RayPose-MVS
A feed-forward model that predicts dense 3D geometry and camera pose in a single pass, without bundle adjustment or RGB inputs, trained on ~1% of the data other methods use.
AI Research Engineer
Shubham Kalantri
Computer vision and 3D reconstruction
About
I am an AI research engineer working in computer vision, mostly on 3D reconstruction, medical imaging and generative models. I am currently at NeuroDiscovery AI, where I am building an MRI analysis assistant along with the GPU infrastructure that serves it.
I did my MSc in Computer Graphics, Vision and Imaging at University College London. For my thesis, I trained a model to predict dense 3D geometry and camera pose in a single forward pass, using roughly 1% of the training data that comparable methods need.
Selected work
Discrete Geometry
Registration, halfedge structures, discrete curvature, the spectrum of the Laplace–Beltrami operator, and denoising by curvature flow, with every operator and solver implemented in NumPy and SciPy.
Planar Geometry
Homography estimation, RANSAC, corner detection, panorama stitching, metric rectification, plane pose, template tracking and Zhang's calibration, implemented without any OpenCV geometry calls.
All projects
The rest of my projects in graphics, geometry and vision, most of them implemented from the maths up.