Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

Online Learning for Indoor Asset Detection

Abstract: 

Building floor plans with locations of safety, security, and energy assets such as IoT sensors, fire alarms, etc. are vital for climate control, emergency response, safety, and maintenance of building infrastructure. Existing approaches to building survey are tedious, error prone, and involve an operator with a clipboard and pen, enumerating and localizing assets in each room. We propose an interactive method for a human operator to use an app on a smartphone, which can detect, classify, and localize assets of interest, to expedite such a task. We must overcome the fact that appearances of a single type of asset, e.g. power outlet, vary greatly from building to building or even from room to room. We propose an online "one-shot learning" approach using a Neural Turing Machine (NTM) architecture with augmented memory capacity, which allows us to rapidly incorporate new data into our model, improving prediction accuracy after only a few examples, without compromising its ability to remember previously learned data. This approach reduces the training time needed to update the model between building survey sessions by up to a factor of 10. Experiments show that our proposed method outperforms the prediction accuracy attained by using more traditional, batch processing deep learning methods where new data is combined with all old data to train the model. The advantage is especially pronounced for assets in never-before-seen buildings.

up
1 user has voted: Adith Balamurugan

Paper Details

Authors:
Adith Balamurugan, Avideh Zakhor
Submitted On:
14 October 2019 - 8:33pm
Short Link:
Type:
Presentation Slides
Event:
Presenter's Name:
Adith Balamurugan
Paper Code:
48
Document Year:
2019
Cite

Document Files

OneShotMLSP.pdf

(18)

Subscribe

[1] Adith Balamurugan, Avideh Zakhor, "Online Learning for Indoor Asset Detection", IEEE SigPort, 2019. [Online]. Available: http://sigport.org/4872. Accessed: Nov. 19, 2019.
@article{4872-19,
url = {http://sigport.org/4872},
author = {Adith Balamurugan; Avideh Zakhor },
publisher = {IEEE SigPort},
title = {Online Learning for Indoor Asset Detection},
year = {2019} }
TY - EJOUR
T1 - Online Learning for Indoor Asset Detection
AU - Adith Balamurugan; Avideh Zakhor
PY - 2019
PB - IEEE SigPort
UR - http://sigport.org/4872
ER -
Adith Balamurugan, Avideh Zakhor. (2019). Online Learning for Indoor Asset Detection. IEEE SigPort. http://sigport.org/4872
Adith Balamurugan, Avideh Zakhor, 2019. Online Learning for Indoor Asset Detection. Available at: http://sigport.org/4872.
Adith Balamurugan, Avideh Zakhor. (2019). "Online Learning for Indoor Asset Detection." Web.
1. Adith Balamurugan, Avideh Zakhor. Online Learning for Indoor Asset Detection [Internet]. IEEE SigPort; 2019. Available from : http://sigport.org/4872