Optimal Number of Cue Objects for Photo-Based Indoor Localization Youngsun Chung Yonsei University, Korea, Republic of Daeyoung Gil Yonsei University, Korea, Republic of Ghang Lee Yonsei University, Korea, Republic of This is a section of CONVR 2023 - Proceedings of the 23rd International Conference on Construction Applications of Virtual Reality (DOI: 10.36253/979-12-215-0289-3) by Pietro Capone, Vito Getuli, Farzad Pour Rahimian, Nashwan Dawood, Alessandro Bruttini, Tommaso Sorbi Firenze University Press Florence 2023 https://doi.org/10.36253/10.36253/979-12-215-0289-3.98

Available for academic research purposes

Open Access

Copyright Author(s)

Content licence CC BY-NC 4.0

Metadata licence CC0 1.0

This is original content, published for academic research purposes

 Digital edition XML powered by Booksflow

Building information modeling (BIM) is widely used to generate indoor images for indoor localization. However, changes in camera angles and indoor conditions mean that photos are much more changeable than BIM images. This makes any attempt at localization based on the similarity between real photos and BIM images challenging. To overcome this limitation, we propose a reasoning-based approach for determining the location of a photo by detecting the cue objects in the photo and the relationships between them. The aim of this preliminary study was to determine the optimal number of cue objects required for an indoor image. If there are too few cue objects in an indoor image, it results in an excessive number of location candidates. Conversely, if there are too many cue objects, the accuracy of object detection in an image decreases. Theoretically, a larger number of cue objects would improve the reasoning process; however, too many cue objects could lead to declining object detection performance. The experimental results demonstrated that of two to five cue objects, three cue objects is most likely to yield optimal performance

 indoor location determination BIM reasoning

It is available online at https://doi.org/10.36253/10.36253/979-12-215-0289-3.98

References Acharya, D., Khoshelham, K., & Winter, S. (2019). BIM-PoseNet: Indoor camera localisation using a 3D indoor model and deep learning from synthetic images. ISPRS Journal of Photogrammetry and Remote Sensing, 150, 245–258. 10.1016/j.isprsjprs.2019.02.020 Acharya, D., Roy, S., Khoshelham, K., & Winter, S. (2020). A recurrent deep network for estimating the pose of real indoor images from synthetic image sequences. Sensors, 20(19), 1–20. 10.3390/s20195492 Alam, M., Hossain, A. K. M., & Mohamed, F. (2022). Performance evaluation of recurrent neural networks applied to indoor camera localization. International Journal of Emerging Technology and Advanced Engineering, 12(8), 116–124. 10.46338/ijetae0822_15 Bay, H., Tuytelaars, T., & Van Gool, L. (2006). SURF: Speeded up robust features. In A. Leonardis, H. Bischof, & A. Pinz (Eds.), Computer Vision – ECCV 2006 (pp. 404–417). Springer. 10.1007/11744023_32 Guan, K., Ma, L., Tan, X., & Guo, S. (2016). Vision-based indoor localization approach based on SURF and landmark. 2016 International Wireless Communications and Mobile Computing Conference (IWCMC), (pp. 655–659). 10.1109/IWCMC.2016.7577134 Ha, I., Kim, H., Park, S., & Kim, H. (2018). Image retrieval using BIM and features from a pretrained VGG network for indoor localization. Building and Environment, 140, 23–31. 10.1016/j.buildenv.2018.05.026 Kang, H., Park, Y., & Kim, Y. (2019). Improvement model of defect information management system for apartment buildings. Korean Journal of Construction Engineering and Management, 20(4), 13–21. 10.6106/KJCEM.2019.20.4.013 Kim, D., & Kim, J. (2023). CT-Loc: Cross-domain visual localization with a channel-wise transformer. Neural Networks, 158, 369–383. 10.1016/j.neunet.2022.11.014 Kim, J. (2022). Identifying indoor locations of close-up photos using deep learning and building information modeling objects. Yonsei University. http://www.riss.kr/link?id=T16372630 Kim, K.-T., Lim, M.-G., & Kim, G.-T. (2014). History management technology of building construction and maintenance using vector photo information and BIM. Journal of the Korea Institute of Building Construction, 14(6), 605–613. 10.5345/JKIBC.2014.14.6.605 Li, Y., Kambhamettu, R. H., Hu, Y., & Zhang, R. (2022). ImPos: An image-based indoor positioning system. 2022 IEEE 19th Annual Consumer Communications & Networking Conference (CCNC), (pp. 144–150). 10.1109/CCNC49033.2022.9700699 Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016). You only look once: Unified, real-time object detection. Computer Vision and Pattern Recognition Conference (CVPR), (pp. 779-788). https://openaccess.thecvf.com/content_cvpr_2016/html/Redmon_You_Only_Look_CVPR_2016_paper.html Simonyan, K., & Zisserman, A. (2015). Very deep convolutional networks for large-scale image recognition. arXiv. 10.48550/arXiv.1409.1556