劉維恆
:wayne270838@gmail.com
:+886-972-793207
國立陽明交通大學, 碩士學位, 人工智慧技術與應用碩士學位學程 (電子所) 2021年9月 ~ 至今
- 已修習深度學習與實務, 機器學習, 應⽤電腦視覺, 演算法,計算機結構,最佳化理論與實務,影像處理,視訊串流與追蹤
- iVSLab智慧視覺系統設計實驗室 (指導教授: 郭峻因)
- 研究⽅向: semi-supervised Camera/Radar Sensor Fusion on Obejct detection Algorithm
- Wistron: Project: Early Fusion (concatenate the vision image and radar image first )
- MTK Project: Proposed Camera/Radar Sensor Fusion Method on semi-supervised learning for Object detection
- 暑期實習: 世界先進擔任影像演算法工程師
國立台北科技大學, 學士學位, 電機 2017年9月 ~ 2021年6月
-
畢業專題: Implement Hybrid Densely Connected U-Net for Liver and Tumor Auto Segmentation from CT Volumes
- 學期校外實習: Engineering intern ( Barco )
- Sensor Fusion
- MTK project
- 研發 Camera/Radar Fusion Method on semi-supervised learning for Object detection ( Hybrid Fusion )
- Hybird fusion with radar/camera on Nvidia AGX Xavier
- By combining camera/radar inforamtion, the impact of the severe weather on the only camera is improved.
- 緯創嵌入式⼈⼯智慧研究中⼼計畫 -交通運輸⼈⼯智慧多元感知融合處理系統
- Deep Learning based Camera/Rader Sensor Fusion on RSU for Target Detection with Scaled-YOLOv4
- AI Model
- Background: Optimization, Machine Learning, Deep Learning, Applied Computer vision,Image Processing,Video streaming and tracking
- Programming: Pytorch, TensorFlow2.0
- Platform: Xavier, Linux(Ubuntu)
- Model: YOLOv3, YOLOv4, Scaled-YOLOv4,YOLOX,YOLOv7
- Basic Skills
- Python ( AI model development, Image Processing,TensorRT)/shell script
- Git (source control)
-
Course Project
- Video streaming and tracking(pytorch/javascript)
- Final Project: Live Streaming Server and perform tasks using deep learning model
- Deep Learning and Practice (pytorch)
- LAB ( RNN, AutoEncoder, GAN)
- Final Project: YOLOF optimization
- Image Processing (C++ without opencv)
-
LAB(Image Constrast Enhansement,White Balance, Image Restoration)
- Final Project: water body segmentation (tradition Method/Deep Learning )
- Machine Learning(python without library)
-
LAB ( Gaussian Process for regression, Kmeans, SVM, Gaussian Mixture Model )