Wen-Yen Chang completed his master’s degree at the department of electrical engineering at National Tsing Hua University, Taiwan in 2020. He was a member of Vision Science Lab (VSLab) advised by Prof. Min Sun and Dr. Ting-Fan Wu. His research interests are centered on computer vision, machine learning, especially in deep learning. In his recently industry experience, He had worked in Qualcomm XR Reearch department 3.5 years develop and optimized deep learning feature on XR edge device with NPU(DSP). Specifically, he own quantization feature impacts Qualcomm XR device reduce at least 75% power consumption and 80% latency with NPU.
I’m actively looking for permanent positions. Seeking AI Engineer / Data Scientist / Software Engineer position.
Qualcomm XR Research system team in Taiwan. 1. Model quantization in order to optimize target products’ latency/power by low-bits (W4A8, W8A8). 2. Model architecture selection and operation fusion. • Applications (The Comparison are FP16 model on NPU) 1. Monocular Depth Prediction, reduce 82% latency , 78% power and improve 12% accuracy. 2. Segmentation for Instance and background, reduce 81% latency, 75% power and keep accuracy. |
July 2021 - Nov. 2024 |
Military Service in Taiwan. |
Sep. 2020 - Dec. 2020 |
Machine Learning Research Intern (Consultant) |
May 2020 - Sep. 2020 |
DSP Summer Intern |
July 2019 - Aug. 2019 |
M.Sc. in Electronic Engineering
(4.30/4.30)
National Tsing Hua University | Hsinchu
|
Feb. 2018 - June 2020 |
B.Sc. in Electronic Engineering
(4.14/4.30)
National Chung Cheng University | Chiayi
|
Sep. 2014 - Feb. 2018 |
Phi Tau Phi Scholastic Honor
top 1 master student in the electrical engineering department of NTHU. |
2018 - 2020 |
Appier Scholarship
for outstanding students in their research with top conference papers. (research:ECCV 2018) |
2018 |
The High Distinction Award
the undergraudated project is award to 1st prizes in graduation exhibition of electronic engineering, CCU. |
2017 |
The Chair Award
the most potential product for undergraudated project in the graduation exhibition of electronic engineering in CCU. |
2017 |
Enhance data selection efficiency with variational auto-encoder for object detection’s active learning Wen-Yen Chang Master Thesis [1] [abs] [pdf] |
|
360-Indoor: Towards Learning Real-World Objects in 360deg Indoor Equirectangular Images Shih-Han Chou, Cheng Sun, Wen-Yen Chang, Wan-Ting Hsu, Min Sun, Jianlong Fu WACV 2020 [2] [abs] [pdf] |
Bias-Aware Heapified Policy for Active Learning Wen-Yen Chang, Wen-Huan Chiang, Shao-Hao Lu, Tingfan Wu, Min Sun CVGIP 2019 [3] [abs] [pdf] |
Leveraging Motion Priors in Videos for Improving Human Segmentation Yu-Ting Chen, Wen-Yen Chang, Hai-Lun Lu, Tingfan Wu, Min Sun ECCV 2018 [4] [abs][pdf] [code] |
JoRaLe Auto-Cinemagraph Tsun-Hsuan Wang, Wen-Yen Chang, Chun-Hung Chao [5] [abs][pdf] [slide] |
|
3D Game world with C++ Wen-Yen Chang, Chun-Hung Chao, Wei-Cheng Tseng, Jason Wu [6] [abs][video] |
|
Taxi-bidding simulation(auction rule) Meng-Li Shih, Wen-Yen Chang, Chun-Hung Chao [7] [pdf] [video] |
Car tracking and finding service in parkinglot system Wen-Yen Chang Undergraudated project [8] [abs][pdf] [video] |
|
Home Care App. Wen-Yen Chang, Kang-Yu Liou, Kuan-Yeh Wu, Ya-Chu Chang, Po-Hsien Wang, Ting-Ying Wang [9] [abs][pdf] |
Programming Languages | C, C++, C#, Java, Python, Matlab, Assembly |
Deep Learning Frameworks | Caffe, Tensorflow, Pytorch |
Web Tools | html, css, javascript, ruby |
UI Tools | MFC, QT |
Hardware/Firmware (wo/ UNIX OS) | 8051, Arduino, Nu-LB-NUC140 |
Hardware/Firmware (w/ UNIX OS) | Raspberry-Pi zero, TX2(ROS) |
Hardware descript language | Verilog |
Misc. | OpenCV, OpenGL, Github, Docker, Vim, Linux, LATEX |
Last updated on 2025-03-25