Indoor home scene recognition through instance segmentation using a combination of neural networks

Amlan Basu; Keerati Kaewrak; Lykourgos Petropoulakis; Gaetano Di Caterina; John J. Soraghan

Indoor home scene recognition through instance segmentation using a combination of neural networks

Amlan Basu, Keerati Kaewrak, Lykourgos Petropoulakis, Gaetano Di Caterina, John J. Soraghan

Electronic And Electrical Engineering

Research output: Contribution to conference › Paper › peer-review

83 Downloads (Pure)

Abstract

This work presents a technique for recognizing indoor home scenes by using object detection. The object detection task is achieved through pre-trained Mask-RCNN (Regional Convolutional Neural Network), whilst the scene recognition is performed through a Convolutional Neural Network (CNN). The output of the Mask-RCNN is fed in input to the CNN, as this provides the CNN with the information of objects detected in one scene. So, the CNN recognizes the scene by looking at the combination of objects detected. The CNN is trained using the various object detection outputs of Mask-RCNN. This helps the CNN learn about the various combinations of objects that a scene can have. The CNN is trained using 500 combinations of 5 different scenes (bathroom, bedroom, kitchen, living room, and dining room) of the indoor home generated by Mask-RCNN. The trained network was tested on 24,000 indoor home scene images. The final accuracy produced by the CNN is 97.14%.

Original language	English
Number of pages	7
Publication status	Published - 19 Jun 2022
Event	2022 IEEE World Conference on Applied Intelligence and Computing - Rajkiya Engineering College, Sonbhadra, India Duration: 17 Jun 2022 → 19 Jun 2022 https://aic2022.scrs.in/

Conference

Conference	2022 IEEE World Conference on Applied Intelligence and Computing
Abbreviated title	AIC 2022
Country/Territory	India
City	Sonbhadra
Period	17/06/22 → 19/06/22
Internet address	https://aic2022.scrs.in/

Keywords

indoor home scenes
object recognition
scene recognition
instance segmentation
transfer learning
mask-RCNN
CNN

Access to Document

Basu-etal-AIC-2022-Indoor-home-scene-recognition-through-instance-segmentation-using-a-combination-of-neural-networksAccepted author manuscript, 1.06 MBLicence: Strath_1

Cite this

@conference{3ef985e8a8fc4c67b3dcb3e3c6ed87d6,

title = "Indoor home scene recognition through instance segmentation using a combination of neural networks",

abstract = "This work presents a technique for recognizing indoor home scenes by using object detection. The object detection task is achieved through pre-trained Mask-RCNN (Regional Convolutional Neural Network), whilst the scene recognition is performed through a Convolutional Neural Network (CNN). The output of the Mask-RCNN is fed in input to the CNN, as this provides the CNN with the information of objects detected in one scene. So, the CNN recognizes the scene by looking at the combination of objects detected. The CNN is trained using the various object detection outputs of Mask-RCNN. This helps the CNN learn about the various combinations of objects that a scene can have. The CNN is trained using 500 combinations of 5 different scenes (bathroom, bedroom, kitchen, living room, and dining room) of the indoor home generated by Mask-RCNN. The trained network was tested on 24,000 indoor home scene images. The final accuracy produced by the CNN is 97.14%.",

keywords = "indoor home scenes, object recognition, scene recognition, instance segmentation, transfer learning, mask-RCNN, CNN",

author = "Amlan Basu and Keerati Kaewrak and Lykourgos Petropoulakis and {Di Caterina}, Gaetano and Soraghan, {John J.}",

note = "{\textcopyright} 2022 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.; 2022 IEEE World Conference on Applied Intelligence and Computing, AIC 2022 ; Conference date: 17-06-2022 Through 19-06-2022",

year = "2022",

month = jun,

day = "19",

language = "English",

url = "https://aic2022.scrs.in/",

}

TY - CONF

T1 - Indoor home scene recognition through instance segmentation using a combination of neural networks

AU - Basu, Amlan

AU - Kaewrak, Keerati

AU - Petropoulakis, Lykourgos

AU - Di Caterina, Gaetano

AU - Soraghan, John J.

N1 - © 2022 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

PY - 2022/6/19

Y1 - 2022/6/19

N2 - This work presents a technique for recognizing indoor home scenes by using object detection. The object detection task is achieved through pre-trained Mask-RCNN (Regional Convolutional Neural Network), whilst the scene recognition is performed through a Convolutional Neural Network (CNN). The output of the Mask-RCNN is fed in input to the CNN, as this provides the CNN with the information of objects detected in one scene. So, the CNN recognizes the scene by looking at the combination of objects detected. The CNN is trained using the various object detection outputs of Mask-RCNN. This helps the CNN learn about the various combinations of objects that a scene can have. The CNN is trained using 500 combinations of 5 different scenes (bathroom, bedroom, kitchen, living room, and dining room) of the indoor home generated by Mask-RCNN. The trained network was tested on 24,000 indoor home scene images. The final accuracy produced by the CNN is 97.14%.

AB - This work presents a technique for recognizing indoor home scenes by using object detection. The object detection task is achieved through pre-trained Mask-RCNN (Regional Convolutional Neural Network), whilst the scene recognition is performed through a Convolutional Neural Network (CNN). The output of the Mask-RCNN is fed in input to the CNN, as this provides the CNN with the information of objects detected in one scene. So, the CNN recognizes the scene by looking at the combination of objects detected. The CNN is trained using the various object detection outputs of Mask-RCNN. This helps the CNN learn about the various combinations of objects that a scene can have. The CNN is trained using 500 combinations of 5 different scenes (bathroom, bedroom, kitchen, living room, and dining room) of the indoor home generated by Mask-RCNN. The trained network was tested on 24,000 indoor home scene images. The final accuracy produced by the CNN is 97.14%.

KW - indoor home scenes

KW - object recognition

KW - scene recognition

KW - instance segmentation

KW - transfer learning

KW - mask-RCNN

KW - CNN

M3 - Paper

T2 - 2022 IEEE World Conference on Applied Intelligence and Computing

Y2 - 17 June 2022 through 19 June 2022

ER -

Indoor home scene recognition through instance segmentation using a combination of neural networks

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this