Meta tags:

Headings (most frequently used words):

welcome, to, satoshi, ikehata, home, page, about, me, research, interests, selected, papers,

Text of the page (most frequently used words):
satoshi (45), #ikehata (45), and (40), proceedings (32), vision (24), conference (24), computer (22), aizawa (22), kiyoharu (21), for (20), ieee (18), paper (13), stereo (13), image (13), photometric (12), international (12), the (11), from (10), sato (10), accepted (9), 360 (9), ikuro (9), 2026 (8), pattern (8), recognition (8), processing (8), icip (8), tokyo (8), cvpr (7), using (7), based (7), oral (7), presentation (7), detection (7), 2025 (7), university (7), cvf (6), 2024 (6), rei (6), kawakami (6), 2022 (5), information (5), 2014 (5), images (5), european (5), eccv (5), learning (5), program (5), machine (4), codes (4), pdf (4), sparse (4), volume (4), transactions (4), via (4), single (4), network (4), universal (4), under (4), leslie (4), wöhler (4), aaai (4), with (4), applicants (4), nii (4), master (4), sokendai (4), 2013 (3), 2023 (3), british (3), bmvc (3), supplementary (3), david (3), wipf (3), yasuyuki (3), matsushita (3), regression (3), annual (3), depth (3), workshop (3), applications (3), map (3), cost (3), filtering (3), procession (3), intelligence (3), general (3), surfaces (3), strategy (3), yasutaka (3), furukawa (3), page (3), arxiv (3), motion (3), geometric (3), 2017 (3), 2018 (3), github (3), non (3), 2021 (3), nerf (3), free (3), urban (3), perception (3), videos (3), research (3), researcher (3), since (3), form (3), topic (3), this (3), students (3), graduate (3), all (2), takashi (2), otonari (2), 2012 (2), robust (2), cho (2), super (2), resolution (2), geometry (2), source (2), bayesian (2), ryosuke (2), furuta (2), toshihiko (2), coarse (2), fine (2), iccv (2), project (2), 2016 (2), neural (2), nips (2), preprint (2), scale (2), geo (2), multimedia (2), tools (2), cnn (2), distortion (2), naoki (2), sugimoto (2), deep (2), miao (2), cao (2), representation (2), object (2), interest (2), field (2), scenes (2), yuta (2), asano (2), facial (2), nakamasa (2), inoue (2), masayuki (2), tanaka (2), artificial (2), diffusion (2), sakuya (2), ota (2), qing (2), kent (2), fujiwara (2), noise (2), generation (2), mizuki (2), takenawa (2), realistic (2), large (2), light (2), limited (2), multi (2), visualization (2), graphics (2), chen (2), april (2), june (2), september (2), selected (2), papers (2), worked (2), visiting (2), 2011 (2), got (2), supervision (2), prof (2), was (2), 2020 (2), lab (2), about (2), interested (2), k00501 (2), please (2), mou (2), number (2), note (2), will (2), accept (2), have (2), jst (2), google (2), internship (2), ist (2), doctor (2), school (2), communication (2), engineering (2), target (2), position (2), opening (2), opportunity (2), informatics (2), institute (2), copyright, rights, reserved, highlight, yuki, sawabe, access, vol, 89124, 89133, signal, association, summit, apsipa, asc, confidence, refinement, corrupted, maps, hyunjin, yoo, margrit, gelautz, 11th, ivmsp, video, technologies, upsampling, inpainting, internal, statistics, appearance, analysis, tpami, diffuse, constrained, bivariate, isotropic, yamasaki, efficient, hang, yan, 2015, structured, indoor, modeling, ivaylo, boyadzhiev, shan, corr, abs, 1612, 01256, panoramic, structure, relationship, hao, xin, systems, 5560, 5570, sparsity, gated, recurrent, nets, kazuya, iwami, 6th, road, scene, understanding, autonomous, driving, eccv2018, drift, correction, camera, localization, tagged, yamaskai, appl, 12469, 12491, efficiency, enhanced, featuring, photometricc, convex, akito, nishiyama, aware, distorted, perspective, intersection, prediction, possible, direction, travel, transformer, self, attention, mechanism, site, global, lighting, contexts, dual, erp, degree, does, physical, interpretability, observation, improve, saliency, multiple, region, uniform, sampling, strategies, code, scalable, detailed, mask, tip, view, iou, website, entity, detecting, removing, moving, entities, ashish, tiwari, shanmuganathan, raman, towards, shot, through, material, estimation, relighting, physics, spectrally, multiplexed, unknown, spectral, composition, yusuke, sekikawa, chingwei, hsu, gumble, representing, unseen, objects, part, compositional, radiance, feilds, acm, applied, issue, investigating, anonymization, techniques, mao, tomita, asian, accv, simple, finetuning, bias, variance, ratios, layer, wise, gradients, ryo, moriai, rectified, lagrangian, out, distribution, modern, hopfield, networks, kazuma, kitazawa, takahito, aoto, tsuyoshi, takatani, eip, event, interval, profile, haruka, kumagai, preface, metric, perceptual, similarity, encanced, face, anomaization, akira, uchida, yuichi, yoshida, measuring, strength, dewarping, models, anomaly, pino, person, interaction, optimization, long, duration, customizable, arbitrary, sized, groups, december, building, evaluating, virtual, world, exploration, king, man, tam, zhaoyi, january, meets, leveraging, priors, illumination, cues, tatsuro, banno, 2420, 2426, 360citygml, interactive, system, integrating, citygml, model, hong, houyuan, chongjie, zhaoxi, bohan, shaocong, xianda, guo, xuhui, liu, yikai, wang, baochang, zhang, representations, iclr, normals, unified, feature, masahiro, kada, ryota, yoshihashi, teacher, guided, routing, mixture, experts, retrieving, refining, winning, tickets, jirong, shuhei, kurita, capframe, text, instructed, viewpoint, grounding, gaussian, pseudo, labels, full, publication, list, available, researchmap, algorithms, sensor, modality, fusion, human, cognition, interests, microsoft, asia, student, supervised, visual, computing, group, psychology, 2009, takao, november, post, doc, washington, louis, yamakata, matsui, utokyo, faq, submit, mainly, targets, universities, institutions, but, can, also, important, time, most, three, you, strong, ability, motivation, feel, apply, not, place, much, emphasis, prior, knowledge, specific, due, cannot, offer, interviews, contact, only, candidates, who, pass, initial, screening, schedule, interview, although, application, deadline, 10th, must, fill, march, 27th, fri, 1st, call, closed, started, currently, looking, talented, both, japan, other, countries, mail, sikehata, address, hitotsubashi, chiyoda, 101, 8430, senior, densoit, working, control, algoritym, laboratory, tech, concurrent, affiliations, advanced, studies, department, science, technology, associate, proffesor, national, phd, news, japanese, english, welcome, home,

Text of the page (random words):
satoshi ikehata welcome to satoshi ikehata s home page english japanese about news selected papers satoshi ikehata phd associate proffesor in national institute of informatics since 2017 concurrent affiliations informatics program graduate institute for advanced studies sokendai department of information communication engineering graduate school of information science and technology the university of tokyo senior researcher at densoit lab working in recognition control and learning algoritym laboratory in tokyo tech since 2023 address 2 1 2 hitotsubashi chiyoda ku tokyo 101 8430 e mail sikehata nii ac jp opportunity ph d position opening sokendai target ph d i am currently looking for talented ph d students both from japan and other countries at sokendai sokendai ph d program opportunity ph d position opening university of tokyo target master doctor i have started to accept master doctor students from graduate school of university of tokyo ist information communication engineering ist university of tokyo internship 2026 1st call of nii international internship program master ph d closed topic k00501 3d computer vision master ph d for applicants interested in my topic 3d computer vision k00501 3d computer vision 3d computer graphics please submit this google form the program mainly targets universities institutions with an mou with nii but a limited number of non mou applicants can also be accepted important note at this time i will accept at most three students if you have strong research ability and motivation please feel free to apply i do not place much emphasis on prior knowledge in my specific field due to the large number of applications i cannot offer interviews to all applicants so i will contact only the candidates who pass the initial screening to schedule an interview note although the application deadline for nii is april 10th jst applicants interested in my topic must fill this google form by march 27th fri jst program page interest form faq about me from 2020 to 2025 i was visiting researcher at aizawa yamakata matsui lab in utokyo since 2020 from june 2014 to november 2016 i was a post doc researcher in washington university in st louis and worked with dr yasutaka furukawa i got my ph d in 2014 and ms in 2011 in university of tokyo under the supervision of prof kiyoharu aizawa i got my ba of psychology from the university of tokyo in 2009 under the supervision of prof takao sato i worked for visual computing group at microsoft research asia as a visiting student in 2011 supervised by dr yasuyuki matsushita research interests 3d computer vision deep learning algorithms multi sensor modality fusion human cognition and perception selected papers full publication list is available at researchmap capframe text instructed viewpoint grounding in 3d gaussian scenes via geometric pseudo labels jirong li satoshi ikehata shuhei kurita ikuro sato in proceedings of european conference on computer vision eccv september 2026 retrieving and refining winning noise tickets for diffusion based motion generation sakuya ota qing yu kent fujiwara satoshi ikehata ikuro sato in proceedings of european conference on computer vision eccv september 2026 teacher guided routing for sparse vision mixture of experts masahiro kada ryota yoshihashi satoshi ikehata rei kawakami ikuro sato in proceedings of ieee cvf conference on computer vision and pattern recognition cvpr june 2026 light of normals unified feature representation for universal photometric stereo hong li houyuan chen chongjie ye zhaoxi chen bohan li shaocong xu xianda guo xuhui liu yikai wang baochang zhang et al international conference on learning representations iclr april 2026 360citygml realistic and interactive urban visualization system integrating citygml model and 360 videos tatsuro banno mizuki takenawa leslie wöhler satoshi ikehata kiyoharu aizawa ieee transactions on visualization and computer graphics 32 2 2420 2426 2026 geometry meets light leveraging geometric priors for universal photometric stereo under limited multi illumination cues king man tam satoshi ikehata yuta asano zhaoyi an rei kawakami in proceedings of annual aaai conference on artificial intelligence aaai accepted as an oral presentation january 2026 building and evaluating a realistic virtual world for large scale urban exploration from 360 videos mizuki takenawa naoki sugimoto leslie wöhler satoshi ikehata kiyoharu aizawa multimedia tools and applications preprint december 2025 pino person interaction noise optimization for long duration and customizable motion generation of arbitrary sized groups sakuya ota qing yu kent fujiwara satoshi ikehata ikuro sato in proceedings of cvf ieee international conference of computer vision iccv 2025 measuring distortion strength with dewarping diffusion models in anomaly detection akira uchida satoshi ikehata yuichi yoshida ikuro sato in proceedings of ieee international conference on image processing icip 2025 preface metric learning in perceptual facial similarity fo encanced face anomaization haruka kumagai leslie wöhler satoshi ikehata kiyoharu aizawa in proceedings of ieee international conference on image processing icip 2025 ps eip robust photometric stereo based on event interval profile kazuma kitazawa takahito aoto satoshi ikehata tsuyoshi takatani in proceedings of cvf ieee computer vision and pattern recognition cvpr 2025 rectified lagrangian for out of distribution detection in modern hopfield networks ryo moriai nakamasa inoue masayuki tanaka rei kawakami satoshi ikehata ikuro sato in proceedings of annual aaai conference on artificial intelligence aaai accepted as an oral presentation 2025 a simple finetuning strategy based on bias variance ratios of layer wise gradients mao tomita ikuro sato rei kawakami nakamasa inoue satoshi ikehata masayuki tanaka rei kawakami satoshi ikehata ikuro sato in proceedings of asian conference on computer vision accv 2024 accepted as an oral presentation investigating the perception of facial anonymization techniques in 360 videos leslie wöhler satoshi ikehata kiyoharu aizawa acm transactions on applied perception volume 21 issue 4 2024 gumble nerf representing unseen objects as part compositional neural radiance feilds yusuke sekikawa chingwei hsu satoshi ikehata rei kawakami ikuro sato in proceedings of international conference on image processing icip 2024 physics free spectrally multiplexed photometric stereo under unknown spectral composition satoshi ikehata yuta asano in proceedings of european conference on computer vision eccv 2024 accepted as an oral presentation towards single shot photometric stereo through material estimation and relighting ashish tiwari satoshi ikehata shanmuganathan raman in proceedings of european conference on computer vision eccv 2024 entity nerf detecting and removing moving entities in urban scenes takashi otonari satoshi ikehata kiyoharu aizawa in proceedings of cvf ieee computer vision and pattern recognition cvpr 2024 paper arxiv website field of view iou for object detection in 360 images miao cao satoshi ikehata kiyoharu aizawa ieee transactions on image procession tip 2023 paper scalable detailed and mask free universal photometric stereo satoshi ikehata in proceedings of cvf ieee computer vision and pattern recognition cvpr 2023 accepted as a highlight paper 2 5 paper arxiv code github non uniform sampling strategies for nerf on 360 images takashi otonari satoshi ikehata kiyoharu aizawa in proceedings of british machine vision conference bmvc 2022 saliency based multiple region of interest detection from a single 360 image yuki sawabe satoshi ikehata kiyoharu aizawa in ieee access vol 10 pp 89124 89133 2022 does physical interpretability of observation map improve the photometric stereo network satoshi ikehata in proceedings of ieee international conference on image processing icip 2022 paper dual erp representation for object detection in 360 degree images miao cao satoshi ikehata kiyoharu aizawa in proceedings of ieee international conference on image processing icip 2022 universal photometric stereo network using global lighting contexts satoshi ikehata in proceedings of cvf ieee computer vision and pattern recognition cvpr 2022 project site ps transformer learning sparse photometric stereo network using self attention mechanism satoshi ikehata in proceedings of british machine vision conference bmvc 2021 paper github intersection prediction from single 360 image via deep detection of possible direction of travel naoki sugimoto satoshi ikehata and kiyoharu aizawa in proceedings of british machine vision conference bmvc 2021 paper 360 single image super resolution via distortion aware network and distorted perspective images akito nishiyama satoshi ikehata kiyoharu aizawa in proceedings of international conference on image processing icip 2021 paper cnn ps cnn based photometricc stereo for general non convex surfaces satoshi ikehata in proceedings of european conference on computer vision eccv 2018 paper supplementary codes github efficiency enhanced cost volume filtering featuring coarse to fine strategy ryosuke furuta satoshi ikehata toshihiko yamaskai kiyoharu aizawa multimedia tools appl 77 10 12469 12491 2018 paper scale drift correction of camera geo localization using geo tagged images kazuya iwami satoshi ikehata kiyoharu aizawa 6th workshop on computer vision for road scene understanding and autonomous driving eccv2018 workshop 2018 preprint from bayesian sparsity to gated recurrent nets hao he bo xin satoshi ikehata david p wipf in proceedings of neural information processing systems nips 2017 accepted as an oral presentation 1 20 5560 5570 2017 nips proceedings panoramic structure from motion via geometric relationship detection satoshi ikehata ivaylo boyadzhiev qi shan yasutaka furukawa corr abs 1612 01256 2016 arxiv paper structured indoor modeling satoshi ikehata hang yan and yasutaka furukawa in proceedings of ieee international conference on computer vision iccv 2015 accepted as an oral presentation 3 30 project page coarse to fine strategy for efficient cost volume filtering ryosuke furuta satoshi ikehata toshihiko yamasaki and kiyoharu aizawa in proceedings of ieee international conference on image procession icip 2014 paper photometric stereo using constrained bivariate regression for general isotropic surfaces satoshi ikehata and kiyoharu aizawa in proceedings of computer vision and pattern recognition cvpr 2014 accepted as an oral presentation 5 76 paper supplementary source codes photometric stereo using sparse bayesian regression for general diffuse surfaces satoshi ikehata david wipf yasuyuki matsushita and kiyoharu aizawa ieee transactions on pattern analysis and machine intelligence tpami 2014 source codes depth map inpainting and super resolution based on internal statistics of geometry and appearance satoshi ikehata ji ho cho and kiyoharu aizawa in proceedings of ieee international conference on image procession icip 2013 pdf depth map upsampling using cost volume filtering ji ho cho satoshi ikehata hyunjin yoo margrit gelautz and kiyoharu aizawa in proceedings of 11th ieee ivmsp workshop 3d image video technologies and applications 2013 pdf confidence based refinement of corrupted depth maps satoshi ikehata and kiyoharu aizawa in proceedings of signal information processing association annual summit and conference apsipa asc 2012 pdf robust photometric stereo using sparse regression satoshi ikehata david wipf yasuyuki matsushita and kiyoharu aizawa in proceedings of computer vision and pattern recognition cvpr 2012 pdf supplementary codes copyright 2013 2026 satoshi ikehata all rights reserved