//assets.cakeresume.com/assets/contentarea/contentarea-2d64eca8dae2a1ffb31e0fb1df887bbad072c7c19734d7d30fa14884bbcc299a.css

<html><head></head><body><div class="row snippet-profile-001"><div class="col-sm-2"><img data-no-retina="true" src="https://res.cloudinary.com/cakeresume/image/upload/s--Pf42lIm5--/c_fill,fl_png8,g_face,h_300,w_300/v1555833556/kcauvs1oml3zztykrr4u.png" alt="Kcauvs1oml3zztykrr4u" loading="" style="aspect-ratio: 300 / 300;"></div><div class="col-sm-10"><h1><b>蔡承哲 (Cheng-Che, Tsai)</b></h1><p>E-mail: tcfsh010128@gmail.com</p><p><br></p></div></div><div class="row snippet-educations-013 snippet-experiences-013"><div class="col-xs-12"><div class="row"><div class="col-xs-12"><h1><b style="font-size: 28px; color: rgb(255, 156, 38);">學歷</b></h1></div></div><div class="row" title=""><div class="col-xs-2 col-sm-1 item-bullet"></div><div class="col-xs-10 col-sm-11 item"><h3>碩士 國立臺灣大學&nbsp;電子工程學研究所 電子設計自動化組 2019/9 ~ 現今 GPA&nbsp;4.13 / 4.30</h3><div style="font-size: 18px;">指導教授： 郭斯彥教授</div></div></div><div class="row" title=""><div class="col-xs-2 col-sm-1 item-bullet"></div><div class="col-xs-10 col-sm-11 item"><h3>學士 國立中央大學&nbsp;電機工程學系&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; 2015 ~ 2019&nbsp; &nbsp;GPA 3.64 / 4.00</h3></div></div></div></div><div class="row snippet-experiences-013"><div class="col-xs-12"><div class="row"><div class="col-xs-12"><h1><b style="font-size: 28px; color: rgb(255, 156, 38);">碩士論文</b><br></h1></div></div><div class="row" title=""><div class="col-xs-2 col-sm-1 item-bullet"></div><div class="col-xs-10 col-sm-11 item"><h3><span style="font-size: 22px;"><b>基於輪廓波的泛用除雨條紋深度神經網路</b></span></h3><div style="font-size: 18px;">有雨場景拍到的圖片含有不同方向的雨以及原本場景的邊緣，現有方法無法很好的分離雨和背景，經實驗發現，透過輪廓波可以捕捉到不同方向、尺寸特徵的性質以及雨條紋，將除雨的問題變成還原輪廓波轉換得到的係數，並透過具有不同尺度感受野的卷積神經網路 (CNN) 得到極佳的還原結果；損失函數的設計上，根據輪廓波多階層的特性提出了多階層判別器 (Multi-level discriminator)，使還原的結果更逼真，實驗結果顯示，提出的方法不論在合成或現實的資料集皆得到最好的還原結果。</div><div style="font-size: 18px;">準備投稿至&nbsp;<span style="font-size: 18px;">IEEE Transactions on Image Processing</span></div></div></div></div></div><div class="row snippet-experiences-013"><div class="col-xs-12"><div class="row"><div class="col-xs-12"><h1><b style="font-size: 28px; color: rgb(255, 156, 38);">論文發表</b><br></h1></div></div><div class="row" title=""><div class="col-xs-2 col-sm-1 item-bullet"></div><div class="col-xs-10 col-sm-11 item"><h3><b style="font-size: 20px;">ContourletNet: A Generalized Rain Removal Architecture Using Multi-Direction Hierarchical Representation</b><br></h3><h3 style="font-size: 16px;"><span style="font-size: 16px;"><u>Cheng-Che Tsai</u></span>, Wei-Ting Chen, Hao-Yu Feng, I-Hsiang Chen, Jian-Jiun Ding, and Sy-Yen Kuo</h3><h3><span style="font-size: 16px;">British Machine Vision Conference (BMVC), 2021.</span><br></h3></div></div><div class="row" title="" style="display: block;"><div class="col-xs-2 col-sm-1 item-bullet"></div><div class="col-xs-10 col-sm-11 item"><h3><b>ALL Snow Removed: Single Image Desnowing Algorithm Using Hierarchical Dual-tree Complex Wavelet Representation and Contradict Channel Loss</b><br></h3><h3><span style="font-size: 16px;">Wei-Ting Chen, Hao-Yu Feng, <u>Cheng-Che Tsai</u>, Cheng-Lin Hsieh, I-Hsiang Chen, Jian-Jiun Ding, and Sy-Yen Kuo<br></span></h3><h3><span style="font-size: 16px;">IEEE&nbsp;International Conference on Computer Vision&nbsp;(ICCV), 2021.</span><br></h3></div></div><div class="row" title="" style="display: block;"><div class="col-xs-2 col-sm-1 item-bullet"></div><div class="col-xs-10 col-sm-11 item"><h3><b>All Characteristics Preservation: Single Image Dehazing based on Hierarchical Detail Reconstruction Wavelet Decomposition Network</b></h3><div><span style="font-size: 16px;"><span style="color: inherit; font-family: &quot;open sans&quot;, &quot;microsoft jhenghei&quot;, sans-serif; letter-spacing: 0px;">Wei-Ting Chen, Hao-Yu Feng, </span><span style="color: inherit; font-family: &quot;open sans&quot;, &quot;microsoft jhenghei&quot;, sans-serif; letter-spacing: 0px;"><u>Cheng-Che Tsai</u></span><span style="color: inherit; font-family: &quot;open sans&quot;, &quot;microsoft jhenghei&quot;, sans-serif; letter-spacing: 0px;">, Jian-Jiun Ding, and Sy-Yen Kuo</span><br></span></div><h3><span style="font-size: 16px;">IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Sep., 2021.</span><br></h3></div></div><div class="row" title="" style="display: block;"><div class="col-xs-2 col-sm-1 item-bullet"></div><div class="col-xs-10 col-sm-11 item"><h3><b>JSTASR: Joint Size and Transparency-Aware Snow Removal Algorithm Based on Modified Partial Convolution and Veiling Effect Removal</b></h3><div><span style="font-size: 16px;">Wei-Ting Chen, Hao-Yu Feng, Jian-Jiun Ding, <u>Cheng-Che Tsai</u>, and Sy-Yen Kuo<br></span></div><h3><span style="font-size: 16px;">European Conference on Computer Vision (ECCV), Aug., 2020.</span><br></h3></div></div></div></div><div class="row snippet-paragraph-image-061" style="font-size: 30px;"><div class="col-sm-4"><b><span style="color: rgb(255, 156, 38);">修課紀錄</span></b><div><ul><li><span style="font-size: 22px; font-weight: inherit;">電腦視覺</span><br></li></ul></div></div><div class="col-sm-4"><div><br></div><div><ul><li><span style="font-size: 22px;">演算法</span></li></ul></div></div><div class="col-sm-4"><div><br></div><div><ul><li><span style="font-size: 22px; font-weight: inherit;">時頻分析和小波轉換</span><br></li></ul></div></div></div><div class="row snippet-paragraphs-002"><div class="col-sm-6"><h1><b><span style="font-size: 28px;"><span style="color: rgb(255, 156, 38);">程式語言</span></span></b></h1><p style="font-size: 20px;">Python, Matlab, C++</p></div><div class="col-sm-6"><h1><b><span style="font-size: 28px;"><span style="color: rgb(255, 156, 38);">機器學習框架</span></span></b></h1><p style="font-size: 20px;">Pytorch, Keras</p></div></div><div class="row snippet-profile-071"><div class="col-sm-12"><h1><br></h1></div></div><div class="row snippet-profile-071"><div class="col-sm-12"><h1><br></h1></div></div><div class="row snippet-profile-071"><div class="col-sm-12"><h1><br></h1></div></div><div class="row snippet-profile-071"><div class="col-sm-12"><h1><br></h1></div></div></body></html>

蔡承哲 (Cheng-Che, Tsai) E-mail: tcfsh010128@gmail.com 學歷 碩士 國立臺灣大學 電子工程學研究所 電子設計自動化組 2019/9 ~ 現今 GPA 4.13 / 4.30 指導教授： 郭斯彥教授 學士 國立中央大學 電機工程學系 2015 ~ ...

My CakeResume

Studying in NCU Dept. EE. Going to enroll in NTU GIEE on Sep 2019. 

#Research direction: Circuit Design + Program => Electronic Design Automation

#Work Experience
Intern at Media System Lab, NCU
QA Intern at CloudMosa Inc

蔡承哲 (Cheng-Che, Tsai)

學歷

碩士國立臺灣大學電子工程學研究所電子設計自動化組 2019/9 ~ 現今 GPA 4.13 / 4.30

學士國立中央大學電機工程學系 2015 ~ 2019 GPA 3.64 / 4.00

碩士論文

基於輪廓波的泛用除雨條紋深度神經網路

論文發表

ContourletNet: A Generalized Rain Removal Architecture Using Multi-Direction Hierarchical Representation

Cheng-Che Tsai, Wei-Ting Chen, Hao-Yu Feng, I-Hsiang Chen, Jian-Jiun Ding, and Sy-Yen Kuo

British Machine Vision Conference (BMVC), 2021.

ALL Snow Removed: Single Image Desnowing Algorithm Using Hierarchical Dual-tree Complex Wavelet Representation and Contradict Channel Loss

Wei-Ting Chen, Hao-Yu Feng, Cheng-Che Tsai, Cheng-Lin Hsieh, I-Hsiang Chen, Jian-Jiun Ding, and Sy-Yen Kuo

IEEE International Conference on Computer Vision (ICCV), 2021.

All Characteristics Preservation: Single Image Dehazing based on Hierarchical Detail Reconstruction Wavelet Decomposition Network

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Sep., 2021.

JSTASR: Joint Size and Transparency-Aware Snow Removal Algorithm Based on Modified Partial Convolution and Veiling Effect Removal

European Conference on Computer Vision (ECCV), Aug., 2020.

程式語言

機器學習框架