Logo Logo
deutsch
directory schraeg
Knowledge
Codecs
Hardware
Camcorders· Cinema-Kamera· Computers· Drohnen· GPU· Kamera-Zubehör· Video-DSLR· accessories
Software
Compositing· Color correction· DV Editing
DV-Movies

HowTo
Shooting· Sound
Misc
Forschung· Reviews· Streaming
/// News
AI generates photorealistic images according to description

AI generates photorealistic images according to description

[10:48 sun,28.March 2021   by ]    

The new episode of "Two Minute Papers" introduces a very interesting new DeepLearning project that can be used in several interesting ways for creative work, as it specializes in generating high-resolution photorealistic images.





Although Open AI&s GPT-3, which is based on Transformer network algorithms and has achieved some notoriety ( here some interesting examples of its universality), was already able to generate not only meaningful texts, but also images via image GPT, but these were still quite poor in quality (unlike the texts) with a maximum resolution of 192x192 pixels.



The new technique developed at the University of Heidelberg, on the other hand, just radically improves the capabilities of the neural Transformer Network with respect to images by combining a Convolutional Network (CNN), in that images are not only seen as sequences of pixels (as by a mere Transformer Network), but are first abstracted into visually meaningful (semantic) image components by a CNN before training with it, which makes images with much higher resolutions possible.

Depth-Map
Image Generation by Depth Map


Based on the learned image material, the new algorithm can generate arbitrary matching images and 3D objects from simple depth maps as well as photorealistic landscape images from a schematic sketch (here sky, there water and mountains), conjure up detailed versions from blurred images via superresolution, generate photos of people assuming a pose given by a sketch or add the cropped half of photos.

Label-to-Image
Image generation of landscapes per sketch



Superresolution
Superresolution



Landscape
Landscapes supplemented


As always, Károly Zsolnai-Fehér also documents the astonishingly fast development of the various AI algorithms by means of comparisons - what was possible more or less poorly a year ago is now already passable and one or two projects and further developments later already almost perfectly possible.

Link more infos at bei www.youtube.com

deutsche Version dieser Seite: Neue KI generiert photorealistische Bilder auf Wunsch

  

  Vorige News lesen Nächste News lesen 
bildARRI SUP 7.0 Update Beta - ALEXA Mini LF jetzt auch mit vielen S35-Optionen bildLG 27GP850: WQHD Monitor mit 98% DCI-P3 und 180Hz


related news:1E0AnkerWork PowerConf C300 Smart HD webcam with integrated AI support 13.April 2021
Video stabilization via AI: Better than anything before? 12.April 2021
Change age, gender or facial expression of a face via AI Photo Editing 4.April 2021
Flim - the AI search engine for film stills 28.February 2021
Topaz Labs Video Enhance AI 2.0: Video enhancements via AI at a special price 28.February 2021
Colourlab Ai 1.1 brings AI color correction via Look Designer for DaVinci Resolve 27.February 2021
JPEG starts standardisation of AI image compression - JPEG AI 11.January 2021
alle Newsmeldungen zum Thema Machine Learning


[nach oben]
















Archiv Newsmeldungen

2021

April - March - February - January

2020
December - November - October - September - August - July - June - May - April - March - February - January

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000






































deutsche Version dieser Seite: Neue KI generiert photorealistische Bilder auf Wunsch



last update : 14.April 2021 - 18:02 - slashCAM is a project by channelunit GmbH- mail : slashcam@--antispam:7465--slashcam.de - deutsche Version