TY - GEN
T1 - Indoor Semantic Segmentation for Robot Navigating on Mobile
AU - Kim, Wonsuk
AU - Seok, Junhee
N1 - Publisher Copyright:
© 2018 IEEE.
PY - 2018/8/14
Y1 - 2018/8/14
N2 - In recent years, there have been many successes of using Deep Convolutional Neural Networks (DCNNs) in the task of pixel-level classification (also called 'semantic image segmentation'). The advances in DCNN have led to the development of autonomous vehicles that can drive with no driver controls by using sensors like camera, LiDAR, etc. In this paper, we propose a practical method to implement autonomous indoor navigation based on semantic image segmentation using state-of-the-art performance model on mobile devices, especially Android devices. We apply a system called 'Mobile DeepLabv3', which uses atrous convolution when applying semantic image segmentation by using MobileNetV2 as a network backbone. The ADE20K dataset is used to train our models specific to indoor environments. Since this model is for robot navigating, we re-label 150 classes into 20 classes in order to easily classify obstacles and road. We evaluate the trade-offs between accuracy and computational complexity, as well as actual latency and the number of parameters of the trained models.
AB - In recent years, there have been many successes of using Deep Convolutional Neural Networks (DCNNs) in the task of pixel-level classification (also called 'semantic image segmentation'). The advances in DCNN have led to the development of autonomous vehicles that can drive with no driver controls by using sensors like camera, LiDAR, etc. In this paper, we propose a practical method to implement autonomous indoor navigation based on semantic image segmentation using state-of-the-art performance model on mobile devices, especially Android devices. We apply a system called 'Mobile DeepLabv3', which uses atrous convolution when applying semantic image segmentation by using MobileNetV2 as a network backbone. The ADE20K dataset is used to train our models specific to indoor environments. Since this model is for robot navigating, we re-label 150 classes into 20 classes in order to easily classify obstacles and road. We evaluate the trade-offs between accuracy and computational complexity, as well as actual latency and the number of parameters of the trained models.
KW - Atrous Convolution
KW - Convolutional Neural Networks
KW - Indoor Navigation
KW - Semantic Segmentation
UR - http://www.scopus.com/inward/record.url?scp=85052521968&partnerID=8YFLogxK
U2 - 10.1109/ICUFN.2018.8436956
DO - 10.1109/ICUFN.2018.8436956
M3 - Conference contribution
AN - SCOPUS:85052521968
SN - 9781538646465
T3 - International Conference on Ubiquitous and Future Networks, ICUFN
SP - 22
EP - 25
BT - ICUFN 2018 - 10th International Conference on Ubiquitous and Future Networks
PB - IEEE Computer Society
T2 - 10th International Conference on Ubiquitous and Future Networks, ICUFN 2018
Y2 - 3 July 2018 through 6 July 2018
ER -