Shuo Meng is a PhD candidate of Biomedical Engineering at City University of Hong Kong. He aims to solve the practical engineering problems. For example, he leads the “fabric recognition” project (”织识“), which is using in the actural production and is highly praised.
Although his major is not computer science, he studied it systematically out of a strong passion. He used deep learning and other computer technologies to solve problems in Textile area, such as fabric defect detection, fabric smoothness evaluation, fabric images style transfer, fabric simulation, etc
.
Shuo Meng is now working for medical signal processing under Prof. Hu Jinlian.
Download my resumé.
Degree: PhD candidate of Biomedical Engineering
GPA: Not Applicable
Degree: Master of Textile Engineering
GPA: 86.8 out of 100, ranking top 5%
Degree: Bachelor of Textile Engineering (Minor in International Economics and Trade)
GPA: 3.58 out of 4, ranking top 5%
70%
80%
70%
Programming:
Frameworks:
Languages:
Certificates:
A graduate project granted under the Postgraduate Research & Practice Innovation of Jiangsu Province 1062050205206039.
The project adopted a wireless device and established a fabric images dataset with elaborated structure parameters. Presented a multi-task and multi-scale convolutional neural network (MTMSnet) to recognize the fabric density, weaving pattern, reaching under 2% and 10% error respectively.
The project is called “织识”, which is now online. Some of the source can be found in my “Github”
A project came from the competition of Tianchi 2019 Guangdong Intelligent Identification of Cloth Defects.
Proposed a two-stage strategy to do local defect prediction and global defect recognition by using the Inception-VI model and LeNet-5 model, reaching 93.2% accuracy based on our cropped dataset.
The decoloration problem was solved by a paired image-to-image translation model built by conditional generative adversarial networks (GAN), which is also promising for fabric defect detection and other problems.
An ordinal classification framework based on label noise estimation (OCF-LNE) to objectively evaluate the fabric smoothness appearance degree. In this project, I helped to establish the dataset and the paper can be found in https://doi.org/10.1109/ACCESS.2019.2959705
A graduate project which was a web-based production management system specially designed for textile production enterprises, which also involved data-collecting devices based on Arduino to monitor weaving process in real-time.
Responsible for the back-end and front-end development and the thesis “Development of APS Software for Scheduling of Multi-varieties Weaving Production” won the “Excellent Theses of Jiangnan University”.
The source code can be found in my GitHub. A demo project can be found in http://www.jntex.cn:88/
Articles:
Patents: