Click here to go to the home page

Research work on Floor Plan Image Analysis



Graphical documents are human made instruments of communication that convey complex concepts using rich diagrammatic information. Floor plans are one of the type of graphical document which shows the structural planning and visually model the structure of a building. In architecture a floor plan is drawing to scale, showing a view from above of the relationship between rooms, spaces and other feature at one level of a structure. Understanding a graphical document require three steps

  1. Understanding/recognizing the symbols and denotations contained in it. As a floor plan contains various symbols example, symbols of walls, door, windows, and furniture (Low level Analysis).
  2. Identifying the relation between different symbols. Understanding the structure of the floor plan. (Mid level Analysis).
  3. Understanding the relation between different types of rooms and other entities. Extracting the fine grain details. (High level Analysis).

Development of Multimodal Search Framework for Architectural Floor plan



  1. Funding Agency: SCIENCE and ENGINEERING RESEARCH BOARD (SERB), India
  2. Duration: 3 years (2017-2020)

As part of this project, we have published two labelled datasets for the purpose of Content Based Floor Plan Image Retrieval.
...

ROBIN (Repository Of BuildIng plaNs)

Description: In ROBIN there are three categories of floor plans, depicting the three broad classes of houses. There are houses with 3, 4 and 5 rooms. The purpose and the subsequent label of each room varies depending on the various decors present in them. To be consistent with the existing floor plan datasets, the furniture symbols used in ROBIN dataset are similar to the ones used in SESYD dataset.


Link to the Repository


...

BRIDGE (Building Plan Repository for Image Description Generation, and Evaluation)

Description: BRIDGE (Building plan Repository for Image Description Generation, and Evaluation) dataset contains more than 13000 images of the floor plan and annotations collected from various websites, as well as publicly available floor plan images in the research domain. The images in BRIDGE also has annotations for symbols, region graphs, and paragraph descriptions. The BRIDGE dataset will be useful for symbol spotting, caption and description generation, scene graph synthesis, retrieval and many other tasks involving building plan parsing.


Link to the Repository


  • Rasika Khade, Krupa Jariwala, Chiranjoy Chattopadhyay; "An Interactive Floor Plan Image Retrieval Framework Based on Structural Features", Accepted for Publication in Arabian Journal for Science and Engineering, 2023.
  • Atharva Kalsekar, Rasika Khade, Krupa Jariwala, Chiranjoy Chattopadhyay; "RISC-Net : Rotation Invariant Siamese Convolution Network for Floor Plan Image Retrieval", DOI: 10.1007/s11042-022-13124-3 , Multimedia Tools and Applications, 2022.
  • Shreya Goyal, Chiranjoy Chattopadhyay , Gaurav Bhatnagar; "A Hybrid Network for Scanned Floor plan Image Recognition and Room Labeling", International Conference on Pattern Recognition and Machine Intelligence, Kolkata, India (December 2021)
  • Shreya Goyal, Naimul Khan, Chiranjoy Chattopadhyay, Gaurav Bhatnagar; "GRIHA: Synthesizing 2-Dimensional Building Layouts from Images Captured using a Smart Phone", Multimedia Tools and Applications, Vol. 81, pp. 14589–14612, 2022.
  • Shreya Goyal, Chiranjoy Chattopadhyay, Gaurav Bhatnagar; "Knowledge driven Description Synthesis for Floor Plan Interpretation", International Journal on Document Analysis and Recognition, pp. 1-14, DOI: 10.1007/s10032-021-00367-3, March 2021 (Available Online).
  • Rasika Khade, Krupa Jariwala, Chiranjoy Chattopadhyay, Umapada Pal; "A Rotation and Scale Invariant Approach for Multi-oriented Floor Plan Image Retrieval", Pattern Recognition Letters, pp. 1-8, DOI: 10.1016/j.patrec.2021.01.020, January, 2021 (Available Online).
  • Shreya Goyal, Satya Bhavsar, Shreya Patel, Chiranjoy Chattopadhyay, Gaurav Bhatnagar; "SUGAMAN: Describing Floor Plans for Visually Impaired by Annotation Learning and Proximity based Sentence Model", IET Image Processing, Vol 13, pp. 1-8, 2019.
  • Mahak Jain, Anurag Sanyal, Shreya Goyal, Chiranjoy Chattopadhyay, Gaurav Bhatnagar ; "Automatic Rendering of Building Floor Plan Images from Textual Descriptions in English", Preprint, arXiv:1811.11938.
  • Divya Sharma, Gaurav Harit, Chiranjoy Chattopadhyay ; "Attributed paths for layout-based document retrieval", 4th Workshop on Document Analysis and Recognition (DAR), Hyderabad, India (December 2018)
  • Divya Sharma, Nitin Gupta, Chiranjoy Chattopadhyay and Sameep Mehta; "REXplore: A Sketch based Interactive Explorer for Real Estates using Building Floor Plan Images", IEEE International Symposium on Multimedia (ISM 2018) , Taichung, Taiwan, December 2018. Poster
  • Shreya Goyal, Chiranjoy Chattopadhyay, Gaurav Bhatnagar; "ASYSST: A Framework for Synopsis Synthesis Empowering Visually Impaired", Workshop on Multimedia for Accessible Human Computer Interface, ACM Multimedia (MM 2018) , Seoul, Korea, October 2018. Poster
  • Divya Sharma, Chiranjoy Chattopadhyay; "High-Level Feature Aggregation for Fine-Grained Architectural Floor Plan Retrieval",   IET Computer Vision, Vol. 12, No. 5, August 2018, DOI: 10.1049/iet-cvi.2017.0581. Graphical Abstract
  • Shreya Goyal, Chiranjoy Chattopadhyay , Gaurav Bhatnagar; "Plan2Text: A Framework for Describing Building Floor Plan Images From First Person Perspective", IEEE Colloquium on Signal Processing and its Applications (CSPA 2018), Malayasia (Mar 2018), DOI: 10.1109/CSPA.2018.8368681. Poster
  • Divya Sharma, Nitin Gupta, Chiranjoy Chattopadhyay and Sameep Mehta; "DANIEL: A Deep Architecture for Automatic Analysis and Retrieval of Building Floor Plans", IAPR International Conference on Document Analysis and Recognition (ICDAR), Kyoto, Japan (Nov 2017). Poster
  • Divya Sharma, Chiranjoy Chattopadhyay and Gaurav Harit; " Analysis and Retrieval of Architectural Floor Plans",  IAPR Summer School on Document Analysis: Document Informatics, Jaipur, India (January 2017). Poster
  • Shreya Goyal, Chiranjoy Chattopadhyay, Gaurav Bhatnagar; " Automatic Textual Description Synthesis from Building Floor Plan Images",  IAPR Summer School on Document Analysis: Document Informatics, Jaipur, India (January 2017). Poster
  • Divya Sharma, Chiranjoy Chattopadhyay and Gaurav Harit; "A Unified Framework for Semantic based Architectural Floor plan Retrieval", International Conference on Pattern Recognition (ICPR), Cancun, Mexico (December, 2016). Poster
  • Divya Sharma, Chiranjoy Chattopadhyay and Gaurav Harit; "Retrieval of Architectural Floor plans based on Layout Semantics", Computer Vision and Pattern Recognition (Women in Computer Vision), Las Vegas, USA (June 2016). Poster