Meta tags:

Headings (most frequently used words):

kilonerf, abstract, video, results, citation, speeding, up, neural, radiance, fields, with, thousands, of, tiny, mlps,

Text of the page (most frequently used words):
and (6), with (6), mlps (6), the (5), #neural (4), #radiance (4), thousands (4), tiny (4), rendering (4), nerf (4), kilonerf (3), speeding (3), fields (3), iccv (3), our (3), real (3), time (3), video (3), scene (3), mlp (3), this (3), christian (2), reiser (2), songyou (2), peng (2), yiyi (2), liao (2), andreas (2), geiger (2), 2021 (2), code (2), quality (2), times (2), paper (2), that (2), can (2), further (2), without (2), for (2), supplementary (2), inproceedings, reiser2021iccv, author, title, booktitle, international, conference, computer, vision, year, you, want, cite, work, please, use, citation, following, videos, were, rendered, interactive, viewer, provided, release, results, representing, many, independent, arranged, voxel, grid, synthesizes, novel, views, unprecedented, fitting, field, rgb, images, however, requires, querying, deep, multi, layer, perceptron, millions, leading, slow, even, modern, gpus, demonstrate, possible, utilizing, instead, one, single, large, setting, each, individual, only, needs, represent, parts, thus, smaller, faster, evaluate, used, combining, divide, conquer, strategy, optimizations, accelerated, three, orders, magnitude, compared, original, model, incurring, high, storage, costs, using, teacher, student, distillation, training, show, speed, achieved, sacrificing, visual, abstract, blog, max, planck, institute, intelligent, systems, eth, zurich, university, tübingen,

Text of the page (random words):
kilonerf speeding up neural radiance fields with thousands of tiny mlps kilonerf speeding up neural radiance fields with thousands of tiny mlps christian reiser songyou peng yiyi liao andreas geiger max planck institute for intelligent systems eth zurich and university of tübingen iccv 2021 paper supplementary code blog iccv video supplementary video abstract nerf synthesizes novel views of a scene with unprecedented quality by fitting a neural radiance field to rgb images however nerf requires querying a deep multi layer perceptron mlp millions of times leading to slow rendering times even on modern gpus in this paper we demonstrate that real time rendering is possible by utilizing thousands of tiny mlps instead of one single large mlp in our setting each individual mlp only needs to represent parts of the scene thus smaller and faster to evaluate mlps can be used by combining this divide and conquer strategy with further optimizations rendering is accelerated by three orders of magnitude compared to the original nerf model without incurring high storage costs further using teacher student distillation for training we show that this speed up can be achieved without sacrificing visual quality tl dr real time rendering of nerf by representing the scene as many independent mlps arranged on a voxel grid video results the following videos were rendered in real time with the interactive viewer provided in our code release citation if you want to cite our work please use inproceedings reiser2021iccv author christian reiser and songyou peng and yiyi liao and andreas geiger title kilonerf speeding up neural radiance fields with thousands of tiny mlps booktitle international conference on computer vision iccv year 2021