Meta tags:
description=LODGE paper. Official web with qualitative comparisons, links to additional materials.;
keywords=lodge,3dgs,nerf,official,code;
author=Jonas Kulhanek;

Headings (most frequently used words):

rendering, lodge, level, of, detail, large, scale, gaussian, splatting, with, efficient, abstract, lod, representation, for, faster, opacity, interpolation, to, remove, cross, chunk, transitions, citation,

Text of the page (most frequently used words):
fps (12), the (10), #rendering (8), and (8), with (7), google (7), ours (6), lod (5), chunks (5), gaussians (5), lodge (4), large (4), scale (4), efficient (4), opacity (4), level (3), detail (3), splatting (3), blending (3), visible (3), chunk (3), representation (3), per (3), based (3), jonas (2), kulhanek (2), marie (2), julie (2), rakotosaona (2), fabian (2), manhardt (2), christina (2), tsalicoglou (2), michael (2), niemeyer (2), torsten (2), sattler (2), songyou (2), peng (2), federico (2), tombari (2), 2025 (2), please (2), citation (2), 258 (2), number (2), pixel (2), 3dgs (2), active (2), cluster (2), gaussian (2), inproceedings, kulhanek2025lodge, title, aussian, author, year, booktitle, proceedings, 39th, international, conference, neural, information, processing, systems, use, following, 317, ensure, temporal, consistency, when, transitioning, between, propose, focus, black, car, notice, how, removes, sharp, changes, interpolation, remove, cross, transitions, full, show, lighter, means, more, adding, reduce, significantly, improving, speed, little, loss, quality, for, faster, 253, h3dgs, 280, octree, 119, 219, zipnerf, 257, scene, represented, multiple, lods, are, selected, during, training, camera, distance, cameras, into, pre, compute, render, two, nearest, right, left, abstract, enables, even, mobile, devices, video, paper, deepmind, ctu, prague, neurips, spotlight,

Text of the page (random words):
lodge level of detail large scale gaussian splatting with efficient rendering lodge level of detail large scale gaussian splatting with efficient rendering neurips 2025 spotlight jonas kulhanek google marie julie rakotosaona google fabian manhardt google christina tsalicoglou google michael niemeyer google torsten sattler ctu in prague songyou peng google deepmind federico tombari google paper video lodge enables efficient rendering of large scale 3dgs even on mobile devices abstract left lod the scene is represented with multiple lods active gaussians are selected during training based on camera distance right cluster based rendering we cluster cameras into chunks pre compute active gaussians per chunk and render the two nearest chunks with opacity blending ours fps 257 zipnerf fps 0 09 ours fps 219 octree gs fps 119 ours fps 280 h3dgs fps 33 ours fps 253 3dgs fps 99 lod representation for faster rendering we show the number of visible gaussians per pixel lighter means more gaussians by adding the lod representation with chunk based rendering we reduce the number of visible gaussians per pixel significantly improving the rendering speed with little loss in quality ours fps 258 full representation fps 66 opacity interpolation to remove cross chunk transitions to ensure temporal consistency when transitioning between chunks we propose opacity blending please focus on the black car notice how opacity blending removes the sharp changes visible in lod chunks ours fps 258 lod chunks fps 317 citation please use the following citation inproceedings kulhanek2025lodge title lodge level of detail large scale g aussian splatting with efficient rendering author jonas kulhanek and marie julie rakotosaona and fabian manhardt and christina tsalicoglou and michael niemeyer and torsten sattler and songyou peng and federico tombari year 2025 booktitle proceedings of the 39th international conference on neural information processing systems