13 0

SGNet: Design of Optimized DCNN for Real-Time Face Detection

Title
SGNet: Design of Optimized DCNN for Real-Time Face Detection
Author
조인휘
Keywords
DCNN; Scalable face detection; Transfer learning; Grid-based one-shot detection method
Issue Date
2019-02
Publisher
Springer Verlag
Citation
Communications in Computer and Information Science, v. 931, Page. 200-209
Abstract
This paper proposes to optimize the deep convolution neural networks for real time video processing on detecting faces and facial landmarks. For that, we have to reduce the existing weight size and duplication of weight parameters. By utilizing the strengths of the two previous powerful algorithms which have shown the best performance, we overcome the weakness of the existing methods. Instead of using the old-fashioned searching method like sliding window, we propose our grid-based one-shot detection method. Furthermore, instead of forwarding one image frame through a very deep CNN, we divide the process into 3 stages for incremental detection improvements to overcome the existing limitation of grid-based detection. After lots of experiments with different frameworks, deep learning frameworks are chosen as the best for integration of 3-stage DCNN. By using transfer learning, we can remove the unnecessary convolution layers in the existing DCNN and retrain hidden layers repeatedly and finally succeed in obtaining the best speed and accuracy which can run on the embedded platform. The performance to find small sized faces is better than YOLO v2.
URI
https://link.springer.com/chapter/10.1007%2F978-981-13-5907-1_21http://repository.hanyang.ac.kr/handle/20.500.11754/108275
ISBN
978-981135906-4; 978-981-13-5907-1
ISSN
1865-0929
DOI
10.1007/978-981-13-5907-1_21
Appears in Collections:
COLLEGE OF ENGINEERING[S](공과대학) > COMPUTER SCIENCE(컴퓨터소프트웨어학부) > Articles
Files in This Item:
There are no files associated with this item.
Export
RIS (EndNote)
XLS (Excel)
XML


qrcode

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

BROWSE