Meta tags:

Headings (most frequently used words):

and, in, models, the, to, for, research, affiliated, graduate, students, forecasting, video, tasks, social, using, language, visual, data, analysis, styles, learning, of, art, shiry, ginosar, news, group, teaching, recent, talks, selected, publications, assistant, professors, former, affiliates, visitors, questioning, platonic, convergence, motion, wild, evaluating, on, empowerment, gain, causal, model, construction, mapping, perception, behavior, artificial, neural, networks, gaussian, splatting, masked, autoencoders, poly, autoregression, pose, priors, from, kiva, kid, inspired, analogies, lmms, synergy, synchrony, couple, dances, diffusion, as, mining, tools, can, learn, listen, historical, image, learning2listen, audio, conditioned, contrastive, textures, body2hands, modeling, minutiae, gestures, temporal, patterns, factorize, relight, city, individual, conversational, gesture, everybody, dance, now, burgeoning, computer, symbiosis, century, portraits, object, detection, abstract, speech, recognition, information, intensive, editable, code, histories, crowdsourced, phetch, human, computation, game, see, google, scholar, full, list,

Text of the page (most frequently used words):
and (145), the (104), for (44), shiry (41), #ginosar (40), pdf (28), models (25), #computer (21), learning (20), with (20), visual (20), from (20), data (19), that (18), video (18), motion (18), project (17), page (17), social (17), can (16), vision (15), this (15), human (14), our (14), image (13), language (13), behavior (13), are (12), undergrad (11), their (11), year (11), art (11), author (11), efros (11), title (10), speech (10), alexei (10), prediction (10), malik (9), jitendra (9), mining (9), pose (9), forecasting (9), research (8), luis (8), pages (8), ieee (8), analysis (8), use (8), code (8), model (8), world (8), based (8), representation (8), learn (8), approach (8), now (7), berkeley (7), 2025 (7), booktitle (7), which (7), workshop (7), recognition (7), not (7), historical (7), dance (7), large (7), phd (6), perception (6), inproceedings (6), computing (6), acm (6), using (6), detection (6), part (6), present (6), dataset (6), into (6), synthesis (6), cvpr (6), conversational (6), gesture (6), hand (6), gestures (6), audio (6), interaction (6), future (6), neural (6), improving (5), game (5), von (5), ahn (5), blum (5), manuel (5), 2007 (5), volume (5), natural (5), current (5), hartmann (5), transactions (5), system (5), multi (5), input (5), but (5), tasks (5), like (5), eccv (5), methods (5), one (5), object (5), theory (5), elements (5), time (5), students (5), chan (5), caroline (5), zhou (5), given (5), novel (5), target (5), realistic (5), andrew (5), single (5), evonne (5), trevor (5), darrell (5), body (5), simons (5), institute (5), machine (4), you (4), web (4), kedia (4), mihir (4), liu (4), conference (4), factors (4), organization (4), phetch (4), computation (4), images (4), used (4), agrawala (4), maneesh (4), journal (4), number (4), 2013 (4), 2014 (4), information (4), has (4), kate (4), imaging (4), portraits (4), iccv (4), much (4), aubry (4), mathieu (4), tinghui (4), 2019 (4), out (4), styles (4), wild (4), speaker (4), city (4), google (4), modeling (4), insight (4), textures (4), synthesize (4), listener (4), first (4), supervision (4), level (4), unified (4), 2024 (4), across (4), framework (4), agent (4), gaussian (4), artificial (4), causal (4), intelligence (4), empowerment (4), work (4), ttic (3), accessibility (3), proceedings (3), chi (3), paper (3), search (3), international (3), evaluation (3), play (3), improve (3), upon (3), article (3), crowdsourced (3), visualization (3), publisher (3), bjoern (3), explore (3), helps (3), authoring (3), stage (3), examples (3), editable (3), histories (3), pombo (3), fernando (3), task (3), study (3), intensive (3), applications (3), been (3), people (3), cubist (3), daniel (3), brown (3), timothy (3), just (3), good (3), objects (3), abstract (3), may (3), propose (3), brain (3), rakelly (3), sachs (3), sarah (3), yin (3), brian (3), lee (3), crystal (3), century (3), record (3), american (3), high (3), school (3), 2017 (3), doi (3), too (3), them (3), what (3), shen (3), honig (3), elizabeth (3), burgeoning (3), symbiosis (3), 2018 (3), 3186655 (3), new (3), computers (3), everybody (3), person (3), only (3), translation (3), between (3), mapping (3), amir (3), bar (3), gefen (3), kohavi (3), owens (3), individual (3), pattern (3), arm (3), generate (3), videos (3), factorize (3), relight (3), permanent (3), scene (3), training (3), scale (3), same (3), learned (3), dissertation (3), non (3), communication (3), settings (3), 2022 (3), conditioned (3), contrastive (3), angjoo (3), kanazawa (3), listen (3), than (3), comes (3), method (3), how (3), sanjay (3), subramanian (3), when (3), predicting (3), these (3), diffusion (3), diverse (3), central (3), jathushan (3), couple (3), physical (3), observe (3), eunice (3), yiu (3), benchmark (3), inspired (3), analogies (3), multimodal (3), children (3), psychology (3), neerja (3), thakkar (3), splatting (3), networks (3), responses (3), understanding (3), variability (3), submission (3), scholar (3), june (3), was (3), see (2), here (2), reality (2), second (2), chicago (2), have (2), position (2), generative (2), ruoran (2), 2006 (2), press (2), honorable (2), best (2), icassp (2), hcir (2), descriptions (2), identifying (2), redundancy (2), exposing (2), provenance (2), willett (2), wesley (2), steinitz (2), avital (2), bjorn (2), graphics (2), analysts (2), workers (2), software (2), changes (2), multiple (2), interactions (2), document (2), coding (2), results (2), available (2), detecting (2), haas (2), 101 (2), recognizing (2), paintings (2), also (2), align (2), goal (2), mimic (2), then (2), must (2), whether (2), find (2), while (2), abstraction (2), further (2), krähenbühl (2), philipp (2), computational (2), yearbooks (2), face (2), deep (2), month (2), about (2), captured (2), they (2), since (2), knowledge (2), frontal (2), allow (2), frame (2), varying (2), content (2), such (2), look (2), dwivedi (2), karan (2), xrds (2), spring (2), 1145 (2), issue (2), sway (2), source (2), transfer (2), subject (2), problem (2), temporal (2), temporally (2), generation (2), perform (2), cross (2), modal (2), snavely (2), noah (2), 2020 (2), disentangle (2), outdoor (2), scenes (2), street (2), where (2), through (2), changing (2), ones (2), patterns (2), hanbyul (2), joo (2), body2hands (2), infer (2), hands (2), dynamics (2), prior (2), shape (2), estimation (2), verbal (2), formulate (2), facial (2), synchronous (2), technically (2), extend (2), show (2), better (2), consistent (2), heritage (2), two (2), watermark (2), repeated (2), artwork (2), collections (2), both (2), problems (2), fine (2), different (2), mid (2), features (2), dan (2), klein (2), text (2), appropriate (2), lexical (2), semantic (2), juice (2), tokens (2), innovation (2), ioannis (2), siglidis (2), tools (2), accurate (2), does (2), explicitly (2), all (2), most (2), previous (2), works (2), focus (2), datasets (2), lea (2), müller (2), rajasegaran (2), synergy (2), synchrony (2), during (2), demonstrate (2), alison (2), gopnik (2), kiva (2), kid (2), critical (2), reasoning (2), lmms (2), priors (2), should (2), contact (2), optimization (2), poly (2), chen (2), talk (2), masked (2), autoencoders (2), reconstructions (2), activity (2), surprisingly (2), ofc (2), aic (2), neurons (2), identify (2), role (2), rich (2), nodes (2), representations (2), gain (2), construction (2), adults (2), interventions (2), controllable (2), jacob (2), walker (2), carl (2), doersch (2), joão (2), carreira (2), evaluating (2), under (2), uncertainty (2), takes (2), convergence (2), platonic (2), recorded (2), token (2), affiliated (2), graduate (2), assistant (2), neuroscience (2), fundamental (2), scientific (2), group (2), capabilities (2), museum (2), spent (2), years (2), visiting (2), faculty (2), postdoctoral (2), fellow (2), university (2), department (2), jose, cruzado, facebook, labs, princeton, tau, metaai, mit, actively, looking, third, math, stats, programs, who, excelled, classes, interested, think, qualified, send, note, 31270, fall, von2006improving, sigchi, systems, coverage, mentioned, nominee, award, von2007improving, acoustics, signal, processing, 1209, online, collects, side, effect, well, engines, willett2013identifying, 2198, 2206, lets, paid, crowd, sets, interactively, examine, build, insights, ginosar2013authoring, 26th, annual, symposium, user, interface, technology, 485, 494, uist, ide, extension, allowing, propagate, insertions, deletions, modifications, throughout, saved, stages, marti, hearst, designing, growing, importance, especially, mobile, less, done, editing, presents, modern, publicly, incollection, ginosar2014detecting, workshops, 116, springer, publishing, visart, other, depictions, form, recognize, outside, however, always, strive, picasso, test, invariance, fragmentation, humans, significantly, outperform, exhibit, similarly, graceful, degradation, increases, corroborating, ginosar2017yearbooks, 421, 431, keywords, market, sociology, statistics, dating, 1109, tci, 2699865, sept, extreme, 2015, september, many, details, written, records, because, mundane, describe, words, fortunately, invention, camera, ever, increasing, photographs, capture, otherwise, lost, plethora, artifacts, documenting, culture, treasure, trove, yet, untapped, historians, 921, facing, yearbook, photos, glimpse, voluminous, evaluated, manually, collected, provide, constant, reference, therefore, consider, issues, decade, defining, style, trends, fashion, norms, over, makes, ginosar2018art, issue_date, apr, issn, 1528, 4972, numpages, url, http, org, acmid, address, york, usa, crossroads, magazine, archive, help, understand, teach, chan2019dance, check, magic, app, dancing, performance, amateur, after, few, minutes, performing, standard, moves, per, spatio, smoothing, detections, intermediate, appearance, adapt, setup, coherent, including, ginosar2019, often, accompanied, plausible, along, sound, specifically, monologue, train, unlabeled, noisy, pseudo, ground, truth, automatic, release, specific, liu2020city, european, illumination, facilitate, assemble, timelapse, imagery, view, locations, repeatedly, disentangled, manipulate, ways, lighting, effects, geometry, appendix, corona, minutiae, 2021, domain, builds, strongly, correlated, alone, medhini, narasimhan, wacv, oral, mention, strumming, beat, creating, frames, associated, liwen, hao, deterministic, dyadic, delicate, expect, listeners, eye, synchronize, ours, mirror, emotions, annotate, raw, vector, quantization, sequence, encoding, vae, coupled, transformer, mechanism, performs, competitive, respond, learning2listen, robin, champenois, ilaria, pastrolin, morgane, rousselot, oumayma, bounou, tom, monnier, spyros, gidaris, françois, bougard, pierre, guillaume, raverdy, marie, françoise, limon, christine, bénévent, marc, smith, olivier, poncet, bender, béatrice, joyeux, prunel, ijcv, spatially, feature, matching, analyze, shot, challenges, representative, encountered, cultural, limited, grained, several, modalities, define, similarity, score, geometric, verification, spatial, consistency, tune, box, weak, 2023, spoken, responds, emotionally, manner, semantics, crucial, example, smile, despite, uneasy, laughter, squeeze, possible, pretrained, finetuning, autoregressively, response, transcript, treat, atomic, easily, ingestible, finetune, llms, sequences, main, squeezing, aleksander, holynski, spotlight, summarize, key, advantages, scales, traditional, correspondence, approaches, require, comparing, pairs, terms, furthermore, allows, translating, class, labels, analyzing, opening, possibilities, vongani, maluleke, georgios, pavlakos, swing, couples, genre, tight, coupling, context, challenging, instead, greatly, benefits, considering, partners, influence, dances, maan, qraitem, anisa, noor, majhi, charlie, rose, wong, yutong, bai, saenko, iclr, colab, closes, gap, benchmarks, foundational, analogical, even, young, poorly, developmental, llm, tell, hug, someone, your, neck, shoulder, waist, bottom, probably, knee, detailed, books, leverage, converting, descriptors, generated, lmm, tractable, losses, constrain, estimates, involving, tara, sadjadpour, rajasegeran, autoregressive, wide, range, simple, unifying, action, car, trajectory, general, autoregression, xinlei, ruilong, christoph, feichtenhofer, gave, trained, starkly, contrasts, usage, restricted, employ, primitives, gaussians, nate, dolensek, shi, doris, tsao, cosyne, poster, set, embed, reconstruct, support, astonishingly, optimal, stimuli, each, region, interpretable, axes, partner, angle, distance, finally, subset, behavioral, production, electrical, microstimulation, evoking, gaze, shifts, movements, suggest, regions, contain, transforming, positioning, cognition, ann, uncovering, kelsey, allen, hard, life, mind, continuity, towards, philosophical, 2026, sensitive, controllability, generalization, directed, prioritize, undirected, shift, toward, embracing, pure, don, seek, variable, pedro, vélez, luisa, polania, cabrera, guangyao, sayna, ebrahimi, rishabh, kabra, maks, ovsjanikov, gfrozen, predict, states, levels, requires, stochastic, frozen, account, via, module, certain, properly, evaluate, dense, point, trajectories, enabling, scalable, complex, species, unseen, agents, generalizes, sophia, koepke, daniil, zverev, back, plato, cave, examining, representational, converging, hypothesis, says, yes, evidence, more, fragile, looks, organisms, perceive, within, umwelt, questioning, full, list, selected, publications, observations, right, next, really, know, neurips, icml, recent, talks, teaching, exceptional, undergraduate, recruting, tsai, hemang, jeetendra, jangle, mai, varsha, ramakrishnan, vivien, nguyen, jasmine, collins, former, affiliates, visitors, uriel, barron, kaylene, stocking, professors, focuses, answer, questions, anticipate, generalize, contexts, intrinsic, motivation, little, creativity, scratch, building, grounded, interdisciplinary, drawing, arts, perspective, higher, lower, recordings, iterations, organized, summer, cluster, organizing, chairing, coming, studying, viewpoint, bringing, together, top, researchers, come, join, news, official, bio, covered, amongst, others, featured, exhibited, collection, helped, inspire, founding, startup, later, acquired, roblox, deutsches, israeli, design, pbs, nova, washington, post, wall, yorker, recently, researcher, previously, advised, completed, science, before, joining, lab, earlier, career, academic, roles, four, senior, engineer, endeca, carnegie, mellon, alyosha, deepmind, professor, intersection, hold, member, global, centre, intelligences, edu, bluesky, twitter, dot, homepage,

Text of the page (random words):
unified evaluation framework for frozen video understanding and synthesis models that takes the uncertainty of the future into account via a diffusion forecasting module jacob c walker pedro vélez luisa polania cabrera guangyao zhou sayna ebrahimi rishabh kabra carl doersch maks ovsjanikov joão carreira shiry ginosar gfrozen forecasting a unified evaluation in submission pdf empowerment gain and causal model construction children adults don t just explore they seek empowerment controllable and variable interventions when it comes to goal directed work people prioritize controllable variability a k a empowerment but in undirected play we shift toward embracing pure variability eunice yiu kelsey allen shiry ginosar alison gopnik empowerment gain and causal model construction children and adults are sensitive to controllability and variability in their causal generalization and interventions world models a g i and the hard problem s of life mind continuity towards a unified understanding of natural and artificial intelligence issue of philosophical transactions a 2026 in press pdf mapping social perception to social behavior using artificial neural networks ann approach for uncovering representations in social brain nodes by training a set of neural networks to embed and reconstruct videos from neural activity we surprisingly observe that neural responses in ofc and aic support astonishingly accurate reconstructions of social videos we further extend this approach to generate optimal stimuli for neurons from each region and identify interpretable axes of neural responses like social partner angle and distance finally we identify a subset of neurons predicting social behavioral responses and observe a causal role of activity in aic and ofc in social behavior production with electrical microstimulation evoking gaze shifts and facial movements these results suggest that these frontal brain regions contain a surprisingly rich code for social scenes and play a critical role in transforming social perception into behavior positioning them as central nodes in social cognition nate dolensek shi chen shiry ginosar and doris tsao mapping social perception to social behavior using artificial neural networks cosyne 2025 poster pdf gaussian splatting for masked autoencoders 3d gaussians as a learned mid level image representation we present the first method to employ 3d gaussian primitives in an image representation learning framework trained on large image datasets our learning based method starkly contrasts with the current usage of gaussian splatting which is restricted to optimization based single scene reconstructions jathushan rajasegaran xinlei chen ruilong li christoph feichtenhofer jitendra malik and shiry ginosar gaussian splatting for masked autoencoders pdf project page the talk i gave at the simons institute at berkeley poly autoregression a general framework for behavior prediction in multi agent settings we demonstrate that we can formulate a wide range of problems as social behavior prediction we present a simple framework for predicting agent behavior in multi agent settings unifying human action forecasting car trajectory prediction and hand object interaction forecasting neerja thakkar tara sadjadpour jathushan rajasegeran shiry ginosar and jitendra malik poly autoregressive prediction for modeling interactions cvpr 2025 pdf project page pose priors from language models improving pose estimates involving physical contact using lmms central insight an llm can tell you e g that to hug someone your arm should go on their neck shoulder waist or bottom but probably not on their knee and all of this knowledge comes from detailed descriptions of physical contact in books we leverage this insight to improve pose estimation by converting natural language descriptors generated by a large multimodal model lmm into tractable losses to constrain 3d pose optimization sanjay subramanian evonne ng lea müller dan klein shiry ginosar and trevor darrell pose priors from language models cvpr 2025 pdf project page kiva kid inspired visual analogies for lmms a benchmark for visual analogies inspired by developmental psychology we present a benchmark that closes a critical gap in current benchmarks for foundational models visual analogical reasoning which even young children can do but in which large multimodal models perform poorly eunice yiu maan qraitem anisa noor majhi charlie rose wong yutong bai shiry ginosar alison gopnik and kate saenko kiva kid inspired visual analogies for large multimodal models iclr 2025 pdf project page benchmark data colab synergy and synchrony in couple dances how does social interaction influence one s behavior we focus our analysis on motion behavior during swing a couples dance genre with tight physical coupling for which we present an in the wild video dataset we demonstrate that single person future motion prediction in this context is challenging instead we observe that prediction greatly benefits from considering the interaction partners behavior vongani maluleke lea müller jathushan rajasegaran georgios pavlakos shiry ginosar angjoo kanazawa and jitendra malik synergy and synchrony in couple dance pdf project page diffusion models as data mining tools image synthesis models can be used for visual data mining central insight since generative models learn an accurate representation of their training data we can use them to summarize the data by mining for visual patterns our analysis by synthesis approach to data mining has two key advantages first it scales much better than traditional correspondence based approaches as it does not require explicitly comparing all pairs of visual elements second while most previous works on visual data mining focus on a single dataset our approach works on diverse datasets in terms of content and scale furthermore our approach allows for translating visual elements across class labels and analyzing consistent changes opening up new possibilities for data mining in the future ioannis siglidis aleksander holynski alexei a efros mathieu aubry and shiry ginosar diffusion models as data mining tools eccv 2024 spotlight pdf project page can language models learn to listen squeezing lexical semantic juice out of large language models given only the spoken text of a speaker we synthesize a realistic synchronous listener our text based model responds in an emotionally appropriate manner when lexical semantics is crucial for example when it is not appropriate to smile despite a speaker s uneasy laughter technically we squeeze out as much semantic juice as possible from a pretrained large language model by finetuning it to autoregressively generate realistic 3d listener motion in response to the input transcript main innovation we treat atomic gesture elements as novel language tokens easily ingestible by language models we can then finetune llms to synthesize motion by predicting sequences of these elements evonne ng sanjay subramanian dan klein angjoo kanazawa trevor darrell and shiry ginosar can language models learn to listen iccv 2023 pdf video project page historical image analysis a unified approach to historical watermark recognition and repeated pattern detection in artwork collections we explore two applications of computer vision to analyze historical data watermark recognition and one shot repeated pattern detection in artwork collections both problems present computer vision challenges which are representative of the ones encountered in cultural heritage applications limited supervision is available the tasks are fine grained recognition and the data comes in several different modalities we use a single method for both tasks we define an image similarity score based on geometric verification of mid level features and show how spatial consistency can be used to fine tune out of the box features for the target dataset with weak or no supervision xi shen robin champenois shiry ginosar ilaria pastrolin morgane rousselot oumayma bounou tom monnier spyros gidaris françois bougard pierre guillaume raverdy marie françoise limon christine bénévent marc smith olivier poncet k bender béatrice joyeux prunel elizabeth honig alexei a efros and mathieu aubry spatially consistent feature matching and learning for heritage image analysis ijcv 2022 pdf project page learning2listen learning to respond like a good listener given a speaker we synthesize a realistic synchronous listener to do this we learn human interaction 101 the delicate dance of non verbal communication we expect good listeners to look us in the eye synchronize their motion with ours and mirror our emotions you can t annotate this so we must learn from raw data technically we are the first to extend vector quantization methods to motion synthesis we show that our novel sequence encoding vq vae coupled with a transformer based prediction mechanism performs much better than competitive methods for motion generation evonne ng hanbyul joo liwen hu hao li trevor darrell angjoo kanazawa and shiry ginosar learning to listen modeling non deterministic dyadic facial motion cvpr 2022 pdf video project page audio conditioned contrastive video textures learning to synthesize video textures conditioned on audio we learn a representation for creating video textures using contrastive learning between the frames of a video and the associated audio medhini narasimhan shiry ginosar andrew owens alexei a efros and trevor darrell strumming to the beat audio conditioned contrastive video textures wacv 2022 best paper honorable mention oral pdf project page body2hands learning to infer 3d hands from conversational gesture body dynamics a novel learned deep prior of body motion for 3d hand shape synthesis and estimation in the domain of conversational gestures our model builds upon the insight that body motion and hand gestures are strongly correlated in non verbal communication settings we formulate the learning of this prior as a prediction task of 3d hand shape given body motion input alone evonne ng shiry ginosar trevor darrell and hanbyul joo body2hands learning to infer 3d hands from conversational gesture body dynamics cvpr 2021 pdf video project page modeling visual minutiae gestures styles and temporal patterns ph d dissertation dissertation appendix dissertation in the time of corona learning to factorize and relight a city disentangle changing factors from permanent ones we disentangle outdoor scenes into temporally varying illumination and permanent scene factors to facilitate training we assemble a city scale dataset of outdoor timelapse imagery from google street view time machine where the same locations are captured repeatedly through time our learned disentangled factors can be used to manipulate novel images in realistic ways such as changing lighting effects and scene geometry andrew liu shiry ginosar tinghui zhou alexei a efros and noah snavely learning to factorize and relight a city eccv 2020 pdf video project page inproceedings liu2020city author liu andrew and ginosar shiry and zhou tinghui and snavely noah and efros alexei a title learning to factorize and relight a city booktitle european conference on computer vision eccv year 2020 learning individual styles of conversational gesture audio to motion translation human speech is often accompanied by hand and arm gestures given audio speech input we generate plausible gestures to go along with the sound specifically we perform cross modal translation from in the wild monologue speech of a single speaker to their hand and arm motion we train on unlabeled videos for which we only have noisy pseudo ground truth from an automatic pose detection system we release a large video dataset of person specific gestures shiry ginosar amir bar gefen kohavi caroline chan andrew owens and jitendra malik learning individual styles of conversational gesture cvpr 2019 pdf project page inproceedings ginosar2019 title learning individual styles of conversational gesture author ginosar shiry and bar amir and kohavi gefen and chan caroline and owens andrew and malik jitendra booktitle computer vision and pattern recognition cvpr year 2019 everybody dance now do as i do motion transfer given a source video of a person dancing we can transfer that performance to a novel amateur target after only a few minutes of the target subject performing standard moves we pose this problem as a per frame image to image translation with spatio temporal smoothing using pose detections as an intermediate representation between source and target we learn a mapping from pose images to a target subject s appearance we adapt this setup for temporally coherent video generation including realistic face synthesis caroline chan shiry ginosar tinghui zhou and alexei a efros everybody dance now iccv 2019 pdf video project page check out the sway magic dance app inproceedings chan2019dance author chan caroline and ginosar shiry and zhou tinghui and efros alexei a title everybody dance now booktitle ieee international conference on computer vision iccv year 2019 the burgeoning computer art symbiosis computers help us understand art art helps us teach computers shiry ginosar xi shen karan dwivedi elizabeth honig and mathieu aubry the burgeoning computer art symbiosis xrds crossroads the acm magazine for students computers and art archive volume 24 issue 3 spring 2018 pages 30 33 pdf article ginosar2018art author ginosar shiry and shen xi and dwivedi karan and honig elizabeth and aubry mathieu title the burgeoning computer art symbiosis journal xrds issue_date spring 2018 volume 24 number 3 month apr year 2018 issn 1528 4972 pages 30 33 numpages 4 url http doi acm org 10 1145 3186655 doi 10 1145 3186655 acmid 3186655 publisher acm address new york ny usa a century of portraits what makes the 60 s look like the 60 s many details about our world are not captured in written records because they are too mundane or too abstract to describe in words fortunately since the invention of the camera an ever increasing number of photographs capture much of this otherwise lost information this plethora of artifacts documenting our visual culture is a treasure trove of knowledge as yet untapped by historians we present a dataset of 37 921 frontal facing american high school yearbook photos that allow us to use computation to glimpse into the historical visual record too voluminous to be evaluated manually the collected portraits provide a constant visual frame of reference with varying content we can therefore use them to consider issues such as a decade s defining style elements or trends in fashion and social norms over time shiry ginosar kate rakelly sarah sachs brian yin crystal lee philipp krähenbühl and alexei a efros a century of portraits a visual historical record of american high school yearbooks extreme imaging workshop iccv 2015 and ieee transactions on computational imaging september 2017 pdf project page articl...