Object Detection Using Deep Convolutional Generative Adversarial Networks Embedded Single Shot Detector with Hyper-parameter Optimization

Ranjith Dinakaran; Li Zhang

doi:10.1109/SSCI50451.2021.9659855

Object Detection Using Deep Convolutional Generative Adversarial Networks Embedded Single Shot Detector with Hyper-parameter Optimization

Ranjith Dinakaran, Li Zhang

Computer Information Sciences Department

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Citations (Scopus)

Abstract

It is a challenging task to identify optimal network configurations for large-scale deep neural networks with cascaded structures. In this research, we propose a hybrid end-to-end model by integrating Deep Convolutional Generative Adversarial Network (DCGAN) with Single Shot Detector (SSD), for undertaking object detection. We subsequently employ the Particle Swarm Optimization (PSO) algorithm to conduct hyperparameter identification for the DCGAN-SSD model. The detected class labels as well as salient regional features are then used as inputs for a Long Short-Term Memory (LSTM) network for image description generation. Evaluated with a video data set in the wild, the empirical results indicate the efficiency of the proposed PSO-enhanced DCGAN-SSD object detector with respect to object detection and image description generation.

Original language	English
Title of host publication	2021 IEEE Symposium Series on Computational Intelligence SSCI Proceedings
Place of Publication	Piscataway, US
Publisher	IEEE
Number of pages	6
ISBN (Electronic)	9781728190488
ISBN (Print)	9781728190495
DOIs	https://doi.org/10.1109/SSCI50451.2021.9659855
Publication status	Published - 5 Dec 2021
Event	2021 IEEE Symposium Series on Computational Intelligence, SSCI 2021 - Orlando, United States Duration: 5 Dec 2021 → 7 Dec 2021

Conference

Conference	2021 IEEE Symposium Series on Computational Intelligence, SSCI 2021
Country/Territory	United States
City	Orlando
Period	5/12/21 → 7/12/21

Keywords

Deep Convolutional Generative Adversarial Network
Hyperparameter tuning
Long Short-Term Memory
Object detection
Particle Swarm Optimization
Recurrent Neural Network
Single Shot Detector

Access to Document

10.1109/SSCI50451.2021.9659855

Cite this

@inproceedings{a712e0098ace44bdacf439fd02fd5ccb,

title = "Object Detection Using Deep Convolutional Generative Adversarial Networks Embedded Single Shot Detector with Hyper-parameter Optimization",

abstract = "It is a challenging task to identify optimal network configurations for large-scale deep neural networks with cascaded structures. In this research, we propose a hybrid end-to-end model by integrating Deep Convolutional Generative Adversarial Network (DCGAN) with Single Shot Detector (SSD), for undertaking object detection. We subsequently employ the Particle Swarm Optimization (PSO) algorithm to conduct hyperparameter identification for the DCGAN-SSD model. The detected class labels as well as salient regional features are then used as inputs for a Long Short-Term Memory (LSTM) network for image description generation. Evaluated with a video data set in the wild, the empirical results indicate the efficiency of the proposed PSO-enhanced DCGAN-SSD object detector with respect to object detection and image description generation.",

keywords = "Deep Convolutional Generative Adversarial Network, Hyperparameter tuning, Long Short-Term Memory, Object detection, Particle Swarm Optimization, Recurrent Neural Network, Single Shot Detector",

author = "Ranjith Dinakaran and Li Zhang",

year = "2021",

month = dec,

day = "5",

doi = "10.1109/SSCI50451.2021.9659855",

language = "English",

isbn = "9781728190495",

booktitle = "2021 IEEE Symposium Series on Computational Intelligence SSCI Proceedings",

publisher = "IEEE",

address = "United States",

note = "2021 IEEE Symposium Series on Computational Intelligence, SSCI 2021 ; Conference date: 05-12-2021 Through 07-12-2021",

}

Dinakaran, R & Zhang, L 2021, Object Detection Using Deep Convolutional Generative Adversarial Networks Embedded Single Shot Detector with Hyper-parameter Optimization. in 2021 IEEE Symposium Series on Computational Intelligence SSCI Proceedings. IEEE, Piscataway, US, 2021 IEEE Symposium Series on Computational Intelligence, SSCI 2021, Orlando, United States, 5/12/21. https://doi.org/10.1109/SSCI50451.2021.9659855

Object Detection Using Deep Convolutional Generative Adversarial Networks Embedded Single Shot Detector with Hyper-parameter Optimization. / Dinakaran, Ranjith; Zhang, Li.
2021 IEEE Symposium Series on Computational Intelligence SSCI Proceedings. Piscataway, US: IEEE, 2021.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Object Detection Using Deep Convolutional Generative Adversarial Networks Embedded Single Shot Detector with Hyper-parameter Optimization

AU - Dinakaran, Ranjith

AU - Zhang, Li

PY - 2021/12/5

Y1 - 2021/12/5

N2 - It is a challenging task to identify optimal network configurations for large-scale deep neural networks with cascaded structures. In this research, we propose a hybrid end-to-end model by integrating Deep Convolutional Generative Adversarial Network (DCGAN) with Single Shot Detector (SSD), for undertaking object detection. We subsequently employ the Particle Swarm Optimization (PSO) algorithm to conduct hyperparameter identification for the DCGAN-SSD model. The detected class labels as well as salient regional features are then used as inputs for a Long Short-Term Memory (LSTM) network for image description generation. Evaluated with a video data set in the wild, the empirical results indicate the efficiency of the proposed PSO-enhanced DCGAN-SSD object detector with respect to object detection and image description generation.

AB - It is a challenging task to identify optimal network configurations for large-scale deep neural networks with cascaded structures. In this research, we propose a hybrid end-to-end model by integrating Deep Convolutional Generative Adversarial Network (DCGAN) with Single Shot Detector (SSD), for undertaking object detection. We subsequently employ the Particle Swarm Optimization (PSO) algorithm to conduct hyperparameter identification for the DCGAN-SSD model. The detected class labels as well as salient regional features are then used as inputs for a Long Short-Term Memory (LSTM) network for image description generation. Evaluated with a video data set in the wild, the empirical results indicate the efficiency of the proposed PSO-enhanced DCGAN-SSD object detector with respect to object detection and image description generation.

KW - Deep Convolutional Generative Adversarial Network

KW - Hyperparameter tuning

KW - Long Short-Term Memory

KW - Object detection

KW - Particle Swarm Optimization

KW - Recurrent Neural Network

KW - Single Shot Detector

UR - http://www.scopus.com/inward/record.url?scp=85125781877&partnerID=8YFLogxK

U2 - 10.1109/SSCI50451.2021.9659855

DO - 10.1109/SSCI50451.2021.9659855

M3 - Conference contribution

AN - SCOPUS:85125781877

SN - 9781728190495

BT - 2021 IEEE Symposium Series on Computational Intelligence SSCI Proceedings

PB - IEEE

CY - Piscataway, US

T2 - 2021 IEEE Symposium Series on Computational Intelligence, SSCI 2021

Y2 - 5 December 2021 through 7 December 2021

ER -

Object Detection Using Deep Convolutional Generative Adversarial Networks Embedded Single Shot Detector with Hyper-parameter Optimization

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this