, called RuralHomeData, containing 800 rural residence floor plans with accurate man-made annotations. Based on the dataset, we propose a new deep learning-based recognition framework using a joint neural network to predict the geometric elements and text information on the floor plan simultaneously. Our insight is that walls and openings (doors and windows) are the basic elements corresponding to the room boundary that a closed 1D loop must form a certain room. Then the semantic information (e.g., the room function) of room regions can be obtained through text detection and identification. Furthermore, we use the MIQP algorithm to divide the area containing multiple room type texts into multiple room areas. Finally, the input floor plan can be transformed into a room layout graph with room attributes and adjacent relationships. The proposed algorithm has been tested on both urban and rural datasets, and the experimental results demonstrate our efficiency and robustness compared with the state-of-the-art methods.

Framework for Indoor Elements Classification via Inductive Learning on Floor Plan Graphs

This paper presents a new framework to classify floor plan elements and represent them in a vector format. Unlike existing approaches using image-based learning frameworks as the first step to segment the image pixels, we first convert the input floor plan image into vector data and utilize a graph neural network. Our framework consists of three steps. (1) image pre-processing and vectorization of the floor plan image; (2) region adjacency graph conversion; and (3) the graph neural network on converted floor plan graphs. Our approach is able to capture different types of indoor elements including basic elements, such as walls, doors, and symbols, as well as spatial elements, such as rooms and corridors. In addition, the proposed method can also detect element shapes. Experimental results show that our framework can classify indoor elements with an F1 score of 95%, with scale and rotation invariance. Furthermore, we propose a new graph neural network model that takes the distance between nodes into account, which is a valuable feature of spatial network data.

A rotation and scale invariant approach for multi-oriented floor plan image retrieval

An automatic system for analysis and retrieval of building floor plans images is helpful for the architects while designing new projects and providing recommendations to the buyers. For such systems, query by example is preferred over query by keyword, for which user’s requirements must be available in digital image form. Floor plans are converted to digital form by scanning and often get rotated slightly by a certain degree of angle during digitization. In this paper, we have proposed a geometric feature-based approach for floor plan image retrieval and our key contribution is to handle different kinds of rotation and scale while retrieving similar floor plan from the database. Our framework is divided into three phases, namely outer shape feature extraction, internal object feature extraction, followed by matching and retrieval. For our experimentation, we have rotated images of ROBIN dataset as currently no rotated floor plan dataset was available. Our experiment shows that the proposed methodology outperforms recent competing methods.

A Study on the Making Indoor Topology Map for the Autonomous Driving of Wearable Robots

When walking with a wearable robot, we may not react properly to the surrounding environment. Therefore, we need to help the wearable robot to walk correctly by providing information of the surrounding environment in the indoor space. SLAM creates a map while the robot moves in real time, so you can use the same map as the actual structure, but it takes a lot of time to create the map. We produce a topology map using the floor plan image. Since the floor plan is produced when a building is made, it is very similar to the actual structure of the building and it is very efficient in terms of time because it does not need to create a map in real-time. In addition, since we generate our topology map using the sensor nodes installed in the building, it matches the coordinates on the floor plan one to one so that the path can be set more accurately and we can find doors, stairs, and obstacles more easily found through object detection. In this study, create a topology map based on a floor plan so that a wearable robot can walk more stably and autonomously, and we can provide the information of object detection by learning a dataset of rooms, doors, and stairs. By synthesizing this information, we transmit it to the wearable robot and provide a path to progress. Conduct an experiment to detect obstacles in the path and transmit the information to wearable robots to prevent accidents.

Graph2Plan

We introduce a learning framework for automated floorplan generation which combines generative modeling using deep neural networks and user-in-the-loop designs to enable human users to provide sparse design constraints. Such constraints are represented by a layout graph. The core component of our learning framework is a deep neural network, Graph2Plan, which converts a layout graph, along with a building boundary, into a floorplan that fulfills both the layout and boundary constraints. Given an input building boundary, we allow a user to specify room counts and other layout constraints, which are used to retrieve a set of floorplans, with their associated layout graphs, from a database. For each retrieved layout graph, along with the input boundary, Graph2Plan first generates a corresponding raster floorplan image, and then a refined set of boxes representing the rooms. Graph2Plan is trained on RPLAN, a large-scale dataset consisting of 80K annotated floorplans. The network is mainly based on convolutional processing over both the layout graph, via a graph neural network (GNN), and the input building boundary, as well as the raster floorplan images, via conventional image convolution. We demonstrate the quality and versatility of our floorplan generation framework in terms of its ability to cater to different user inputs. We conduct both qualitative and quantitative evaluations, ablation studies, and comparisons with state-of-the-art approaches.

Indoor mapping and modeling by parsing floor plan images

ABSTRACT A large proportion of indoor spatial data is generated by parsing floor plans. However, a mature and automatic solution for generating high-quality building elements (e.g., walls and doors) and space partitions (e.g., rooms) is still lacking. In this study, we present a two-stage approach to indoor mapping and modeling (IMM) from floor plan images. The first stage vectorizes the building elements on the floor plan images and the second stage repairs the topological inconsistencies between the building elements, separates indoor spaces, and generates indoor maps and models. To reduce the shape complexity of indoor boundary elements, i.e., walls and openings, we harness the regularity of the boundary elements and extract them as rectangles in the first stage. Furthermore, to resolve the overlaps and gaps of the vectorized results, we propose an optimization model that adjusts the rectangle vertex coordinates to conform to the topological constraints. Experiments demonstrate that our approach achieves a considerable improvement in room detection without conforming to Manhattan World Assumption. Our approach also outputs instance-separate walls with consistent topology, which enables direct modeling into Industry Foundation Classes (IFC) or City Geography Markup Language (CityGML).

Indoor Reconstruction from Floorplan Images with a Deep Learning Approach

Although interest in indoor space modeling is increasing, the quantity of indoor spatial data available is currently very scarce compared to its demand. Many studies have been carried out to acquire indoor spatial information from floorplan images because they are relatively cheap and easy to access. However, existing studies do not take international standards and usability into consideration, they consider only 2D geometry. This study aims to generate basic data that can be converted to indoor spatial information using IndoorGML (Indoor Geography Markup Language) thick wall model or the CityGML (City Geography Markup Language) level of detail 2 by creating vector-formed data while preserving wall thickness. To achieve this, recent Convolutional Neural Networks are used on floorplan images to detect wall and door pixels. Additionally, centerline and corner detection algorithms were applied to convert wall and door images into vector data. In this manner, we obtained high-quality raster segmentation results and reliable vector data with node-edge structure and thickness attributes that enabled the structures of vertical and horizontal wall segments and diagonal walls to be determined with precision. Some of the vector results were converted into CityGML and IndoorGML form and visualized, demonstrating the validity of our work.

SUGAMAN: describing floor plans for visually impaired by annotation learning and proximity‐based grammar

In this study, the authors propose a framework SUGAMAN (Supervised and Unified framework using Grammar and Annotation Model for Access and Navigation). SUGAMAN is a Hindi word meaning ‘easy passage from one place to another’. SUGAMAN synthesises textual description from a given floor plan image, usable by visually impaired to navigate by understanding the arrangement of rooms and furniture. It is the first framework for describing a floor plan and giving direction for obstacle-free movement within a building. The model learns five classes of room categories from 1355 room image samples under a supervised learning paradigm. These learned annotations are fed into a description synthesis framework to yield a holistic description of a floor plan image. Authors demonstrate the performance of various supervised classifiers on room learning and provided a comparative analysis of system generated and human-written descriptions. The contribution of this study includes a novel framework for description generation from document images with graphics while proposing a new feature representing the floor plans, text annotations for a publicly available data set, and an algorithm for door to door obstacle avoidance navigation. This work can be applied to areas like understanding floor plans and design of historical monuments, and retrieval.

A novel feature transform framework using deep neural network for multimodal floor plan retrieval

In recent past, there has been a steep increase in the use of online platforms for the search of desired products. Real estate industry is no exception and has started initiating rent/sale of houses through online platforms. In this paper, we propose a deep neural network framework to facilitate automatic search of homes based on their floor plans. The salient features of this framework are that the query can be either an image (existing floor plan) or a sketch through a sketch pad interface. Our proposed framework automatically determines the type of query (image or sketch) and retrieves similar floor plan images from the database. The critical contributions of our proposed approach are: (1) a novel unified floor plan retrieval framework using multimodal query, i.e., an intuitive and convenient sketch query mode as well as a query by example mode ; (2) a conjunction of autoencoder, Cyclic GAN and CNN for the task of domain mapping and floor plan image retrieval. We have reported results of extensive experimentation and comparison with baseline results to establish the effectiveness of our approach.

SU‐G‐206‐17: RadShield: Semi‐Automated Shielding Design for CT Using NCRP 147 and Isodose Curves

Purpose:Computed tomography (CT) exam rooms are shielded more quickly and accurately compared to manual calculations using RadShield, a semi‐automated diagnostic shielding software package. Last year, we presented RadShield's approach to shielding radiographic and fluoroscopic rooms calculating air kerma rate and barrier thickness at many points on the floor plan and reporting the maximum values for each barrier. RadShield has now been expanded to include CT shielding design using not only NCRP 147 methodology but also by overlaying vendor provided isodose curves onto the floor plan.Methods:The floor plan image is imported onto the RadShield workspace to serve as a template for drawing barriers, occupied regions and CT locations. SubGUIs are used to set design goals, occupancy factors, workload, and overlay isodose curve files. CTDI and DLP methods are solved following NCRP 147. RadShield's isodose curve method employs radial scanning to extract data point sets to fit kerma to a generalized power law equation of the form K(r) = ar^b. RadShield's semiautomated shielding recommendations were compared against a board certified medical physicist's design using dose length product (DLP) and isodose curves.Results:The percentage error found between the physicist's manual calculation and RadShield's semi‐automated calculation of lead barrier thickness was 3.42% and 21.17% for the DLP and isodose curve methods, respectively. The medical physicist's selection of calculation points for recommending lead thickness was roughly the same as those found by RadShield for the DLP method but differed greatly using the isodose method.Conclusion:RadShield improves accuracy in calculating air‐kerma rate and barrier thickness over manual calculations using isodose curves. Isodose curves were less intuitive and more prone to error for the physicist than inverse square methods. RadShield can now perform shielding design calculations for general scattering bodies for which isodose curves are provided.

Mobile robot localization using the Hausdorff distance

SUMMARYThis paper presents a novel method for localization of mobile robots in structured environments. The estimation of the position and orientation of the robot relies on the minimisation of the partial Hausdorff distance between ladar range measurements and a floor plan image of the building. The approach is employed in combination with an extended Kalman filter to obtain accurate estimates of the robot's position, heading and velocity. Good estimates of these variables were obtained during tests performed using a differential drive robot, thus demonstrating that the approach provides an accurate, reliable and computationally feasible alternative for indoor robot localization and autonomous navigation.