Info: I am applying for a PhD program in the field of computer vision starting Fall 2019. I am also looking for an internship opportunity in the US in Summer 2019.

News


  • 11 / 2018:   One paper on representation learning accepted to AAAI 2019.
  • 10 / 2018:   We won the Third Place in IEEE Video and Image Processing (VIP) Cup.
  • 09 / 2018:   Work as a full time research assistant at Academia Sinica.
  • 07 / 2018:   One paper on semantic matching accepted to ACCV 2018.
  • 07 / 2017:   Start my internship at Dragon Cloud AI (Saratoga, CA).
  • 07 / 2017:   Work as a part time research assistant at Academia Sinica.
  • 05 / 2017:   One paper on malicious flow detection accepted to PIMRC 2017.
  • 11 / 2016:   One paper on ransomware detection accepted to IEICE 2016.

About me


I am a research assistant in the Computer Vision Lab at Academia Sinica under the supervision of Dr. Yen-Yu Lin. I am also fortunate to be co-advised by Prof. Jia-Bin Huang (Virginia Tech) and Prof. Ming-Hsuan Yang (UC Merced). In the meantime, I work closely with Prof. Yu-Chiang Frank Wang (NTUEE) and Prof. Winston H. Hsu (NTUCSIE).

My research interests lie in the intersection of Computer Vision and Deep Learning. I'm particularly interested in Unsupervised Domain Adaptation, as well as Representation Learning and Semantic Segmentation.

I received my Bachelor of Science degree from the Department of Electrical Engineering at National Taiwan University in 2018. During the past summer, I interned at Dragon Cloud AI (Saratoga, CA), where I worked as a software engineer on computer vision related tasks, face recognition in particular.

Publications


2018
Anonymous Title
Yun-Chun Chen, Yen-Yu Lin, Ming-Hsuan Yang, and Jia-Bin Huang
Submitted to CVPR 2019
Anonymous Title
Yun-Chun Chen*, Yu-Jhe Li*, Yen-Yu Lin, Xiaofei Du, and Yu-Chiang Frank Wang
Submitted to CVPR 2019
Show, Match and Segment: Weakly Supervised Learning for Joint Semantic Matching and Object Co-segmentation
Yun-Chun Chen, Kuang-Jui Hsu, Ming-Hsuan Yang, Jia-Bin Huang, and Yen-Yu Lin
Will be submitted to PAMI
Learning Resolution-Invariant Deep Representations for Person Re-Identification
Yun-Chun Chen*, Yu-Jhe Li*, Xiaofei Du, and Yu-Chiang Frank Wang
AAAI 2019 (acceptance rate: 16.2%)
[Paper] [Project Page] [Supplement] [Slides]
Saliency Aware: Weakly Supervised Object Localization
Yun-Chun Chen and Winston H. Hsu
ICASSP 2019 (Under review)
[Paper] [Project Page]
Accurate and Efficient Volumetric Lung Tumor Segmentation
Jhih-Yuan Lin*, Min-Sheng Wu*, Yu-Cheng Chang*, Yun-Chun Chen, Chao-Te Chou, Chun-Ting Wu, and Winston H. Hsu
ICASSP 2019 (Under review)
[Paper] [Project Page]
Learning Volumetric Segmentation for Lung Tumor
Jhih-Yuan Lin, Min-Sheng Wu, Yu-Cheng Chang, Yun-Chun Chen, Chao-Te Chou, Chun-Ting Wu, and Winston H. Hsu
IEEE Video and Image Processing (VIP) Cup
ICIP 2018
[Paper] [Project Page] [Slides] [Demo]
Deep Semantic Matching with Foreground Detection and Cycle Consistency
Yun-Chun Chen, Po-Hsiang Huang, Li-Yu Yu, Jia-Bin Huang, Ming-Hsuan Yang, and Yen-Yu Lin
ACCV 2018
[Paper] [Project Page] [Code] [Supplement]
2017
Deep Learning for Malicious Flow Detection
Yun-Chun Chen, Yu-Jhe Li, Aragorn Tseng, and Tsungnan Lin
PIMRC 2017
[Paper] [Project Page] [Slides]
2016
Deep Learning for Ransomware Detection
Aragorn Tseng, Yun-Chun Chen, YiHsiang Kao, and Tsungnan Lin
IEICE Technical Report 2016
[Paper]

Projects


Co-saliency Detection with Object Proposal and Background Prior
Co-saliency detection may serve as pusedo ground truth masks for image co-segmentation. In this project, we aim at developing an end-to-end trainable model for co-saliency detection.
Advisor: Dr. Yen-Yu Lin
Project Member: Yun-Chun Chen, Dr. Chung-Chi Tsai
Weakly Supervised 3D Object Localization
Localizing objects under weakly supervised settings is a challenging task, not to mention 3D scenario. In this project, we aim at leveraging class activation maps as an attention to precisely localize objects.
Advisor: Prof. Winston H. Hsu
Project Member: Yun-Chun Chen, Hung-Yueh Chiang, Yen-Shi Wang
Image-based Voxel Segmentation
Segmenting volumetric objects is often intractable due to heavy computations and the difficulty of obtaining manually annotated ground truth. In this project, we propose to project 3D objects to different angles then apply 2D segmentation techniques to perform voxel segmentation.
Advisor: Prof. Winston H. Hsu
Project Member: Yun-Chun Chen, Hung-Yueh Chiang, Yen-Shi Wang
Cross Domain Image-Based 3D Shape Retrieval by View Sequence Learning
We propose a cross-domain image-based 3D shape retrieval method that learns a joint embedding space for natural images and 3D shapes in an end-to-end manner.
Advisor: Prof. Winston H. Hsu
Project Member: Yun-Chun Chen, Tang Lee

Teaching


Working


Awards