Dwerft - Potsdam’s Rese­arch Alliance

Dwerft: have you ever heard this term befo­re? Dwerft is the name of a recent­ly laun­ched rese­arch pro­ject in Pots­dam-Babels­berg. Dwerft does­n’t come from dwarf. That would be com­ple­te­ly inap­pro­pria­te. Rather, dwerft is a so-cal­led port­man­teau word from “digi­tal” and “shi­py­ard”. Thus, Dwerft is ope­ra­ting digi­tal boat buil­ding in the figu­ra­ti­ve sen­se. Nine mil­li­on Euros are being chan­ne­led into the pro­ject. Near­ly four mil­li­on of this comes from the Fede­ral Minis­try of Edu­ca­ti­on and Rese­arch, with the lar­gest share coming from the part­ner com­pa­nies them­sel­ves. What is at sta­ke is not­hing less than fun­da­men­tal solu­ti­ons for the film indus­try. As is gene­ral­ly known, the sec­tor is in the midd­le of a tran­si­ti­on to the digi­tal world. 

The red, green and blue in the dwerft logo repre­sent screens. They repre­sent three cen­tral are­as of the media indus­try: pro­duc­tion, dis­tri­bu­ti­on and archi­ve. The sub­tit­le “lin­ked meta­da­ta for media” shows the alliance’s visi­on: to net­work meta­da­ta in the value chain in order to then play digi­tal film out on all kinds of plat­forms. The term “media” was cho­sen becau­se moving images now also play a major role in the fields of print and radio.

The ques­ti­on that the part­ner com­pa­nies of the rese­arch pro­ject in Pots­dam are curr­ent­ly inten­si­ve­ly con­side­ring: how can digi­tal film data best be made available for use and explo­ita­ti­on? Or more con­cre­te­ly: how can this data be stored wit­hout loss and how can the meta­da­ta be seman­ti­cal­ly lin­ked within a data­ba­se? So, it is a mat­ter of lin­king pro­duc­tion, post­pro­duc­tion, dis­tri­bu­ti­on and archi­ving up tog­e­ther. Among other things, films should then be able to be trans­fer­red direct­ly to the various dis­tri­bu­ti­on plat­forms such as Net­flix or Ama­zon Prime with the neces­sa­ry data requirements.
In Pots­dam, rese­arch is curr­ent­ly being con­duc­ted in six are­as: algo­rith­ms and arti­fi­ci­al intel­li­gence for image and speech reco­gni­ti­on, block chains for bil­ling sys­tems, file-based media cloud work­flows, auto­ma­ted sub­tit­ling for cross-media dis­tri­bu­ti­on, appli­ca­ti­ons for non-fic­tion pro­duc­tions and, last but not least, intel­li­gent inter­faces to exter­nal archi­ves and edi­to­ri­al databases.

By the way, the name of dwerft isn’t that new. An initi­al pro­ject of the same name had alre­a­dy exis­ted bet­ween 2014 and 2017 with 10 part­ners coming from indus­try and sci­ence. In this first stage, the focus had been, inter alia, on rese­arch into the copy­right frame­work for the fur­ther use of cine­ma­tic works, new busi­ness models for the pro­duc­tion of online video, and on sup­port for the deve­lo­p­ment of busi­ness models for aggre­ga­ted meta­da­ta manage­ment. The pro­ject has now ente­red into its second round – now loca­ted in Pots­dam-Babels­berg - as part of the Media­Tech Hub Potsdam.
Rese­arch results are expec­ted in 2021. That means we will have to wait a litt­le lon­ger. But it is alre­a­dy clear that they will have a rele­van­ce far bey­ond the film indus­try. Anyo­ne who has to deal with lar­ge amounts of data and iso­la­ted solu­ti­ons can bene­fit from this.

dwerft con­sis­ting of:

• DRA -Stif­tung Deut­sches Rundfunkarchiv
• Film­wer­te GmbH
• Inter­la­ke Media GmbH
• IRT – Insti­tut für Rundfunktechnik
• Rotor Film Früh­mor­gen und Leh­mann OGH
• trans­fer­me­dia pro­duc­tion ser­vices GmbH

About MTH Blog

The media technologies of the future are already being used today – not only in the entertainment sector, but also in a wide variety of industries. Christine Lentz meets up with tech enthusiasts, established companies and researchers for our monthly MediaTech Hub Potsdam blog to tell the stories behind the innovative business models.