Meta tags:
description=;

Headings (most frequently used words):

tutorial, and, cvpr, 2019, on, action, classification, video, modelling, program, organizers, description, of, the, its, relevance, topics,

Text of the page (most frequently used words):
the (32), and (31), video (25), #action (16), models (16), classification (12), for (12), 2018 (12), learning (10), are (9), 2017 (8), videos (7), spatiotemporal (7), feichtenhofer (7), deep (7), carreira (6), supervised (6), link (6), not (6), torresani (5), long (5), networks (5), 2019 (5), discuss (5), tran (5), representations (5), tutorial (5), lorenzo (4), gavves (4), self (4), time (4), recognition (4), 2014 (4), 2016 (4), ghodrati (4), been (4), representation (4), always (4), large (4), training (4), well (4), end (4), fernando (3), joao (3), efstratios (3), evaluation (3), detection (3), architectures (3), including (3), bilen (3), modeling (3), while (3), computer (3), vision (3), have (3), unsupervised (3), has (3), however (3), this (3), does (3), classifiers (3), benchmarks (3), how (3), topics (3), research (3), scientist (3), analysis (3), with (3), christoph (2), organizers (2), temporal (2), using (2), machine (2), datasets (2), convolutions (2), program (2), two (2), stream (2), simonyan (2), dynamic (2), various (2), 2015 (2), image (2), designed (2), their (2), kalchbrenner (2), tracking (2), still (2), domain (2), even (2), target (2), dataset (2), more (2), results (2), manner (2), one (2), short (2), processing (2), that (2), specifically (2), hussein (2), careful (2), clear (2), capacity (2), evaluating (2), professor (2), facebook (2), fair (2), what (2), state (2), art (2), train (2), strong (2), when (2), shallow (2), relevant (2), motion (2), model (2), cvpr (2), 2013, all, rights, reserved, shapebootstrap, basura, feichtenohofer, multimodal, correspondence, axis, untrimmed, break, link2, link1, revisiting, speaker, event, date, june, optical, flow, latest, modern, c3d, i3d, fusion, images, analyze, options, tsn, wang, factorized, timealigned, densenets, beyond, community, primarily, applicability, extends, generative, compression, visualization, velocity, estimation, kampelmuhler, tao, object, future, prediction, analogously, customary, fine, tune, pretrained, optimal, due, gaps, between, source, cite, shkodrani, because, unconventional, importantly, certainly, highly, accurate, take, advantage, plethora, unlabelled, available, alternatives, either, arrow, wei, audio, synchronization, odd, out, term, understanding, majority, systems, focus, rather, sequences, typically, than, seconds, applications, often, require, much, longer, streaming, handle, capture, intricacies, involved, like, timeception, videograph, scale, crucial, developing, next, generation, interestingly, current, measure, accuracy, evaluate, sequence, itself, frameworks, directly, focuses, following, organized, experts, assistant, university, amsterdam, netherlands, deepmind, star, singapore, senior, researcher, associate, dartmouth, college, usa, them, constitute, treated, equally, system, labels, enough, challenges, aims, giving, answers, aforementioned, questions, core, explored, late, years, great, force, change, most, tasks, problems, such, progress, arguably, slower, remaining, surprisingly, competitive, recent, developments, demonstrated, design, annotated, finally, led, details, secrets, achieve, good, accuracies, transparent, furthermore, whether, resulting, from, truly, providing, better, instead, simply, overfitting, idiosyncrasies, each, description, its, relevance, together, pattern, modelling, about, home, brave, new, ideas,

Text of the page (random words):
brave new ideas in motion representations home about program organizers cvpr 2019 tutorial on action classification and video modelling together with the computer vision and pattern recognition cvpr 2019 description of the tutorial and its relevance in the late years deep learning has been a great force of change on most computer vision and machine learning tasks in video analysis problems such as action recognition and detection motion analysis and tracking the progress has arguably been slower with shallow models remaining surprisingly competitive recent developments tran et al 2014 feichtenhofer et al 2016 carreira et al 2017 tran et al 2015 bilen et al 2017 ghodrati et al 2018 feichtenhofer et al 2018 simonyan et al 2014 kalchbrenner et al 2017 have demonstrated that careful model design and end to end model training as well as large and well annotated datasets have finally led to strong results using deep architectures for video analysis however the details and the secrets to achieve good accuracies with deep models are not always transparent furthermore it is not always clear whether the networks resulting from end to end training are truly providing better video models or if instead are simply overfitting their large capacity to the idiosyncrasies of each dataset this tutorial aims at giving answers to the aforementioned questions specifically the core topics to be explored are what are the state of the art video representations and action classification models and how does one train them what does constitute a strong video representation are short and long videos to be treated equally when training action classifiers are shallow models still relevant for state of the art classification how to train an action classification system in an unsupervised manner when supervised labels are not enough relevant benchmarks and challenges this tutorial is organized by experts on action classification and video representation learning a dr e gavves assistant professor in the university of amsterdam the netherlands b j carreira research scientist in deepmind uk c dr b fernando research scientist a star singapore d dr c feichtenhofer senior researcher in facebook fair us and e dr l torresani associate professor at dartmouth college and research scientist at facebook fair usa topics the tutorial focuses on the following topics deep learning for action classification and optical flow discuss the latest modern deep networks for action classification including c3d tran et al 2014 i3d carreira et al 2017 two stream models simonyan et al 2014 two stream fusion feichtenhofer et al 2016 dynamic images bilen et al 2016 deep networks for video modeling discuss and analyze various options for modeling videos including tsn wang et al 2018 spatiotemporal tran et al 2015 and factorized spatiotemporal convolutions tran et al 2018 timealigned densenets ghodrati et al 2018 dynamic image networks bilen et al 2017 deep spatiotemporal models beyond classification while in the computer vision community video models have been designed primarily for action classification their applicability extends to video generative models kalchbrenner et al 2017 video compression wu et al 2018 visualization feichtenhofer et al 2018 velocity estimation kampelmuhler et al 2018 tracking tao et al 2016 and spatiotemporal object detection feichtenhofer et al 2017 and future video prediction ghodrati et al 2018 unsupervised video representation learning analogously to the still image domain even for video it has been customary to fine tune pretrained video models on the target dataset however this is not always optimal due to large gaps between the source and the target domain cite shkodrani et al 2018 or because of unconventional architectures ghodrati et al 2018 more importantly while supervised learning certainly results in highly accurate models it does not take advantage of the plethora of unlabelled video available we discuss alternatives on training video representations models either in an unsupervised or self supervised manner including arrow of time wei et al 2018 audio video synchronization or odd one out models fernando et al 2017 long term video understanding the majority of action classification and video representation systems focus on rather short video sequences typically no more than 10 seconds long however applications often require processing much longer videos or even streaming videos we discuss models that are specifically designed to handle long videos and capture the spatiotemporal intricacies involved like timeception hussein et al 2019 and videograph hussein et al 2019 large scale video processing and evaluation careful evaluation of action classifiers and video representations is crucial for developing the next generation of models interestingly while current benchmarks do measure well the accuracy of action classifiers it is not always clear how to evaluate the capacity of temporal models in modeling the sequence itself we discuss various benchmarks and frameworks for evaluating action classification as well as for evaluating directly video representations program date june 16 2019 time event speaker 13 00 13 25 revisiting spatiotemporal convolutions for action recognition link lorenzo torresani 13 25 14 10 action classification and detection architectures link christoph feichtenhofer 14 10 14 40 3d spatiotemporal networks datasets and evaluation link joao carreira 14 40 15 15 the machine learning of time in long videos link1 link2 efstratios gavves 15 15 15 35 break 15 35 15 45 action recognition in untrimmed videos link lorenzo torresani 15 45 16 00 self supervised learning using the time axis link lorenzo torresani 16 00 16 15 self supervised learning of temporal correspondence joao carreira 16 15 16 30 self supervised and multimodal video learning link efstratios gavves organizers efstratios gavves lorenzo torresani christoph feichtenohofer joao carreira basura fernando 2013 shapebootstrap all rights reserved