Summary
Kunal Chelani is a perception researcher with 11 years of experience, currently focused on 3D computer vision and visual localization after completing a PhD at Chalmers and joining Ericsson Research. His work blends rigorous geometry and machine learning to build accurate, robust 3D reconstruction and privacy-preserving localization methods, and he has exposed practical vulnerabilities in existing privacy representations while proposing concrete mitigations. He introduced a novel privacy attack vector based on querying image sequences and developed an edge-specialized 3D Gaussian splatting pipeline that produces parametric lines and Bézier curves from multi-view edge images. Kunal has industry experience from Meta Reality Labs contributing to AR scene generation and furniture placement, and earlier contributed to low-cost LiDAR–camera calibration for autonomous vehicles. Based in Gothenburg, he combines academic depth with product-oriented research, often pursuing efficient, edge-focused representations that make visual localization more practical for XR applications.
10 years of coding experience
9 years of employment as a software developer
Master of Technology - MTech, Computer Science, 9.0/10, Master of Technology - MTech, Computer Science, 9.0/10 at Indian Institute of Science (IISc)
BITS Pilani, Birla Institute of Technology and Science
High School, Science, 93.4 %, High School, Science, 93.4 % at Seedling Modern High School
Doctor of Philosophy - PhD, 3D Computer Vision, Doctor of Philosophy - PhD, 3D Computer Vision at Chalmers University of Technology