PhD Candidate @ Trinity College Dublin, Ireland
Research Focus: Computer Vision & Graphics

I am currently pursuing a PhD in Computer Science at Trinity College Dublin, specializing in Stereo Matching and Depth Estimation for advanced robotics, augmented reality (AR), and virtual reality (VR) applications. My research addresses critical challenges in 3D reconstruction and autonomous driving, focusing on improving the real-time performance of deep learning models in these domains. Under the supervision of Subrahmanyam Murala (TCD), Peter Corcoran (UoG), and Carol O’Sullivan (TCD), my goal is to develop lightweight deep learning models that enhance the accuracy and efficiency of depth estimation systems.

Visit my Google Scholar | LinkedIn

Education

• Doctor of Philosophy (PhD), Computer Science
  Trinity College Dublin (Sep 2023 - Aug 2027)
  Research Areas: Computer Vision, Graphics, Stereo Matching, Depth Estimation

• Master of Technology (M.Tech), Signal Processing and Communication
  Indian Institute of Technology, Ropar (Jan 2021 - May 2023)
  Skills: Deep Learning, Computer Vision, Programming, Pattern Recognition

• Bachelor of Technology (B.Tech), Electronics and Communications Engineering
  Rajasthan Technical University, Kota (2014 - 2018)

Work Experience

• Project Associate
  Indian Institute of Technology, Ropar
  • Developed a novel architecture, Spectroformer, which applies a Multi-Domain Query Cascaded Transformer Network for underwater image enhancement.
  • Conducted research on Underwater Image Enhancement using using Signal Processing and Transformer based Deep learning techniques, significantly improving image quality in challenging underwater conditions.

Research Projects

1. Phaseformer: Phase-based Attention Mechanism for Underwater Image Restoration and Beyond
   GitHub Repository

2. Spectroformer: Multi-Domain Query Cascaded Transformer Network for Underwater Image Enhancement
   GitHub Repository

3. Underwater Image Enhancement with Phase Transfer and Attention (UIEPTA)
   GitHub Repository

AI-Based Ball and Wicket Detection with Pitch Mapping and Impact Analysis in Cricket”

Project Summary: This project focuses on developing an computer vision model to detect the cricket ball and wicket in real time, map the pitch, and accurately identify the ball’s impact point after being thrown. Using advanced computer vision techniques, the model tracks the ball’s trajectory, determines the exact landing spot on the pitch, and assesses interactions with the wicket. This information provides insights into ball behavior, such as bounce and spin, which are critical for performance analysis and strategic planning. The project aims to enhance decision-making in cricket through precise ball tracking and pitch analytics.

Skills

• Signal Processing
• Image Processing
• Deep Learning
• Computer Vision
• GitHub
• Programming: Python, MATLAB, C/C++
• Pattern Recognition

Publications

1. “Phase-based Attention Mechanism for Underwater Image Restoration and Beyond” IEEE/CVF Winter Conference on Applications of Computer Vision (WACV 2025) (Accepted h-index:109)

2. ACFTNet: Attentive Color Fusion Transformer Network (ACFTNet) for Underwater Image Enhancement” International Conference on Document Analysis and Recognition (ICPR 2025) (Accepted h-index:52)

3. “Spectroformer: Multi-domain query cascaded transformer network for underwater image enhancement.”
   In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision, 2024, pp. 1454-1463.
   This work introduces Spectroformer, a transformer-based network for multi-domain underwater image enhancement using cascaded query transformers.
   GitHub Repository

4. “Underwater image enhancement with phase transfer and attention.”
   In 2023 International Joint Conference on Neural Networks (IJCNN), IEEE, 2023, pp. 1-8.
   The paper presents a novel method for underwater image enhancement combining phase transfer with attention mechanisms to address visibility and color distortions.
   GitHub Repository

5. “NTIRE 2024 image shadow removal challenge report.”
   In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024, pp. 6547-6570.
   A comprehensive challenge report detailing state-of-the-art techniques and solutions for image shadow removal.