Symbol Spotting on Digital Architectural Floor Plans Using a Deep Learning-based Framework

Rezvanifar, Alireza; Cote, Melissa; Albu, Alexandra Branzan

doi:10.1109/cvprw50498.2020.00292

Cited by 26 publications

(36 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In Chapter 4 we relaxed the on-the-fly property of symbol spotting methods and leveraged the recent advances in DL-based object detection neural networks to detect architectural symbols (published in [20]). We used a tiling strategy for training YOLOv2 which allows us to 1) detect small symbols in large input images, 2) collect semantic information around each symbol, 3) not change the aspect ratio of the input image and consequently the morphology of symbols and finally 4) data augmentation.…”

Section: Discussionmentioning

confidence: 99%

“…• By relaxing the on-the fly property of symbol spotting, a Deep Learning (DL)-based framework is proposed to address the practical challenges of detecting symbols in real-world architectural floor plans [20].…”

Section: Thesis Objectives and Contributionsmentioning

confidence: 99%

“…Chapter 4 (Ref. [20]) focuses on symbol detection on real-world digital architectural floor plans with a DL-based framework. By relaxing the on-the-fly property of spotting approaches, this method can detect symbols with different notations.…”

Section: Outline Of Thesismentioning

confidence: 99%

“…The main requirement for symbol spotting relates to performing "on-the-fly" queries [9], which precludes us from using learning-based methods, including deep learning. While some recent neural networkbased methods relax the on-the-fly requirement [20,87,101] and reframe symbol spotting as an object detection problem, this chapter focuses on traditional symbol spotting, described in detail in Chapter 2, where the main objective is to retrieve all the image patches which contain the "user-selected" symbol.…”

Section: Born-digital Architectural Floor Plansmentioning

confidence: 99%

“…Symbol spotting on digital architectural floor plans using a Deep Learning-based framework (published in[20]), I was the first author of this work. The co-authors of this paper helped me with the structural design of the manuscript, literature review, writing and revising the paper in the submission process.…”

mentioning

confidence: 99%

See 4 more Smart Citations

Geometry-based symbol spotting in born-digital architectural floor plans

2021

Self Cite

View full text Add to dashboard Cite

Architectural floor plans are scale-accurate 2D drawings of one level of a building, seen from above, which convey structural and semantic information related to rooms, walls, symbols, textual data, etc. They consist of lines, curves, symbols, and textual markings, showing the relationships between rooms and all physical features, required for the proper construction or renovation of the building.First, this thesis provides a thorough study of state-of-the-art on symbol spotting methods for architectural drawings, an application domain providing the document image analysis and graphic recognition communities with an interesting set of challenges linked to the sheer complexity and density of embedded information, that have yet to be resolved.Second, we propose a hybrid method that capitalizes on strengths of both vector-based and pixel-based symbol spotting techniques. In the description phase, the salient geometric constituents of a symbol are extracted by a variety of vectorization techniques, including a proposed voting-based algorithm for finding partial ellipses. This enables us to better handle local shape irregularities and boundary discontinuities, as well as partial occlusion and overlap. In the matching phase, the spatial relationship between the geometric primitives is encoded via a primitive-aware proximity graph. A statistical approach is then used to rapidly yield a coarse localization of symbols within the plan. Localization is further refined with a pixel-based step implementing a modified cross-correlation function. Experimental results on the public SESYD synthetic dataset and real-world images demonstrate that our approach clearly outperforms other popular symbol spotting approaches.Traditional on-the-fly symbol spotting methods are unable to address the semantic challenge of graphical notation variability, i.e. low intra-class symbol similarity, an issue that is particularly important in architectural floor plan analysis. The presence of occlusion and clutter, characteristic of real-world plans, along with a varying graphical symbol complexity from almost trivial to highly complex, also pose challenges to existing spotting methods. iv Third, we address all the above issues by leveraging recent advances in deep learningbased neural networks and adapting an object detection framework based on the YOLO (You Only Look Once) architecture. We propose a training strategy based on tiles, avoiding many issues particular to deep learning-based object detection networks related to the relatively small size of symbols compared to entire floor plans, aspect ratios, and data augmentation. Experimental results demonstrate that our method successfully detects architectural symbols with low intra-class similarity and of variable graphical complexity,

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Thesis Objectives and Contributionsmentioning

confidence: 99%

Section: Outline Of Thesismentioning

confidence: 99%

Section: Born-digital Architectural Floor Plansmentioning

confidence: 99%

mentioning

confidence: 99%

See 3 more Smart Citations

Geometry-based symbol spotting in born-digital architectural floor plans

2021

Self Cite

View full text Add to dashboard Cite

show abstract

Extracting Information from Old and Scanned Engineering Drawings of Existing Buildings for the Creation of Digital Building Models

Al-Wesabi,

Bach,

Schönfelder

et al. 2023

Lecture Notes in Civil Engineering

View full text Add to dashboard Cite

MuraNet: Multi-task Floor Plan Recognition with Relation Attention

Huang¹,

Wu²,

Wei³

et al. 2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

The recognition of information in floor plan data requires the use of detection and segmentation models. However, relying on several single-task models can result in ineffective utilization of relevant information when there are multiple tasks present simultaneously. To address this challenge, we introduce MuraNet, an attention-based multitask model for segmentation and detection tasks in floor plan data. In MuraNet, we adopt a unified encoder called MURA as the backbone with two separated branches: an enhanced segmentation decoder branch and a decoupled detection head branch based on YOLOX, for segmentation and detection tasks respectively. The architecture of MuraNet is designed to leverage the fact that walls, doors, and windows usually constitute the primary structure of a floor plan's architecture. By jointly training the model on both detection and segmentation tasks, we believe MuraNet can effectively extract and utilize relevant features for both tasks. Our experiments on the CubiCasa5k public dataset show that MuraNet improves convergence speed during training compared to single-task models like U-Net and YOLOv3. Moreover, we observe improvements in the average AP and IoU in detection and segmentation tasks, respectively. Our ablation experiments demonstrate that the attention-based unified backbone of MuraNet achieves better feature extraction in floor plan recognition tasks, and the use of decoupled multi-head branches for different tasks further improves model performance. We believe that our proposed Mu-raNet model can address the disadvantages of single-task models and improve the accuracy and efficiency of floor plan data recognition.

show abstract

Symbol Spotting on Digital Architectural Floor Plans Using a Deep Learning-based Framework

Cited by 26 publications

References 33 publications

Geometry-based symbol spotting in born-digital architectural floor plans

Geometry-based symbol spotting in born-digital architectural floor plans

Extracting Information from Old and Scanned Engineering Drawings of Existing Buildings for the Creation of Digital Building Models

MuraNet: Multi-task Floor Plan Recognition with Relation Attention

Contact Info

Product

Resources

About