Automatic Segmentation Method of Garment Figure Based on Convolutional Neural Network

DOI

10.2991/amms-17.2017.55How to use a DOI?

Keywords

computer vision; image segmentation; convolutional neural networks; multi-scale feature fusion; residual connection

Abstract

Aiming to the problem that the rapid acquisition of clothing style map requires manual participation and takes a long time and heavy task, an automatic image segmentation method based on convolution neural network is proposed. Firstly, we denoised normalized and semantically annotated the sequence image in order to produce the data set. Then, we trained the convolution neural networks, which were improved with fusion multi-scale feature and residual connection, and obtained the optimized convolution neural network segmentation model. Finally, it loaded the pre-segmentation image into the optimized model to get the normalized mask pattern, and used the cubic spline interpolation methods to restore the resolution and result of the HD segmentation with the original mask. In this paper, the results of Photoshop segmentation are as the reference standard. The experimental results show that the accuracy of the method is close to the reference standard, and the batch segmentation can be realized automatically, which can solve the heavy problem of the target segmentation task in 3D reconstruction.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Chong Chen
PY  - 2017/11
DA  - 2017/11
TI  - Automatic Segmentation Method of Garment Figure Based on Convolutional Neural Network
BT  - Proceedings of the 2017 International Conference on Applied Mathematics, Modeling and Simulation (AMMS 2017)
PB  - Atlantis Press
SP  - 249
EP  - 252
SN  - 1951-6851
UR  - https://doi.org/10.2991/amms-17.2017.55
DO  - 10.2991/amms-17.2017.55
ID  - Chen2017/11
ER  -