Meta tags:
description=;

Headings (most frequently used words):

comparisons, disco, portrait, distortion, correction, with, perspective, aware, 3d, gans, more, results, of, cropped, faces, abstract, method, overview, visual, quantitative, ablation, study, reference, bibtex, acknowledgements,

Text of the page (most frequently used words):
and (14), perspective (12), fried (11), the (10), our (10), tog (10), from (7), portrait (7), images (6), zhao (6), aware (6), distance (6), ours (6), input (6), #comparisons (6), results (5), iccv (5), method (5), collected (4), wild (4), distortion (4), with (4), wang (4), huang (4), gan (4), facial (4), siggraph (4), wacv (4), shih (4), camera (4), visual (4), distortions (4), for (3), their (3), disco (3), correction (3), gans (3), jia (3), bin (3), arxiv (3), tzaban (3), asia (3), inversion (3), optimization (3), cvpr (3), cmdp (3), photos (3), reference (3), ablation (3), study (3), man (3), face (3), code (3), more (3), university (3), data (2), are (2), ohad (2), yajie (2), zhixiang (2), lun (2), liu (2), shin (2), ichi (2), satoh (2), sizhuo (2), guru (2), krishnan (2), jian (2), 2023 (2), stitch (2), 2022 (2), yang (2), using (2), portraits (2), direct (2), paste (2), scheduling (2), all (2), woman (2), girl (2), close (2), captured (2), short (2), often (2), suffer (2), distorted (2), focal (2), length (2), corrects (2), show (2), that (2), website, design, adapted, hypernerf, sunstage, special, thanks, providing, sharing, web, internet, under, sources, here, common, creative, acknowledgements, article, wang2023disco, title, author, journal, preprint, year, bibtex, rotem, ron, mokady, rinon, gal, amit, bermano, daniel, cohen, time, based, editing, real, videos, jaehoon, kyusun, cho, daewon, choi, kwangrok, ryoo, seungryong, kim, pose, meng, johannes, kopf, photography, context, layered, depth, inpainting, 2020, burgos, artizzu, xavier, ronchi, matteo, ruggero, perona, pietro, caltech, multi, set, caltechdata, https, doi, org, 22002, 20110, zeng, tianye, weikai, chen, chloe, legendre, xinglei, ren, jun, xing, ari, shapiro, hao, learning, undistortion, 2019, eli, shechtman, dan, goldman, adam, finkelstein, manipulation, acm, proc, 2016, background, warping, tuning, pipeline, reparameterization, init, opt, proposed, conduct, borrowed, denotes, implementation, demo, quantitative, used, photo, two, sequence, overview, distances, resulting, exaggerated, features, unnatural, unattractive, appearances, propose, simple, yet, effective, correcting, single, first, perform, image, jointly, optimizing, intrinsic, extrinsic, parameters, latent, address, ambiguity, joint, develop, starting, reparametrizations, geometric, regularization, rendering, proper, effectively, produces, natural, looking, experiments, compares, favorably, against, previous, approaches, qualitatively, quantitatively, showcase, numerous, examples, validating, applicability, will, release, evaluation, protocol, facilitate, future, work, abstract, correct, can, process, diverse, cropped, faces, near, range, undesired, these, synthesizes, pleasant, views, enlarging, subject, virtually, paper, snap, inc, maryland, college, park, national, ming, chiao, tung, nii, tokyo,

Text of the page (random words):
disco portrait distortion correction with perspective aware 3d gans d i s c o portrait distortion correction with perspective aware 3d gans zhixiang wang yu lun liu jia bin huang shin ichi satoh sizhuo ma guru krishnan jian wang university of tokyo nii national yang ming chiao tung university university of maryland college park snap inc paper arxiv code portrait photos captured from a near range distance often suffer from undesired perspective distortions disco corrects these perspective distortions and synthesizes more pleasant views by virtually enlarging focal length and camera to subject distance more results of cropped faces we show that our method can process diverse in the wild face images correct in the wild distorted images collected by us and by zhao iccv 19 abstract close up facial images captured at short distances often suffer from perspective distortion resulting in exaggerated facial features and unnatural unattractive appearances we propose a simple yet effective method for correcting perspective distortions in a single close up face we first perform gan inversion using a perspective distorted input facial image by jointly optimizing the camera intrinsic extrinsic parameters and face latent code to address the ambiguity of joint optimization we develop starting from a short distance optimization scheduling reparametrizations and geometric regularization re rendering the portrait at a proper focal length and camera distance effectively corrects perspective distortions and produces more natural looking results our experiments show that our method compares favorably against previous approaches qualitatively and quantitatively we showcase numerous examples validating the applicability of our method on in the wild portrait photos we will release our code and the evaluation protocol to facilitate future work method overview visual comparisons sequence man 1 woman 1 two man girl 1 man 2 girl 2 woman 2 3d photo shih cvpr 20 fried tog 16 ko wacv 23 tzaban siggraph asia 22 ours visual comparisons on our collected images input ours fried tog 16 ko wacv 23 visual comparisons on images collected by zhao iccv 19 input ours fried tog 16 zhao iccv 19 visual comparisons on cmdp used in fried tog 16 input fried tog 16 ours reference quantitative comparisons we conduct comparisons on cmdp results of fried tog 16 is borrowed from their demo fried tog 16 denotes our re implementation of fried tog 16 ablation study ablation study of our proposed perspective aware 3d gan inversion input w o all w o all camera opt w o distance init w o scheduling w o reparameterization ours reference ablation study of our pipeline input direct paste stitch tuning background warping bw bw direct paste ours reference fried tog 16 ohad fried eli shechtman dan b goldman and adam finkelstein perspective aware manipulation of portrait photos acm tog proc siggraph 2016 zhao iccv 19 yajie zhao zeng huang tianye li weikai chen chloe legendre xinglei ren jun xing ari shapiro and hao li learning perspective undistortion of portraits iccv 2019 burgos artizzu xavier ronchi matteo ruggero perona pietro 2022 caltech multi distance portraits cmdp 1 0 data set caltechdata https doi org 10 22002 d1 20110 shih cvpr 20 meng li shih shih yang su johannes kopf jia bin huang 3d photography using context aware layered depth inpainting cvpr 2020 ko wacv 23 jaehoon ko kyusun cho daewon choi kwangrok ryoo seungryong kim 3d gan inversion with pose optimization wacv 2023 tzaban siggraph asia 22 rotem tzaban ron mokady rinon gal amit bermano daniel cohen or stitch it in time gan based facial editing of real videos siggraph asia 2022 bibtex article wang2023disco title disco portrait distortion correction with perspective aware 3d gans author zhixiang wang yu lun liu jia bin huang shin ichi satoh sizhuo ma guru krishnan jian wang journal arxiv preprint arxiv year 2023 acknowledgements special thanks to yajie zhao for providing their results and data ohad fried for sharing their results on web our collected in the wild images are from internet under common creative sources are here the website design is adapted from sunstage and hypernerf