Rejoy Chakraborty

Rejoy Chakraborty

Hi all, currently, I am a PhD Scholar at the Computer Vision and Pattern Recognition Unit, Indian Statistical Institute, Kolkata, under the supervision of Prof. Umapada Pal.

I have received my M. Tech. in Computer Science and Engineering with the Institute Silver Medal from the Indian Institute of Technology Ropar (IIT Ropar). During M.Tech. thesis, I have worked under the supervision of Dr. Puneet Goyal. Before that, I also completed my M.Sc. in Computer Science from West Bengal State University, where I worked on my Master's thesis under Prof. Kaushik Roy. I have earned my B.Sc. (Hons) in Computer Science with a Gold Medal from Ramakrishna Mission Vivekananda Centenary College.

Research

Research Interest: Computer Vision, Generative AI, Deep Learning, Image Processing

> Selected Publications _
	(NEW!) DRG-Font: Dynamic Reference-Guided Few-shot Font Generation via Contrastive Style-Content Disentanglement Rejoy Chakraborty, Prasun Roy, Saumik Bhattacharya, Umapada Pal arXiv, 2026 Project Page / Code / arXiv / BibTex We introduce a few-shot font generation strategy that learns complex glyph attributes by decomposing style and content embedding spaces.
	Source Camera Model Identification via Federated Learning using Laplacian-based Patches Rejoy Chakraborty, Puneet Goyal IEEE Transactions on Artificial Intelligence, 2025 Project Page / Code / arXiv / BibTex We introduced first federated learning based framework for source camera identification where a dual branch architecture has been used for classifiation of patches selected through laplacian filter.
	Camera Model Identification with SPAIR-Swin and Entropy based Non-Homogeneous Patches Protyay Dey, Rejoy Chakraborty, Abhilasha S. Jadav, Kapil Rana, Gaurav Sharma, Puneet Goyal arXiv, 2025 Project Page / Code / arXiv / BibTex We introduced SPAIR-Swin method incorporating Swin Transformer and SPAIR module for source camera classification of non-homogenous patches extracted through Shanon entropy.
	MSBNet: Handwritten Bangla Character Recognition Using Lightweight Multi-scale CNN Architecture Rejoy Chakraborty, Chayan Halder, Kaushik Roy, Shivam Gupta, Shashi Shekhar Jha CODS COMAD (Dec'24), 2024 Project Page / Code / arXiv / BibTex We introduced MSBNet, a multi-scale lightweight Bangla character recognition architecture.