|
Hiya Roy Research Scientist, Rakuten Institute of Technology PhD, The University of Tokyo hiyaroy12 at gmail.com [GitHub] [Google Scholar] [CV] [ LinkedIn] |
About me
I am a research scientist at Rakuten Institute of Technology working on Computer Vision. I obtained my Ph.D. degree from the University of Tokyo working with Prof. Tatsuaki Hashimoto and Prof. Toshihiko Yamasaki on the topic of analyzing planetary images using deep learning. I received my Master's degree in Electrical Engineering and Information Systems from the University of Tokyo, Japan in 2017 and Bachelor's in Electrical Engineering from Jadavpur University, India in 2012.News
[March 2022] Our work on Enhancing Spatial Resolution of Remotely Sensed Imagery Using Deep Learning and/or Data Restoration has been published as a book chapter in Machine Learning for Planetary Science, 1st Edition, Elsevier Science and Technology Books.[April 2021] Our work on image inpainting using frequency-domain priors has been introduced in multiple tech blogs: SPIE news, Marktechpost.com, and TechXplore.
[April 2021] Our work on Image Inpainting using Frequency domain Priors has been accepted to SPIE Journal of Electronic Imaging.
[April 2021] I have graduated from the University of Tokyo in March and have joined Rakuten Institute of Technology as a Research Scientist in Tokyo!
[Nov 2020] Our work on "Toward Better Planetary Surface Exploration by Orbital Imagery Inpainting" has been accepted to IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.
Work Experience/Internship
I have had the opportunity to work with some wonderful mentors.-
NASA Jet Propulsion Laboratory, USA:
Position: Visiting Student Researcher [Oct 2019-Jan 2020]
At NASA JPL, I did a collaborative research under Dr. Masahiro Ono where I worked on Machine learning-based Analytics for Autonomous Rover Systems.
-
NEC Data Science Research Laboratories, Japan:
Position: Research Intern [Jan 2017-Mar 2017]
At NEC, I did an internship under Dr. Masato Tsukada and Mr. Kenta Senzaki on satellite image analysis using deep learning.
-
Tata Power Company Limited, India:
Position: Lead Engineer [July 2012-July 2015]
At Trombay Thermal Power Station, Tata Power Mumbai, I worked under the supervision of Mr. Sanjay Patki, leading the core Electrical Testing team whose job is to conditionally monitor and test all Electrical equipments(all over Tata Power), to carry out comissioning tests of new switchgear equipments and relay panels.
Research
My research interests are in computer vision, machine learning, deep learning, and image processing. Most of my research works lie in the intersection of machine learning and planetary sciences where I develop machine learning solutions to parse through the huge amount of data captured from different space exploration missions to Moon and Mars for facilitating scientific discovery, and on-board data interpretation for next generation planetary rovers.
Publications
Image Inpainting using Frequency domain Priors
Hiya Roy, Subhajit Chaudhury,
Toshihiko Yamasaki, Tatsuaki Hashimoto.
In SPIE Journal of Electronic Imaging, 30(2), 023016 (2021).
[Paper]
[Bibtex]
Toward Better Planetary Surface Exploration by Orbital Imagery Inpainting
Hiya Roy, Subhajit Chaudhury,
Toshihiko Yamasaki, Tatsuaki Hashimoto.
In IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (2020).
[Paper]
[Bibtex]
MAARS: Machine Learning-Based Analytics for Rover Systems
Masahiro Ono, Brandon Rothrock, Kyohei Otsu, Shoya Higa, Yumi Iwashita, Annie Didier, Tanvir Islam,
Christopher Laporte, Vivian Sun, Kathryn Stack, Jacek Sawoniewicz, Shreyansh Daftry, Virisha Timmaraju,
Sami Sahnoune, Chris A. Mattmann, Olivier Lamarre, Sourish Ghosh, Dicong Qiu, Shunichiro Nomura, Hiya Roy,
Hemanth Sarabu, Gabrielle Hedrick, Larkin Folsom, Sean Suehr, Hyoshin Park.
In IEEE Aerospace conference, 2020.
[Paper]
Lunar surface image restoration using U-net based deep neural networks
Hiya Roy, Subhajit Chaudhury,
Toshihiko Yamasaki,
Danielle DeLatte, Makiko Ohtake, Tatsuaki Hashimoto.
In Lunar and Planetary Science Conference, 2019.
[Paper]
[Bibtex]
Predicting Image Aesthetics using Objects in the Scene
Hiya Roy,Toshihiko Yamasaki,
Tatsuaki Hashimoto.
In International Joint Workshop on Multimedia Artworks Analysis and Attractiveness Computing
in Multimedia (MMArt and ACM) in conjuction with ICMR 2018.
[Paper]
[Bibtex]
Do hashtags help? – Image aesthetics prediction using only hashtags
Hiya Roy,Toshihiko Yamasaki, Tatsuaki Hashimoto.
In Women in Computer Vision Workshop (WICV), CVPR 2018.
[Poster]
Can fully convolutional networks perform well for general image restoration problems?
Subhajit Chaudhury, Hiya Roy
In International Conference on Machine Vision Applications (MVA), 2017.
[Paper]
[Poster]
[Bibtex]
Planetary image recognition using deep learning
Hiya Roy, Tatsuaki Hashimoto
In The 31st International Symposium on Space Technology and Science, ISTS 2017.
[Paper]
Doctoral Thesis
|
Deep Learning for Planetary Exploration: Improving image analysis capabilities under limited data resources Hiya Roy Advisors: Tatsuaki Hashimoto, Toshihiko Yamasaki; March 2021 [Thesis] [Slides] |
Book chapter
- Hiya Roy, Subhajit Chaudhury, Toshihiko Yamasaki, Tatsuaki Hashimoto, Chapter 10: Enhancing Spatial Resolution of Remotely Sensed Imagery Using Deep Learning and/or Data Restoration, "Machine Learning for Planetary Science", 1st Edition, Elsevier Science and Technology Books (to be published). [Book chapter]
Other Publications
- An Early History of the Philippine Space Development program; Quentin Verspieren, Giulio Coral, Budhaditya Pyne, Hiya Roy; In Acta Astronautica, Volume 151, October 2018, Pages 919- 927. [Paper] [Bibtex]
- The Necessity of National Space Agencies to Enhance Disaster Risk Preparedness in Southeast Asia: A Case Study of the Philippines; Quentin Verspieren, Giulio Coral, Budhaditya Pyne, Hiya Roy; In Proceedings of the United Nations International Conference on Space-based Technologies for Disaster Risk Reduction - "Enhancing Disaster Preparedness for Effective Emergency Response", Beijing, China, October 2018. [Paper]
- Space-based Disaster Risk Management by Location-based Emergency Shelter Awareness and Training (LESAT); Goutham Karthikeyan, Hiya Roy, Janet Han, M.C. MacDonald; 68th International Astronautical Congress (IAC) Adelaide, Sept 2017.
- Location based Emergency Shelter Awareness and Training (LESAT); Goutham Karthikeyan, Hiya Roy, Janet Han, M.C. MacDonald; Proceedings of the Global Public Policy Network (GPPN) Conference, Paris, February 2017. [Poster]
- Space Innovation Policy for Disaster Management capabilities-SIPDMC; Budhaditya Pyne, Giulio Coral, Goutham Karthikeyan, Hiya Roy, Lee Hee Woon, MacDonald M.C.; Conference on Space Based Technologies for Disaster Risk Reduction “Understanding Disaster Risk”, United Nations Office for Outer space affairs (UN-SPIDER), Beijing, China, September 2016.