.footer { } Logo Logo
deutsch
/// News
Videotutorial: Who is Who in the credits: Film Credits and which job behind it

Put in the mouth: artificial intelligence renders mouth after audio file

[11:34 Fri,14.July 2017   by Rudi Schmidts]    



The SIGGRAPH 2017 already throws back its shadows, and the topic of artificial intelligence also occupies an ever-widening place within the framework of the presented research results. A good example is the GRAIL (Computer Graphics and Imaging Laboratory, University of Washington's Department of Computer Science and Engineering). PIC1 In this, the mouth movement of a video recording by Barack Obama was generated synthetically by Barack Obama in another video recording. If you do not know this before, one of the newly generated video clip is authentic. For a better understanding, here is a more detailed explanation of the project: VY = 9Yq67CjDqvw That such a technology is now available brings a lot of implications: First of all, the basic credibility of video material is destroyed. And this should indirectly also increase the mistrust of the media. The only possible solution is the digital signature of a video file. So similar to the cryptoerations and the blockchain could be a video file with hashed hash values u200bu200bas genuine sign. A subsequent manipulation would be immediately recognizable because the signature no longer fits. These keys must of course also originate from trustworthy sources. This sounds as if a completely new market (also for camera manufacturers) could arise. But also in the artistic sense the technology could be interesting. Particularly in the case of post-synchronization, a German translation could be made at the same timeCorrect lip movements. And conversely, the export of German film to the English-speaking region could be fired. Because it is often said that especially Americans are extremely allergic to synchro-responses, if the lip movements do not fail exactly. And last but not least, this technology could also help in principle, instead of O-Ton in scenic productions the voices in the post easier to use overdub. Even smaller language errors (which would otherwise make a scene equally useless) could be thereby comfortably ironed, by the lips times briefly automated. (T) yes, people, it will come so whether we want or not ... u003cimg src=


Bild zur Newsmeldung:
obama_siggraph

Link more infos at bei grail.cs.washington.edu

deutsche Version dieser Seite: In den Mund gelegt: Künstliche Intelligenz rendert Mund nach Audio-Datei

  



[nach oben]












Archiv Newsmeldungen

2025

May - April - March - February - January

2024
December - November - October - September - August - July - June - May - April - March - February - January

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000






































deutsche Version dieser Seite: In den Mund gelegt: Künstliche Intelligenz rendert Mund nach Audio-Datei



last update : 11.Mai 2025 - 15:02 - slashCAM is a project by channelunit GmbH- mail : slashcam@--antispam:7465--slashcam.de - deutsche Version