Jan Zdenek

PhD Student at the Graduate School of Information Science and Technology, the University of Tokyo

Nakayama Laboratory, Graduate School of Information Science and Technology, the University of Tokyo

Biography

Jan is a third-year Ph.D. student in information science and technology particularly interested in artificial intelligence, computer vision, and machine learning. He is a member of the Machine Perception Group (Nakayama Laboratory) at the University of Tokyo, which he joined in 2015. Before commencing his graduate studies at the University of Tokyo, he earned his bachelor’s degree in computer and information science at the Czech Technical University in Prague, Czech Republic, where he was born and grew up.

His research focuses on computer vision, in particular on image generation and text in natural scene images. Recently, he has been interested in application of generative adversarial networks for handwritten text and text in natural scene images.

Interests

Computer Vision
Machine Learning
Artifical Intelligence
Image Generation

Education

PhD in Information Science and Technology, present

The University of Tokyo
Master of Information Science and Technology, 2017

The University of Tokyo
Bachelor of Computer and Information Science, 2014

Czech Technical University in Prague

Publications

International Conferences

Jan Zdenek, Hideki Nakayama (2021). JokerGAN: Memory-Efficient Model for Handwritten Text Generation with Text Line Awareness. ACMMM 2021.

PDF

Ryuichiro Hataya, Jan Zdenek, Kazuki Yoshizoe, Hideki Nakayama (2020). Faster AutoAugment: Learning Augmentation Strategies using Backpropagation. ECCV 2020.

PDF DOI

Jan Zdenek, Hideki Nakayama (2020). Erasing Scene Text With Weak Supervision. WACV 2020.

PDF DOI

Jan Zdenek, Hideki Nakayama (2017). Bag of Local Convolutional Triplets for Script Identification in Scene Text. ICDAR 2017.

PDF DOI

Other Publications

Preprints and Domestic Conferences

Ryuichiro Hataya, Jan Zdenek, Kazuki Yoshizoe, Hideki Nakayama (2021). Meta Approach to Data Augmentation Optimization. arXiv preprint.

PDF

Jan Zdenek, Hideki Nakayama (2021). Efficient Model for Handwritten Text Generation with Text Line Awareness. MIRU 2021.

Ryuichiro Hataya, Jan Zdenek, Kazuki Yoshizoe, Hideki Nakayama (2020). MADAO: データ拡張最適化のためのメタ的アプローチ. MIRU 2020.

Ryuichiro Hataya, Jan Zdenek, Kazuki Yoshizoe, Hideki Nakayama (2019). Faster AutoAugment: Learning Augmentation Strategies using Backpropagation. IBIS 2019.

Jan Zdenek, Hideki Nakayama (2019). Erasing Scene Text Using a General Inpainting Network. MIRU 2019.

See all publications

Experience

Research Assistant

The University of Tokyo, Graduate School of Information Science and Technology

Oct 2020 – Present Tokyo, Japan

Participating in Beyond AI, a joint research program of The University of Tokyo and Softbank. Performing research on image generation of handwritten text and text in natural scene images.

Machine Learning Engineer Intern

RESTAR

Feb 2019 – May 2019 Tokyo, Japan

Participated in development of an OCR system for automatic conversion of real estate data documents from PDF to machine-readable form for an automatic generation of database of real estate data.

Research Intern

Center for Machine Perception

Jul 2014 – Dec 2014 Prague, Czech Republic

Participated in research on scene text detection and recognition as a research intern.

Competitions

3rd place in the ICDAR 2017 Robust Reading Competition on Multilingual Scene Text Detection and Script Identification in the task of script identification

Awards

MIRU 2020 Student Paper Award (Hataya R., Zdenek J., Yoshizoe K., Nakayama H.)
The Japanese Society for Artificial Intelligence (JSAI) Annual Conference Student Incentive Award 2017
Japanese Government (MEXT) Scholarship for Research Students
Dean’s Award for an Outstanding Bachelor Thesis, Czech Technical University, 2014

Other Activities

Teaching

I have worked as a teaching assistant for the following courses at the University of Tokyo, Graduate School of Information Science and Technology:

Data Science

Peer review

I have reviewed papers for conferences such as ICLR, AAAI, SIGGRAPH, ICMR, and journals such as TIP.

Contact

4F, I-REF, Graduate School of Information Science and Technology, the University of Tokyo, 1-1-1 Yayoi, Bunkyo, Tokyo, 113-0032, Japan
Twitter
Github
LinkedIn
Google Scholar
Research Gate