Meta tags:

Headings (most frequently used words):

arun, mallya, selected, research, image, video, generation, facial, animation, neural, rendering, model, efficiency, interesting, properties, tutorials, workshops, writeups, notes,

Text of the page (most frequently used words):
and (14), video (12), #neural (11), arun (11), mallya (11), vision (10), with (10), computer (9), arxiv (9), models (8), image (7), the (6), network (6), from (6), project (6), page (6), animation (6), generation (6), cvpr (5), ming (5), liu (5), editing (5), research (5), code (4), tasks (4), pattern (4), recognition (4), just (4), 2021 (4), facial (4), meta (4), iccv (3), data (3), 2022 (3), training (3), weights (3), svetlana (3), lazebnik (3), single (3), conference (3), transfer (3), oral (3), rendering (3), using (3), ting (3), chun (3), wang (3), for (3), diffusion (3), receptive (2), field (2), introduction (2), rnns (2), github (2), accelerating (2), mixed (2), precision (2), eccv (2), 2020 (2), learning (2), new (2), pruning (2), 2018 (2), adding (2), multiple (2), fixed (2), hongxu (2), yin (2), jose (2), alvarez (2), jan (2), kautz (2), pavlo (2), molchanov (2), free (2), via (2), paper (2), batch (2), gradients (2), see (2), model (2), efficiency (2), properties (2), view (2), worlds (2), international (2), serge (2), belongie (2), zekun (2), hao (2), information (2), processing (2), systems (2), neurips (2), implicit (2), controllable (2), motion (2), warping (2), speech (2), technical (2), report (2), nvidia (2), 2024 (2), gen (2), website (2), scholar (2), currently (2), their (2), evaluation (2), previously (2), real (2), time (2), scientist (2), lab (2), visualization, neuron, connections, cnn, including, dilation, jupyter, notebook, find, size, effective, stride, supports, dilated, convs, illustrated, explanation, lstm, forward, backward, pass, backpropagation, refresher, hosted, edit, requests, additions, corrections, are, welcome, writeups, notes, 2019, machine, synthetic, tutorials, workshops, add, small, number, packnet, iterative, make, learn, simply, masking, existing, european, dillon, davis, piggyback, mask, reproduce, samples, trained, zhizhong, derek, hoiem, niraj, jha, dreaming, distill, knowledge, deepinversion, invert, entire, images, averaged, arash, vahdat, through, recovery, gradinversion, interesting, generating, realistic, consistent, renders, block, gancraft, unsupervised, minecraft, like, moe, but, hierarchy, position, dependent, periodic, representations, levels, experts, one, shot, talking, head, synthesis, conferencing, complex, replaced, cross, attention, layer, sets, lip, sync, input, 2023, rafael, valle, siddharth, gururani, space, driven, portrait, expression, pixel, edify, text, movie, cast, media, foundation, tweet, 15s, latency, 1080p, 10s, released, deployed, instagram, muse, all, google, selected, webpage, linkedin, focuses, generative, content, creation, language, have, dabbled, well, probing, unexpected, work, received, siggraph, live, showcase, cutting, edge, technology, demos, best, show, award, phd, advised, prof, uiuc, tech, cse, iit, kharagpur, university, illinois, urbana, champaign, senior, group, now, cosmos, part, since, its, inception, when, began, members, deep, imagination, dir, msl, working, auto,

Text of the page (random words):
arun mallya arun mallya currently a research scientist at meta gen ai msl working on video generation video editing models and model auto evaluation previously a senior research scientist in the deep imagination research dir group at nvidia now cosmos lab part of the lab since its inception when it began with just 3 members phd from the university of illinois at urbana champaign advised by prof svetlana lazebnik ms in cs from uiuc b tech in cse from iit kharagpur my research currently focuses on generative content creation and their evaluation with vision language models i have previously dabbled in neural rendering and neural network efficiency as well as probing their unexpected properties my work on facial animation received the best in show award at the 2021 siggraph real time live a showcase of cutting edge real time technology demos linkedin scholar github webpage selected research see all on google scholar image video generation muse image video generation models image and video generation models deployed in meta ai and instagram website video editing released in meta ai video editing diffusion models 15s latency for 1080p 10s video editing tweet website movie gen a cast of media foundation models technical report meta 2024 text to video video editing diffusion models arxiv project page edify image generation technical report nvidia 2024 image generation with pixel diffusion models arxiv project page facial animation space speech driven portrait animation with controllable expression siddharth gururani arun mallya ting chun wang rafael valle ming yu liu international conference on computer vision iccv 2023 facial animation with lip sync using just input speech arxiv project page implicit warping for animation with image sets arun mallya ting chun wang ming yu liu neural information processing systems neurips 2022 complex motion transfer warping replaced by a single cross attention layer arxiv project page one shot free view neural talking head synthesis for video conferencing oral ting chun wang arun mallya ming yu liu computer vision and pattern recognition cvpr 2021 3 d controllable facial animation using motion transfer from video arxiv project page neural rendering implicit neural representations with levels of experts zekun hao arun mallya serge belongie ming yu liu neural information processing systems neurips 2022 like moe but using a hierarchy of position dependent and periodic weights paper gancraft unsupervised 3d neural rendering of minecraft worlds oral zekun hao arun mallya serge belongie ming yu liu international conference on computer vision iccv 2021 generating realistic view consistent renders of 3d block worlds arxiv project page code model efficiency interesting properties see through gradients image batch recovery via gradinversion hongxu yin arun mallya arash vahdat jose alvarez pavlo molchanov jan kautz computer vision and pattern recognition cvpr 2021 invert an entire batch of training images just from averaged gradients paper dreaming to distill data free knowledge transfer via deepinversion oral hongxu yin pavlo molchanov zhizhong li jose alvarez arun mallya derek hoiem niraj jha jan kautz computer vision and pattern recognition cvpr 2020 reproduce training data samples just from a trained neural network arxiv code piggyback adding multiple tasks to a single fixed network by learning to mask arun mallya dillon davis svetlana lazebnik european conference on computer vision eccv 2018 make a neural network learn new tasks by simply masking existing fixed weights arxiv code packnet adding multiple tasks to a single network by iterative pruning arun mallya svetlana lazebnik computer vision and pattern recognition cvpr 2018 add new tasks to a neural network by pruning and training a small number of weights arxiv code tutorials workshops machine learning with synthetic data cvpr 2022 accelerating computer vision with mixed precision eccv 2020 accelerating computer vision with mixed precision iccv 2019 writeups notes hosted on github edit requests additions and corrections are welcome a backpropagation refresher an illustrated explanation of the lstm forward backward pass introduction to rnns introduction to rnns ii jupyter notebook to find receptive field size and effective stride supports dilated convs visualization of neuron connections and receptive field of a cnn including dilation