Tseng-Hung Chen
Vision Science Lab
Department of Electrical Engineering
National Tsing Hua University, Hsinchu, Taiwan
paul516800 [at] gmail.com
About Me
Tseng-Hung Chen received his M.S. degree in the Department of Electrical Engineering at National Tsing Hua University in 2017, supervised by Prof. Min Sun.
My research interests lie in the intersection of Computer Vision, Natural Language Processing and Machine Learning. My research projects focus on natural language generation, visual semantic understanding and unsupervised domain adaptation in text generation.
Here is my CV.
News
- 12/2017: Start the mandatory military service in Taiwan, which will end in Nov. 2018.
- 07/2017: One paper accepted to ICCV 2017.
- 06/2017: Start my summer intern at Appier Inc.
- 11/2016: One paper accepted to AAAI 2017.
- 09/2016: One paper accepted to ACCV 2016 Workshop.
- 07/2016: One paper accepted to ECCV 2016.
Publications
Show, Adapt and Tell: Adversarial Training of Cross-domain Image Captioner
IEEE International Conference on Computer Vision (ICCV) 2017
Conference
Leveraging Video Descriptions to Learn Video Question Answering
Association for the Advancement of Artificial Intelligence (AAAI) 2017
Conference
Video Captioning via Sentence Augmentation and Spatio-Temporal Attention
Asian Conference on Computer Vision (ACCV) 2016 - Workshop on Assistive Vision
Workshop
Ranking Highlights in Personal Videos by Analyzing Edited Videos
IEEE Transactions on Image Processing 2016
Journal
Title Generation for User Generated Videos
European Conference on Computer Vision (ECCV) 2016
Conference
Projects
Education
M.S. in EE, National Tsing Hua University, Hsinchu, Taiwan
Sep. 2015 - Jul. 2017
B.S. in EECS, National Tsing Hua University, Hsinchu, Taiwan
Sep. 2011 - Jun. 2015
Work Experience
M.S. Student, Research Assistant
Vision Science Lab, National Tsing Hua University, Hsinchu, Taiwan
Image/Video Captioning,
Video Question Answering,
Video Summarization,
Domain Adaptation
Video Question Answering,
Video Summarization,
Domain Adaptation
May 2015 - June 2017
Intern
Appier, Taipei, Taiwan
Fashion Landmark Detection,
Fashion Item Retrieval
Fashion Item Retrieval
Summer 2017
Paper Notes
Feb 25, 2018
In computer vision, CNNs trained on ImageNet have been the must-have initialization for a set of downstream tasks including object detection, semantic segmentation and pose estimation. NLP shares the similar concept about transferring from large-scale data. Distributional representations such as word2vec (Mikolov et al., 2013) and GloVe (Pennington et al., 2014) have been common initializations for word embeddings in deep networks...(Read more)
