Ecole d'ingénieur et centre de recherche en Sciences du numérique

From raw data to semantically enriched hyperlinking: Recent advances in the LinkedTV analysis workflow

Stein, D; Öktem, A; Apostolidis, E; Mezaris, V; Redondo García, José Luis; Troncy, Raphaël; Sahuguet, Mathilde; Huet, Benoit

NEM Summit 2013, Networked & Electronic Media, 28-30 October 2013, Nantes, France

Enriching linear videos by offering continuative and related information via, e.g., audio streams, web pages, as well as other videos, is typically hampered by its demand for massive editorial work. While a large number of analysis techniques that extract knowledge automatically from video content exists, their produced raw data are typically not of interest to the end user. In this paper, we review our analysis efforts as defined within the LinkedTV project and present the recent advances in core technologies for automatic speech recognition and object-redetection. Furthermore, we introduce our approach for an automatically generated localized person identification database. Finally, the processing of the raw data into a linked resource available in a web compliant format is described.

Document Bibtex

Titre:From raw data to semantically enriched hyperlinking: Recent advances in the LinkedTV analysis workflow
Mots Clés:Automatic Speech Recognition, Object Redetection, Person Identification, NERD Ontology
Type:Conférence
Langue:English
Ville:Nantes
Pays:FRANCE
Date:
Département:Data Science
Eurecom ref:4153
Bibtex: @inproceedings{EURECOM+4153, year = {2013}, title = {{F}rom raw data to semantically enriched hyperlinking: {R}ecent advances in the {L}inked{TV} analysis workflow}, author = {{S}tein, {D} and {\"{O}}ktem, {A} and {A}postolidis, {E} and {M}ezaris, {V} and {R}edondo {G}arc{\'i}a, {J}os{\'e} {L}uis and {T}roncy, {R}apha{\"e}l and {S}ahuguet, {M}athilde and {H}uet, {B}enoit}, booktitle = {{NEM} {S}ummit 2013, {N}etworked \& {E}lectronic {M}edia, 28-30 {O}ctober 2013, {N}antes, {F}rance}, address = {{N}antes, {FRANCE}}, month = {10}, url = {http://www.eurecom.fr/publication/4153} }
Voir aussi: