About Me

Hi, I’m a computer science researcher working in computer vision and 3D reconstruction, with a strong interest in turning research methods into practical, easy-to-use systems. Most recently, I worked on R3CAP, an open-source platform for real-time scene capture, editing, and collaboration across web and VR. The system combines SLAM, digital twins, and multi-user 3D editing, and was showcased at SIGGRAPH 2024 Labs and CHI 2025 Interactivity.

Earlier, my PhD research focused on multi-view stereo, where I explored numerical and metaheuristic optimization techniques to tackle ill-posed problems in multi-view geometry. This work shaped my core interests in 3D vision and continues to inform how I approach newer representations and pipelines.

Over time, I have increasingly enjoyed working on systems that move beyond isolated algorithms and toward usable, end-to-end tools. My primary motivation is to develop technically sound 3D vision methods that can be deployed, evaluated, and extended in real-world settings.

I’m particularly interested in:

• 3D reconstruction (Photogrammetry, SLAM, Gaussian Splatting)
• Computer vision methods for depth, geometry, and scene understanding
• Interactive spatial tools for web and VR (Babylon.js, WebXR)
• Translational research that moves ideas from papers into working software

---

Professional Experience

Research Fellow

Centre for Immersification, Singapore Institute of Technology
May 2023 - July 2025 (2 years, 3 months)

• Co-developed an open-source platform for real-time, progressive 3D reconstruction and collaborative scene editing in web and VR.
• Evaluated and integrated MVS, SLAM, and Gaussian Splatting techniques to enhance system capabilities.
• Delivered 25+ technical presentations and live demonstrations to industry stakeholders, leading to ongoing research collaborations.
• Co-authored publications at peer-reviewed research venues.

---

Projects

R3CAP

SLAM, Gaussian Splatting | Python, NeRF Studio, Babylon.js, WebXR

• Previously known as demoConstruct, R3CAP is an open-source platform that enables accessible and collaborative scene authoring, powered by progressive 3D reconstruction.
• Showcased at ACM SIGGRAPH 2024 and ACM CHI 2025.

Multi-View Stereo Reconstruction

3D Reconstruction, Computer Vision, Graph Cuts, Metaheuristic Optimization | Python, COLMAP, MATLAB

• A depth map-based multi-view stereo method using metaheuristics and graph cuts for depth estimation and refinement.
• Published in IEEE Signal Processing Letters 2022.

---

Publications

Collaborative Scene Authoring with Near Real-Time 3D Reconstruction

Leon Foo, Nirmal S. Nair, Liuziyi Liu, Jeannie Su Ann Lee, Songjia Shen, Indriyati Atmosukarto, Alvin Chan, Jing Shi, Yong Joo Loh, Yih Yng Ng, Michael Chia, Chek Tien Tan
ACM CHI 2025 Interactivity (Yokohama, Japan)

demoConstruct: Democratizing Scene Construction for Digital Twins through Progressive Reconstruction

Leon Foo, Chek Tien Tan, Liuziyi Liu, Nirmal S. Nair, Songjia Shen, Jeannie Lee
ACM SIGGRAPH 2024 Labs (Denver, USA)

Multi-View Stereo Using Graph Cuts-Based Depth Refinement

Nirmal S. Nair, Madhu S. Nair
IEEE Signal Processing Letters, 2022

Multi-view Stereo using Cross-view Depth Map Completion and Row-column Depth Refinement

Nirmal S. Nair, Madhu S. Nair
International Conference on Digital Image Processing, ICDIP 2021 (Singapore)

Scalable Multi-view Stereo using CMA-ES and Distance Transform-based Depth Map Refinement

Nirmal S. Nair, Madhu S. Nair
International Conference on Machine Vision, ICMV 2020 (Rome, Italy)

On Evolutionary Computation Techniques for Multi-view Triangulation

Nirmal S. Nair, Madhu S. Nair
Machine Vision and Applications, Springer, 2020