Filters








5 Hits in 4.7 sec

COBE: Contextualized Object Embeddings from Narrated Instructional Video [article]

Gedas Bertasius, Lorenzo Torresani
2020 arXiv   pre-print
Instead of relying on manually-labeled data for this task, we propose a new framework for learning Contextualized OBject Embeddings (COBE) from automatically-transcribed narrations of instructional videos  ...  We leverage the semantic and compositional structure of language by training a visual detector to predict a contextualized word embedding of the object and its associated narration.  ...  Learning Contextualized Object Embeddings We now present COBE, our approach for learning contextualized object embeddings from instructional videos of HowTo100M.  ... 
arXiv:2007.07306v2 fatcat:5jue4bytinhhnel2cafk5ydpxy

AAS WorldWide Telescope: A Seamless, Cross-platform Data Visualization Engine for Astronomy Research, Education, and Democratizing Data

Philip Rosenfield, Jonathan Fay, Ronald K Gilchrist, Chenzhou Cui, A. David Weigel, Thomas Robitaille, Oderah Justin Otor, Alyssa Goodman
2018 Astrophysical Journal Supplement Series  
images, data, and stories to be viewed and shared among researchers, exhibited in science museums, projected into full-dome immersive planetariums and virtual reality headsets, and taught in classrooms from  ...  WWT's unique contextual narrative layer (Wong 2008) has set it apart from the myriad of other available data visualization systems.  ...  Clockwise from top left: View of Earth with near Earth objects imported as Two Line Element (TLE) files (data from International Astronomical Union's Minor Planet Center); 3D ISS Model (ISS Model from:  ... 
doi:10.3847/1538-4365/aab776 fatcat:4qy3d3p2wvf55byfpc3xre4cye

Unraveling egocentric actions with vision and language

Razvan-George Pasca, Xi Wang, Yen-Ling Kuo, Otmar Hilliges
2022
This work is one of the first to treat next-active object (NAO) prediction as an end-to-end deep learning multimodal prediction problem.  ...  The NAO prediction task requires localizing the object that an actor will interact with in the short-term future and classifying the type of interaction that will take place.  ...  token besides the contextualized word embeddings.  ... 
doi:10.3929/ethz-b-000551596 fatcat:ljuopn7zjzfudadckzqll62jne

Soft in the middle: the contemporary softcore feature in its contexts

2007 ChoiceReviews  
Casting Lori Cobe. Perf. Raelyn Saalman, Will Potter. Cameo (Playboy), 1995. Orion, 1996. SC. The Agony of Love. Dir., writ. William Rotsler. Prod. Harry Novak. 1966. Alley Tramp (a.k.a.  ...  Female masturbation, by contrast, seems to qualify as a contextual absence.  ...  I-I moved away from the theatrical end of things and into the export end of distribution, television, cable, and even their own home video label, Super Video.  ... 
doi:10.5860/choice.44-4949 fatcat:ejjooci7qrcfvcqfvyctsc6cqm

Panel Members: Panel Members

Wednesday, October
unpublished
Sociologist Anders Blok will engage in a dialogue with Rune Boserup from COBE architects, the designers of the new North Harbour, on how future, sustainable cities will take form, and the diverse challenges  ...  Approaching dimensions of technology in practice: Forming an ecological view from the views of teachers and nurses.  ...  Jamie Wallace, University of Aarhus The Social Practice of Flow: Design from a practice lens. Graham Dean, Lancaster University  ... 
fatcat:nmkjosflarfkpo7dwt6vmoasbe