Meta tags:
author= chen;
Headings (most frequently used words):
portrait, neural, radiance, fields, from, single, image, abstract, links, paper, bibtex,
Text of the page (most frequently used words):
and (8), the (8), portrait (6), gao (4), #neural (4), radiance (4), fields (4), from (4), #single (4), chen (3), image (3), arxiv (3), 2020 (2), shih (2), yichang (2), lai (2), wei (2), sheng (2), liang (2), chia (2), kai (2), huang (2), jia (2), bin (2), bibtex (2), data (2), coming (2), soon (2), links (2), method (2), for (2), nerf (2), images (2), captures (2), mlp (2), models (2), using (2), generalization (2), abstract (2), copyright, article, portraitnerf, author, title, journal, preprint, 2012, 05903, year, paper, testing, training, code, present, estimating, headshot, while, has, demonstrated, high, quality, view, synthesis, requires, multiple, static, scenes, thus, impractical, casual, moving, subjects, this, work, propose, pretrain, weights, multilayer, perceptron, which, implicitly, volumetric, density, colors, with, meta, learning, framework, light, stage, dataset, improve, unseen, faces, train, canonical, coordinate, space, approximated, face, morphable, quantitatively, evaluate, controlled, demonstrate, real, showing, favorable, results, against, state, arts, virginia, tech, google, acknowledgments, examples, home,
Text of the page (random words):
portrait neural radiance fields from a single image home examples abstract links bibtex acknowledgments portrait neural radiance fields from a single image chen gao 1 yichang shih 2 wei sheng lai 2 chia kai liang 2 jia bin huang 1 1 virginia tech 2 google abstract we present a method for estimating neural radiance fields nerf from a single headshot portrait while nerf has demonstrated high quality view synthesis it requires multiple images of static scenes and thus impractical for casual captures and moving subjects in this work we propose to pretrain the weights of a multilayer perceptron mlp which implicitly models the volumetric density and colors with a meta learning framework using a light stage portrait dataset to improve the generalization to unseen faces we train the mlp in the canonical coordinate space approximated by 3d face morphable models we quantitatively evaluate the method using controlled captures and demonstrate the generalization to real portrait images showing favorable results against state of the arts links code coming soon training data coming soon testing data paper arxiv bibtex article gao portraitnerf author gao chen and shih yichang and lai wei sheng and liang chia kai and huang jia bin title portrait neural radiance fields from a single image journal arxiv preprint arxiv 2012 05903 year 2020 copyright chen gao 2020
|
