Meta tags:

Headings (most frequently used words):

-

Text of the page (most frequently used words):
the (79), and (67), for (33), that (24), with (23), learning (21), qingqing (19), zhao (19), paper (17), from (15), our (15), motion (14), data (13), gordon (13), world (13), wetzstein (12), models (12), model (11), can (11), abstract (11), physical (11), based (11), this (11), using (10), are (9), inverse (9), bibtex (9), video (9), physics (9), author (8), title (8), year (8), simulation (8), 2024 (8), pose (8), vision (8), visual (8), human (8), such (7), scale (7), demonstrate (7), problems (7), code (7), webpage (7), introduce (7), method (7), state (6), arxiv (6), tasks (6), vla (6), action (6), first (5), inproceedings (5), solve (5), pde (5), journal (5), neural (5), have (5), more (5), multi (5), spatial (5), which (5), lamp (5), error (5), existing (5), yang (5), physavatar (5), humans (5), humanoid (5), humanoids (5), foundation (5), cot (5), reasoning (5), language (5), isospin (4), low (4), 22al (4), wave (4), not (4), zhang (4), via (4), art (4), graph (4), learned (4), gnns (4), parameters (4), system (4), deep (4), policy (4), challenging (4), underground (4), scenes (4), estimate (4), wang (4), yifan (4), control (4), avatars (4), xiang (4), donglai (4), zipeng (4), finn (4), chelsea (4), shadowing (4), real (4), cosmos (4), vlas (4), high (3), mirror (3), well (3), large (3), but (3), while (3), halo (3), another (3), dielectric (3), resonator (3), mode (3), highly (3), computational (3), applications (3), constrained (3), networks (3), 2022 (3), accurate (3), principled (3), solvers (3), simulating (3), systems (3), many (3), engineering (3), however (3), only (3), forward (3), these (3), 2023 (3), controllable (3), adaptive (3), time (3), evolution (3), open (3), their (3), surrogate (3), needs (3), mesh (3), imaging (3), permittivity (3), learn (3), prior (3), methods (3), retargeting (3), motions (3), characters (3), create (3), source (3), character (3), different (3), guandao (3), leonidas (3), guibas (3), dressed (3), observations (3), rendering (3), novel (3), training (3), digital (3), skills (3), customized (3), trained (3), platform (3), chain (3), thought (3), here (2), one (2), proton (2), 22si (2), properties (2), decay (2), transition (2), transitions (2), lying (2), states (2), compared (2), 22o (2), value (2), asymmetry (2), found (2), excited (2), interaction (2), orbit (2), bound (2), results (2), structure (2), provide (2), identify (2), 2020 (2), lang (2), owen (2), miller (2), minimum (2), volumes (2), show (2), structures (2), applied (2), nonlinear (2), david (2), lindell (2), icml (2), been (2), dynamics (2), interested (2), also (2), equation (2), given (2), measurements (2), initial (2), combined (2), conditions (2), other (2), tailin (2), takashi (2), maruyama (2), jure (2), leskovec (2), iclr (2), small (2), dynamic (2), fine (2), resolution (2), majority (2), required (2), compute (2), achieve (2), work (2), full (2), optimizes (2), network (2), gnn (2), allows (2), importance (2), computation (2), test (2), benchmark (2), long (2), born (2), operator (2), reflection (2), tomographic (2), yanting (2), petros (2), boufounos (2), saleh (2), nabi (2), hassan (2), mansour (2), recent (2), gpr (2), goal (2), distribution (2), posed (2), machine (2), under (2), approach (2), dataset (2), achieves (2), peizhuo (2), olga (2), sorkine (2), hornung (2), sca (2), article (2), cross (2), domain (2), various (2), has (2), computer (2), current (2), synthesis (2), datasets (2), hand (2), images (2), through (2), effectively (2), combines (2), features (2), noisy (2), poses (2), appearance (2), present (2), variates (2), automatic (2), setups (2), zheng (2), dubost (2), florian (2), lagun (2), dmitry (2), beeler (2), thabo (2), tombari (2), federico (2), modeling (2), photorealistic (2), crucial (2), build (2), view (2), clothes (2), purpose (2), temporal (2), physically (2), capabilities (2), humanplus (2), imitation (2), booktitle (2), robots (2), lack (2), pipeline (2), autonomous (2), egocentric (2), train (2), body (2), autonomously (2), objects (2), robot (2), demonstrations (2), twin (2), help (2), into (2), pre (2), nvidia (2), yao (2), moo (2), jin (2), kim (2), zhuoyang (2), yecheng (2), max (2), qianli (2), song (2), han (2), ankur (2), handa (2), ming (2), liu (2), tsung (2), lin (2), robotic (2), manipulation (2), goals (2), stanford (2), website, template, delayed, emissions, lightest, nucleus, projection, studied, silicon, array, surrounded, purity, germanium, detectors, branches, reduced, probabilities, determined, largest, far, 209ð96þ, shell, calculation, nonconserving, forces, including, related, introduces, explicitly, symmetry, breaking, force, describes, loosely, nature, functions, reproduce, observed, consistently, explain, observation, occurs, second, supporting, might, means, nuclei, lee, ribll, collaboration, review, letters, gamow, teller, reveals, misc, qzhaomodev, url, https, org, abs, 2008, 13241, publisher, global, lower, bounds, volume, computed, lagrangian, duality, designs, rely, sharp, tips, appear, sub, optimal, nanometer, feature, sizes, design, offers, orders, magnitude, possible, improvements, geometries, simultaneously, resonant, multiple, frequencies, efficiency, optics, unavoidable, penalties, must, accompany, multiresonant, qzhao2022graphpde, recently, established, fast, alternatives, application, domains, across, science, solving, constraints, defined, partial, differential, explore, sparse, set, recovering, condition, autodecoder, style, priors, suited, achieving, estimates, than, approaches, when, navier, stokes, equations, speedups, 90x, spotlight, tailingraphpde, pivotal, scientific, challenge, fraction, extremely, requires, very, grained, changing, slowly, modeled, coarser, scales, typical, use, uniform, resolve, finest, waste, huge, accuracy, jointly, learns, appropriate, resolutions, devote, regions, consists, actor, critic, refinement, coarsening, techniques, weighted, sum, cost, objective, adapt, varying, relative, tradeoff, inference, pdes, outperforms, reduction, able, adaptively, trade, off, improve, term, prediction, openreview, borngpr, under_review, developments, sensor, technologies, ground, penetrating, radar, new, opportunities, scattered, electromagnetic, wavefield, ill, difficult, formulate, computationally, expensive, propose, inspired, matter, setting, signal, recover, unknown, 400, maps, three, layer, background, numerical, improvement, mean, squared, over, reconstructing, layered, icassp, zhao2023pose, creating, believable, graphics, depend, extensive, often, impossible, obtain, accessible, since, static, easier, even, extracted, advancements, utilize, alternative, generates, plausible, transferring, capture, drastically, skeletons, experiments, target, performs, robustly, sets, ranging, few, artist, created, estimated, directly, additionally, conducted, user, study, indicated, participants, retargeted, enjoyable, watch, lifelike, exhibiting, fewer, artifacts, uses, arbitrary, architectures, differentiation, unbiased, variance, numerically, stable, monte, carlo, estimators, problem, zilu, deng, bharath, hariharan, wetzsteinn, siggraph, integration, zheng2024physavatar, others, preprint, 2404, 04421, avatar, struggle, reconstruct, clothed, framework, automatically, shape, along, fabric, adopt, aligned, gaussian, technique, spatio, tracking, renderer, intrinsic, material, integrates, simulator, garments, gradient, optimization, manner, enable, quality, renderings, loose, fitting, lighting, seen, marks, significant, advancement, towards, loop, eccv, fu2024humanplus, key, arguments, building, similar, form, factors, beings, leverage, massive, yet, doing, remained, practice, due, complexities, perception, lingering, gaps, between, morphologies, actuation, stack, level, reinforcement, hour, transfers, follow, rgb, camera, operators, teleoperate, collect, whole, collected, then, perform, supervised, behavior, cloning, skill, policies, allowing, complete, imitating, dof, 180cm, completing, wearing, shoe, stand, walk, unloading, warehouse, racks, folding, sweatshirt, rearranging, typing, greeting, 100, success, rates, corl, best, award, finalist, top, digitally, itself, developers, position, general, tuned, downstream, covers, curation, examples, post, tokenizers, builders, most, critical, society, make, weight, permissive, licenses, available, predict1, 2025, zhao2024cotvla, shown, potential, leveraging, pretrained, diverse, generalizable, sensorimotor, paradigm, utilizes, both, non, sources, primarily, focus, direct, input, output, mappings, lacking, intermediate, steps, complex, result, planning, incorporates, explicit, predicting, future, image, frames, autoregressively, before, generating, short, sequence, understand, generate, tokens, experimental, strong, performance, outperforming, benchmarks, research, scholar, github, email, cyanzhao, dot, edu, robotics, tesla, previously, completed, electrical, advised, prof, 赵青青,

Text of the page (random words):
g zhao 赵青青 i work on robotics at tesla ai previously i completed my ph d in electrical engineering at stanford advised by prof gordon wetzstein cyanzhao at stanford dot edu email github x scholar research cot vla visual chain of thought reasoning for vision language action models qingqing zhao yao lu moo jin kim zipeng fu zhuoyang zhang yecheng wu max li qianli ma song han chelsea finn ankur handa ming yu liu donglai xiang gordon wetzstein tsung yi lin webpage paper abstract bibtex video vision language action models vlas have shown potential in leveraging pretrained vision language models and diverse robot demonstrations for learning generalizable sensorimotor control while this paradigm effectively utilizes large scale data from both robotic and non robotic sources current vlas primarily focus on direct input output mappings lacking the intermediate reasoning steps crucial for complex manipulation tasks as a result existing vlas lack temporal planning or reasoning capabilities in this paper we introduce a method that incorporates explicit visual chain of thought cot reasoning into vision language action models vlas by predicting future image frames autoregressively as visual goals before generating a short action sequence to achieve these goals we introduce cot vla a state of the art 7b vla that can understand and generate visual and action tokens our experimental results demonstrate that cot vla achieves strong performance outperforming the state of the art vla model by 17 in real world manipulation tasks and 6 in simulation benchmarks inproceedings zhao2024cotvla author qingqing zhao yao lu moo jin kim zipeng fu zhuoyang zhang yecheng wu max li qianli ma song han chelsea finn ankur handa ming yu liu donglai xiang gordon wetzstein tsung yi lin title cot vla visual chain of thought reasoning for vision language action models booktitle arxiv year 2024 cosmos world foundation model platform for physical ai nvidia 2025 webpage paper abstract code video physical ai needs to be trained digitally first it needs a digital twin of itself the policy model and a digital twin of the world the world model in this paper we present the cosmos world foundation model platform to help developers build customized world models for their physical ai setups we position a world foundation model as a general purpose world model that can be fine tuned into customized world models for downstream applications our platform covers a video curation pipeline pre trained world foundation models examples of post training of pre trained world foundation models and video tokenizers to help physical ai builders solve the most critical problems of our society we make cosmos open source and our models open weight with permissive licenses available via nvidia cosmos predict1 humanplus humanoid shadowing and imitation from humans zipeng fu qingqing zhao qi wu gordon wetzstein chelsea finn corl 2024 best paper award finalist top 6 webpage paper abstract bibtex code video one of the key arguments for building robots that have similar form factors to human beings is that we can leverage the massive human data for training yet doing so has remained challenging in practice due to the complexities in humanoid perception and control lingering physical gaps between humanoids and humans in morphologies and actuation and lack of a data pipeline for humanoids to learn autonomous skills from egocentric vision in this paper we introduce a full stack system for humanoids to learn motion and autonomous skills from human data we first train a low level policy in simulation via reinforcement learning using existing 40 hour human motion datasets this policy transfers to the real world and allows humanoid robots to follow human body and hand motion in real time using only a rgb camera i e shadowing through shadowing human operators can teleoperate humanoids to collect whole body data for learning different tasks in the real world using the data collected we then perform supervised behavior cloning to train skill policies using egocentric vision allowing humanoids to complete different tasks autonomously by imitating human skills we demonstrate the system on our customized 33 dof 180cm humanoid autonomously completing tasks such as wearing a shoe to stand up and walk unloading objects from warehouse racks folding a sweatshirt rearranging objects typing and greeting another robot with 60 100 success rates using up to 40 demonstrations inproceedings fu2024humanplus author fu zipeng and zhao qingqing and wu qi and wetzstein gordon and finn chelsea title humanplus humanoid shadowing and imitation from humans booktitle arxiv year 2024 physavatar learning the physics of dressed 3d avatars from visual observations yang zheng qingqing zhao guandao yang wang yifan donglai xiang florian dubost dmitry lagun thabo beeler federico tombari leonidas guibas gordon wetzstein eccv 2024 webpage paper abstract bibtex code video modeling and rendering photorealistic avatars is of crucial importance in many applications existing methods that build a 3d avatar from visual observations however struggle to reconstruct clothed humans we introduce physavatar a novel framework that combines inverse rendering with inverse physics to automatically estimate the shape and appearance of a human from multi view video data along with the physical parameters of the fabric of their clothes for this purpose we adopt a mesh aligned 4d gaussian technique for spatio temporal mesh tracking as well as a physically based inverse renderer to estimate the intrinsic material properties physavatar integrates a physics simulator to estimate the physical parameters of the garments using gradient based optimization in a principled manner these novel capabilities enable physavatar to create high quality novel view renderings of avatars dressed in loose fitting clothes under motions and lighting conditions not seen in the training data this marks a significant advancement towards modeling photorealistic digital humans using physically based inverse rendering with physics in the loop article zheng2024physavatar title physavatar learning the physics of dressed 3d avatars from visual observations author zheng yang and zhao qingqing and yang guandao and yifan wang and xiang donglai and dubost florian and lagun dmitry and beeler thabo and tombari federico and guibas leonidas and others journal arxiv preprint arxiv 2404 04421 year 2024 neural control variates with automatic integration zilu li guandao yang qingqing zhao xi deng leonidas guibas bharath hariharan gordon wetzsteinn siggraph 2024 paper abstract bibtex we present a method that uses arbitrary neural network architectures as control variates with automatic differentiation to create unbiased low variance and numerically stable monte carlo estimators for various problem setups pose to motion cross domain motion retargeting with pose prior qingqing zhao peizhuo li wang yifan olga sorkine hornung gordon wetzstein sca 2024 webpage paper abstract bibtex code creating believable motions for various characters has long been a goal in computer graphics current learning based motion synthesis methods depend on extensive motion datasets which are often challenging if not impossible to obtain on the other hand pose data is more accessible since static posed characters are easier to create and can even be extracted from images using recent advancements in computer vision in this paper we utilize this alternative data source and introduce a neural motion synthesis approach through retargeting our method generates plausible motions for characters that have only pose data by transferring motion from an existing motion capture dataset of another character which can have drastically different skeletons our experiments show that our method effectively combines the motion features of the source character with the pose features of the target character and performs robustly with small or noisy pose data sets ranging from a few artist created poses to noisy poses estimated directly from images additionally a conducted user study indicated that a majority of participants found our retargeted motion to be more enjoyable to watch more lifelike in appearance and exhibiting fewer artifacts article zhao2023pose title pose to motion cross domain motion retargeting with pose prior author zhao qingqing and li peizhuo and yifan wang and sorkine hornung olga and wetzstein gordon journal sca year 2024 deep born operator learning for reflection tomographic imaging qingqing zhao yanting ma petros t boufounos saleh nabi hassan mansour icassp 2023 paper abstract bibtex code video recent developments in wave based sensor technologies such as ground penetrating radar gpr provide new opportunities for accurate imaging of underground scenes given measurements of the scattered electromagnetic wavefield the goal is to estimate the spatial distribution of the permittivity of the underground scenes however such problems are highly ill posed difficult to formulate and computationally expensive in this paper we propose a physics inspired machine learning based method to learn the wave matter interaction under the gpr setting the learned forward model is combined with a learned signal prior to recover the permittivity distribution of the unknown underground scenes we test our approach on a dataset of 400 permittivity maps with a three layer background which is challenging to solve using existing methods we demonstrate via numerical simulation that our method achieves a 50 improvement in mean squared error over the benchmark machine learning based solvers for reconstructing layered underground scenes inproceedings borngpr title deep born operator learning for reflection tomographic imaging author qingqing zhao yanting ma petros t boufounos saleh nabi hassan mansour journal under_review year 2022 learning controllable adaptive simulation for multi resolution physics tailin wu takashi maruyama qingqing zhao gordon wetzstein jure leskovec iclr 2023 spotlight webpage openreview paper abstract bibtex code simulating the time evolution of physical systems is pivotal in many scientific and engineering problems an open challenge in simulating such systems is their multi scale dynamics a small fraction of the system is extremely dynamic and requires very fine grained resolution while a majority of the system is changing slowly and can be modeled by coarser spatial scales typical learning based surrogate models use a uniform spatial scale which needs to resolve to the finest required scale and can waste a huge compute to achieve required accuracy in this work we introduce learning controllable adaptive simulation for multi scale physics lamp as the first full deep learning based surrogate model that jointly learns the evolution model and optimizes appropriate spatial resolutions that devote more compute to the highly dynamic regions lamp consists of a graph neural network gnn for learning the forward evolution and a gnn based actor critic for learning the policy of spatial refinement and coarsening we introduce learning techniques that optimizes lamp with weighted sum of error and computational cost as objective which allows lamp to adapt to varying relative importance of error vs computation tradeoff at inference time we test our method in a 1d benchmark of nonlinear pdes and a challenging 2d mesh based simulation we demonstrate that our lamp outperforms state of the art deep learning surrogate models with up to 60 5 error reduction and is able to adaptively trade off computation to improve long term prediction error inproceedings tailingraphpde title learning controllable adaptive simulation for multi scale physics author tailin wu takashi maruyama qingqing zhao gordon wetzstein jure leskovec journal iclr year 2023 learning to solve pde constrained inverse problems with graph networks qingqing zhao david b lindell gordon wetzstein icml 2022 webpage paper abstract bibtex code video learned graph neural networks gnns have recently been established as fast and accurate alternatives for principled solvers in simulating the dynamics of physical systems in many application domains across science and engineering however we are not only interested in a forward simulation but also in solving inverse problems with constraints defined by a partial differential equation pde here we explore gnns to solve such pde constrained inverse problems given a sparse set of measurements we are interested in recovering the initial condition or parameters of the pde we demonstrate that gnns combined with autodecoder style priors are well suited for these tasks achieving more accurate estimates of initial conditions or physical parameters than other learned approaches when applied to the wave equation or navier stokes equations we also demonstrate computational speedups of up to 90x using gnns compared to principled solvers inproceedings qzhao2022graphpde title learning to solve pde constrained inverse problems with graph networks author qingqing zhao and david b lindell and gordon wetzstein journal icml year 2022 minimum dielectric resonator mode volumes qingqing zhao lang zhang owen d miller paper abstract bibtex we show that global lower bounds to the mode volume of a dielectric resonator can be computed via lagrangian duality state of the art designs rely on sharp tips but such structures appear to be highly sub optimal at nanometer scale feature sizes and we demonstrate that computational inverse design offers orders of magnitude possible improvements our bound can be applied for geometries that are simultaneously resonant at multiple frequencies for high efficiency nonlinear optics applications and we identify the unavoidable penalties that must accompany such multiresonant structures misc qzhaomodev url https arxiv org abs 2008 13241 author qingqing zhao and lang zhang and owen d miller title minimum dielectric resonator mode volumes publisher arxiv year 2020 large isospin asymmetry in 22si 22o mirror gamow teller transitions reveals the halo structure of 22al j lee et al ribll collaboration physical review letters 2020 paper abstract β delayed one proton emissions of 22si the lightest nucleus with an isospin projection tz ¼ 3 are studied with a silicon array surrounded by high purity germanium detectors properties of β decay branches and the reduced transition probabilities for the transitions to the low lying states of 22al are determined compared to the mirror β decay of 22o the largest value of mirror asymmetry in low lying states by far with δ ¼ 209ð96þ is found in the transition to the first 1þ excited state shell model calculation with isospin nonconserving forces including the t ¼ 1 j ¼ 2 3 interaction related to the s1 2 orbit that introduces explicitly the isospin symmetry breaking force and describes the loosely bound nature of the wave functions of the s1 2 orbit can reproduce the observed data well and consistently explain the ob...