Brent Zoomers

I am a PhD student at Hasselt University under the supervision of Prof. Dr. Nick Michiels and currently doing an internship at the TU Braunschweig under supervision of Prof. Dr.-Ing. Marcus Magnor. My research focuses on Computer Vision & Graphics, specifically Novel View Synthesis.

I received my Bachelor's and Master's degree from Hasselt University.

Email / Scholar / Twitter

Research

Representative papers are highlighted.

	NVGS: Neural Visibility for Occlusion Culling in 3D Gaussian Splatting Brent Zoomers, Florian Hahlbohm, Joni Vanherck, Lode Jorissen, Marcus Magnor, Nick Michiels project page / arXiv / code (coming soon) We propose a novel pipeline for rendering large composed scene consisting of 3DGS assets. By baking visibility data into a lightweight MLP, and using our novel instanced rasterizer tailored for 3DGS, we allow faster rendering and drastically lower VRAM usage.
	CRISTAL: Real-time Camera Registration in Static LiDAR Scans using Neural Rendering Joni Vanherck, Steven Moonen, Brent Zoomers, Kobe Werner, Jeroen Put, Lode Jorissen, Nick Michiels arXiv By utilizing neural point cloud rendering on LiDAR point clouds, we enable both real-time camera registration, and a way to create a map that can be used for localization directly from LiDAR.
	Real-time Neural Rendering of LiDAR Point Clouds Joni Vanherck, Brent Zoomers, Tom Mertens, Lode Jorissen, Nick Michiels (=equal contribution) Eurographics, 2025* arXiv / code Real-time rendering of point clouds by using depth-filter that preserves edges, after which we utilize a pre-trained UNet for hole filling. Due to the edge preserving nature of the depth filter, his image allows feature matching between these renders and actual images.
	PRoGS: Progressive Rendering of Gaussian Splats Brent Zoomers, Maarten Wijnants, Ivan Molenaers, Joni Vanherck, Jeroen Put, Lode Jorissen, Nick Michiels WACV, 2025 arXiv By considering contribution of Gaussians instead, we can better prioritize which Gaussians to render first to get a better representation of a scene faster.
	Cut-and-splat: Leveraging gaussian splatting for synthetic data generation Bram Vanherle, Brent Zoomers, Jeroen Put, Frank Van Reeth, Nick Michiels ROBOVIS, 2025 Best Paper Award arXiv / code Synthetic data generation using 3D Gaussian Splatting by pasting 3DGS objects into custom environments.
	Projector-camera calibration with non-overlapping fields of view using a planar mirror Joni Vanherck, Lode Jorissen, Brent Zoomers, Nick Michiels (=equal contribution) Virtual Reality, 2025* Springer Calibration of projecter and camera using planar mirror
	Large-Area Spatially Aligned Anchors Joni Vanherck, Lode Jorissen, Brent Zoomers, Nick Michiels XR Salento, 2024 Springer Acurate tracking in large-environments using spatial anchors
	Brains Over Brawn: Small AI Labs in the Age of Datacenter-Scale Compute Jeroen Put, Nick Michiels, Bram Vanherle, Brent Zoomers DeLTA, 2024 Springer Position paper on the future of small AI labs
	Incorporating view-dependent effects into point clouds (Master's Thesis) Brent Zoomers 2023 Link OpenGL implementation of an Unstructure lumigraph-inspired approach to render view-dependent effects in real-time for point clouds.
	Adaptive streaming and rendering of static light fields in the web browser Hendrik Lievens, Maarten Wijnants, Brent Zoomers, Jeroen Put, Nick Michiels, Peter Quax, Wim Lamotte IC3D, 2021 IEEEXplore Static light-field rendering in browser

This website is based on the source code of Jon Barron.