Jonas Kulhanek
@jonaskulhanek
PhD student at CTU in Prague, supervised by @SattlerTorsten and Marc Pollefeys, focusing on 3D scene reconstruction.
EdgeGaussians - 3D Edge Mapping via Gaussian Splatting Kunal Chelani, @_abenbihi, @SattlerTorsten, @FredrikKahl tl;dr: explicit representation of the 3D edge points; 3D edge point->center of a 3D Gaussian; edge direction->principal axis of the Gaussian arxiv.org/pdf/2409.12886
CL-Splats: Continual Learning of Gaussian Splatting with Local Optimization @jan_on_x, @jonaskulhanek, @prime_cai, @haofeixu, @mapo1, @GordonWetzstein, Leonidas Guibas, @songyoupeng tl;dr: DINOv2-based 2D binary change masks->voting+sampling->3D masks->3DGS-based local…
LODGE: Level-of-Detail Large-Scale Gaussian Splatting with Efficient Rendering @jonaskulhanek, Marie-Julie Rakotosaona, Fabian Manhardt, Christina Tsalicoglou, @Mi_Niemeyer, @SattlerTorsten, @songyoupeng, @fedassa tl;dr: camera distance->active Gaussians; pre-compute active…
🚀 Just wrapped up my internship at Google with a paper submission! LODGE is a new large-scale level-of-detail 3DGS method, enabling efficient rendering even on mobile devices. Thanks to the amazing team at Google Zurich for making this possible! 📄lodge-gs.github.io
We are releasing uCO3D! Built to supercharge 3D GenAI and digital-twin models, this evolution of CO3D features more and higher-quality object videos from 1k categories, 3D Gaussian Splats, and streamlined OSS tools. 💻Data&code: github.com/facebookresear… 📄Paper:…
I’ll present our #NeurIPS2024 Spotlight Poster on Friday, 13 Dec 4:30-7:30pm PST in East Exhibit Hall A-C, booth #1505. Please step by to talk large-scale dynamic 3D Gaussians! :)
Dynamic 3D Gaussian Fields for Urban Areas tobiasfshr.github.io/pub/4dgf/ We optimize a single dynamic scene representation from a set of heterogeneous input sequences of a common geographic area, rendering at interactive speeds while modeling huge appearance and geometry changes.
I will be presenting the WildGaussians paper this afternoon at @NeurIPSConf . To learn more about our paper, visit me at poster stand #1307 in East Exhibit Hall 16:30-19:30
WildGaussians: 3D Gaussian Splatting in the Wild arxiv.org/abs/2407.08447 Project: wild-gaussians.github.io Code: github.com/jkulhanek/wild… Method ⬇️ 1 I 2
Heads up! Mosaic dropped a pretty wild dataset of 1.26 million 360° images of Prague 🤯 If you're a researcher, creator or developer into 3D/AI/Geo, I think you're gonna wanna play with this Here's the scoop on this 15 TERAPIXEL dataset & the crazy things you can do with it 🧵
Our WildGaussians paper got accepted to @NeurIPSConf 2024 🎉!
WildGaussians: 3D Gaussian Splatting in the Wild arxiv.org/abs/2407.08447 Project: wild-gaussians.github.io Code: github.com/jkulhanek/wild… Method ⬇️ 1 I 2
Meta 3D AssetGen has been accepted to #neurips2024!! assetgen.github.io Big congrats to the team @ovrdr, @davnov134, Andrea Vedaldi, @filippos_kok, @t_monnier, @mahendrakariya, @NataliaNeverova and everyone involved! :)
Tired of 3D asset generation approaches with baked in lighting effects? Our latest work, Meta 3D AssetGen, can generate high quality meshes with PBR materials given text prompts in seconds! assetgen.github.io The work was done with the amazing GenAI 3D team @AIatMeta
Happy to share that our paper was accepted as spotlight @NeurIPSConf 2024! See you in Vancouver 🇨🇦
Dynamic 3D Gaussian Fields for Urban Areas tobiasfshr.github.io/pub/4dgf/ We optimize a single dynamic scene representation from a set of heterogeneous input sequences of a common geographic area, rendering at interactive speeds while modeling huge appearance and geometry changes.
Thanks a lot for sharing!
Check out this project! It has a lot of bells and whistles and is highly interactive. Highly recommended!
🚀 Ready to take 3D reconstruction to the next level? Whether you're working on NeRF or 3DGS, our new method, GLOMAP, is here to impress! 🌟 It's faster and more accurate than COLMAP on several datasets. 🌐 Website: lpanaf.github.io/eccv24_glomap/ @mapo1, @LinfeiPan, J. Schönberger
WildGaussians 3D Gaussian Splatting in the Wild While the field of 3D scene reconstruction is dominated by NeRFs due to their photorealistic quality, 3D Gaussian Splatting (3DGS) has recently emerged, offering similar quality with real-time rendering speeds. However, both…
WildGaussians: 3D Gaussian Splatting in the Wild arxiv.org/abs/2407.08447 Project: wild-gaussians.github.io Code: github.com/jkulhanek/wild… Method ⬇️ 1 I 2
I wrote this useful Python script to prepare your paper (arXiv) submissions. It crops and resizes all images to fit a specified DPI, removes all comments, removes unused files, and generates a zip file that can be directly uploaded. gist.github.com/jkulhanek/b48f…
Excited to announce 🚀gsplat v1.0🚀: a ⏩efficient⏩ CUDA backend for 3D Gaussian Splatting! docs.gsplat.studio A drop-in replacement of the official impl. with: - Up to 2x faster training; - Up to 4x less GPU memory; - Render millions of GSs in real-time; - And more;
Dynamic 3D Gaussian Fields for Urban Areas @TobiasFischer11, @jonaskulhanek, Samuel Rota Bulò, Lorenzo Porzi, @mapo1, Peter Kontschieder tl;dr: 3D Gaussian primitives->geometry scaffold; neural fields->appearance; scene graph->scene dynamics arxiv.org/pdf/2406.03175