Curriculum Vitae

Education

2019 Jan - 2020 Mar(Intern), 2020 Jul - Present(FTE): Computer Vision Researcher and Engineer
- AI Lab, Bytedance Inc.
- Duties included: Developing services frameworks, video render based on OpenGL, research topics in human pose transfer, video object segmentation, font effect generation(Deeplearning).
- Supervisor: Ke Li, Yuning Jiang
2016 Jul - 2016 Aug: Web Front End Developer Intern
- Boonray.
- Duties included: Developing a web app to present 3D mesh models of roofs which will be built a Distributed photovoltaic power station through Three.js, including measuring roofs, simulation of placing Photovoltaic boards, automatically generating investigation report.
- This webapp increasing 50% income than before in sales of roof-measuring-UAVs.

2020 Apr - 2020 Jul: DeepLearning Method in Pattern Generation
- Using PG-GAN, DC-GAN to generate patterns with artists styles, training keypoint detection networks, deploying services includes super-resolution, image style transfer, color-transfer, border extraction.
- Email to me for more information.
- Supervisor: Associate Professor Yilu Zhou

2018 Sep - 2019 Jul: DeepLearning Method in 3D Medical Image Segmentation
- SIST, ShanghaiTech University
- Designed a new framework to consider the 3d information, it provided extra embedded information in the decoding module. I also considered an axis-aligned trident network.
- Email to me for more information.
- Supervisor: Associate Professor Xuming He.
2019 Jul - 2019 Aug: Dynamic Font Effect Transfer
- AI Lab, Bytedance Inc.
- Implemented font effect transfer algorithm individually. To transfer one natural dynamic effect by computing the nearest neighbour field of reference dynamic image and target generated image. This work saved time for our graphics designer and enriched our dynamic effects font library.
- Supervisor: Mr. Yuning Jiang
2019 Aug - 2020 Mar, 2020 Jul - Present: Video Effect Rendering
- AI Lab, Bytedance Inc.
- Developed video effects, transition effects, green-screen effects, artificial intelligence effects rendering framework, based on OpenGL, accelerated by GPU. In the process of software upgrades, I learned to deal with exceptions, avoid server stress, optimize frameworks by four revisions, and object-oriented programming. The construction of the video effects rendering framework has enabled the company’s advertising business traffic to increase significantly, providing interfaces for computer vision technology access.
- Supervisor: Dr. Ke Li

Yifang Men, Yiming Mao, Yuning Jiang, Weiying Ma and Zhou Lian. Controllable Human Image Synthesis based on Attribute-Decomposed GAN. In Proceedings of the IEEE conference on computer vision and pattern recognition, 2020. [project page] [paper]

Coding
- Strongly skills in Python3.6+/Python2.7
- Medium skills in R language, golang, Matlab, Arduino, JavaScript, Java, C/C++
DeepLearning Frameworks
- PyTorch1.3.0+
- Tensorflow2.0(Keras)
- Numpy, OpenCV, Scikit-learn, Pandas, Matplotlib
Other Frameworks
- The Apache Thrift software framework.(Scalable cross-language services development, clients and servers based on Python3.6)
- Tensorflow Serving. (Deploy a DeepLearning model on a GPU-server)
Other Skills
- Program Architecture.(See my projects in Github for insights.)
- Multi-Processing/Threading Programing.