Shubhajit Basak

Shubhajit Basak is a 3rd year PhD student at the School of Computing, National University of Ireland, Galway. He completed his  MSc in Computer Science (Data Analytics) from School of Computing, National University of Ireland, Galway. He has more than 6 years of industrial experience as a software developer and data engineer.

Research Activities

Working on 3D computer vision and deep learning problems
Current project deals with the 3D human face reconstruction from single image and application of generative model like GAN to generate random 3D human faces
Previously worked on Virtual Human and 3D synthetic facial data. Particularly worked on Head Pose Estimation and Facial Monocular Depth Estimation with the help of 3D synthetic data using semi-supervised domain adaptation and supervised methods
Also have experience in NLP tasks like word and sentence embedding from knowledge graph libraries like Word Net and DBPedia

Journal Publications

Basak, S., Corcoran, P., Khan, F., Mcdonnell, R., & Schukat, M. (2021). Learning 3D Head Pose From Synthetic Data: A Semi-Supervised Approach. IEEE Access, 9, 37557-37573.

Basak, S., Javidnia, H., Khan, F., McDonnell, R., & Schukat, M. (2020, June). Methodology for building synthetic datasets with virtual humans. In 2020 31st Irish Signals and Systems Conference (ISSC) (pp. 1-6). IEEE.

Basak, S., Khan, F., McDonnell, R., & Schukat, M. (2021, January). Learning Accurate Head Pose for Consumer Technology From 3D Synthetic Data. In 2021 IEEE International Conference on Consumer Electronics (ICCE) (pp. 1-6). IEEE.

Khan, F., Basak, S., & Corcoran, P. (2021, January). Accurate 2D Facial Depth Models Derived from a 3D Synthetic Dataset. In 2021 IEEE International Conference on Consumer Electronics (ICCE) (pp. 1-6). IEEE.

Khan, F., Hussain, S., Basak, S., Lemley, J., & Corcoran, P. (2021). An efficient encoder–decoder model for portrait depth estimation from single images trained on pixel-accurate synthetic data. Neural Networks, 142, 479-491.

Khan, F., Basak, S., Javidnia, H., Schukat, M., & Corcoran, P. (2020, June). High-Accuracy Facial Depth Models derived from 3D Synthetic Data. In 2020 31st Irish Signals and Systems Conference (ISSC) (pp. 1-5). IEEE.

Conference Participation & Presentations

Presented the work on Virtual Humans in Irish Signal and System (ISSC 2020) conference

Given a talk on working with 3D Data with Blender at the D-Real Summer School 2021