PARTNERSHIP / VIDISPINE

Vidispine Arvato Systems Interface with AI

About Vidispine

Vidispine provides media companies and broad­casters with innov­ative technologies to support their business needs. Customers benefit from a broad portfolio of media management tools, appli­ca­tions, and services, including solutions for mdia asset management, production asset management, and media supply chain. 

DeepVA product used

Deep Media Analyzer
Deep Model Customizer
Deep Indexer
Deep Collector

Benefits

  • Automatic indexing of faces and easy search for people in videos
  • Faces are automat­i­cally finger­printed and stored in the DeepVA data model, so they can be used in all subse­quent analyses
  • Ability to manage and categorize faces using the “Found By Face Analysis” collection sorting feature

Links

DeepVA Face Training Theme for VidiCore

The DeepVA Face Training Theme for VidiCore is an appli­cation for visual­izing the capabil­ities of the DeepVA Face Recog­nition & Analysis Services in VidiNet. The appli­cation includes features for analyzing video content with DeepVA AI services in VidiNet, managing and organizing the face data model/database with VidiCore, training the data model with known faces before analyzing content, manually capturing faces in videos, and more. The powerful indexing services in VidiCore make it easy to find and display people and faces using time-coded metadata combined with AI. At its core, the Face Training Theme is a light­weight appli­cation with minimal config­u­ration required. The only prereq­uisite is a live instance of VidiCore with an S3 storage, a connected VidiCoder, and one or more connected DeepVA services. The DeepVA Services in VidiNet The “Face Training and Analysis Powered by DeepVA” service will enable the powerful AI capabil­ities of DeepVA for analyzing video content to find faces both unknown and known to the system. Each finding will be finger­printed and compared, with dupli­cates being grouped into one finger­print. All findings will preserve the timecode where the face was found, in order to properly display search results in the user interface. This service also offers the ability to pre-train a face model before analyzing large amounts of content. The “Face and Label Extractor Powered by DeepVA”-service will do a thorough frame-by-frame analysis of videos to find faces and match this with any textual infor­mation found in the lower third area of the video. Analysis results from this service will NOT preserve the timecode from where the face was found. The output, however, can be used to pre-train the face data model for usage with the other DeepVA service, i.e. “Face Training and Analysis by DeepVA”. The output of the “Face and Label Extractor powered by DeepVA”-service is, simply put, a set of still images of a person’s face alongside a suggested name of the person based on the findings in the lower thirds.

Vidispine Media Portal

Vidispine Media­Portal is a web interface for accessing all media and non-media files managed by Vidispine, and provides function­ality for collecting and processing this content within a single search logic and intuitive design. It simplifies your workflows for uploading, house­keeping, editing projects, and distrib­uting content. As a highly integrated part of the Vidispine product portfolio, you can easily customize the appli­cation to your specific needs when it comes to production workflows, media supply chains, and collab­o­rative work via a conve­nient config­u­ration interface.

VidiCore

VidiCore is a Media Management backend service that sees its placement in the base of layer a media supply system. As implied, at the core of your system, its essential role is to handle, and ultimately reduce, the complexity of connecting multiple sources as well as the harvesting and management of media assets and their respective metadata. Coupled together with VidiCoder, VidiCore is able to provide media awareness and proxy gener­ation right out of the box. Another aspect which grants VidiCore a high degree of flexi­bility is the fact that it is a REST API service. This enables one to easily develop custom solutions on top of it. VidiCore malleable nature allows one to fit it into various appli­cation scenarios: From filling a gap in your existing media workflow, to building sophis­ti­cated media management solutions. VidiCore’s archi­tecture is kept simple. It can easily be deployed on an on-prem, or in cloud environment as well as utilizing vendor specify compo­nents for easy operation and scala­bility.
Vidispine Interface
Vidispine powered by DeepVa_1
Vidispine powered by DeepVa_2
Vidispine powered by DeepVa_3

Benefits for the user

Automatic face finger­printing

Detected faces are automat­i­cally stored in the DeepVA data model for subse­quent analysis

Efficient search function­ality

Easily locate individuals in your media assets

Face Management

The software allows you to effec­tively manage and categorize the faces detected by the DeepVA Face Analysis and Training service

Deep media analyzer functions

DeepVA features integrated into Vidispine

Face recog­nition

Face Recog­nition detects and identifies the faces of public figures in a variety of categories such as politics, sports, business, and enter­tainment.

Learn more

Face attributes

Face Attributes recog­nizes emotions and facial charac­ter­istics such as “beard”, “eyes closed” or “glasses” of all persons appearing in pictures or videos.

Learn more

Landmark recog­nition

Landmark Recog­nition identifies all important sights, archi­tec­tural struc­tures and natural monuments across Europe and North America.

Learn more
DeepVA Face Training Function

Face Training

In just a few seconds, new faces are trained using our face training technology and are ready for use in Recog­nition Services.

Learn more

Speech-to-Text

Often the spoken word alone is not enough: transcription is needed. Our Speech to Text function automates this process.

Learn more

Face Index

Using Face Index, each person in video and image material can be assigned a number, allowing them to be trans­lated into the metadata after­wards.

Face Dataset creation

Using images of the faces and text infor­mation (e.g., from belly bands), Face Dataset Creation automat­i­cally creates entire datasets, without any manual labor.

Learn more
Fabian Linder

Contact us

Do you have any questions?

Enter your email below, and we will get back to you as soon as possible.

Get the latest AI news delivered to your inbox

Don’t worry, we reserve our newsletter for important news, so we only send a few updates once in a while.

  • News on Artificial Intelligence
  • Use cases and success stories
  • Events

Designed and engineered in Germany.

aiconix GmbH
Stollwerckstraße 17-19
51149 Köln
Germany

aiconix
Product
Functions
Resources
Stadiem project

The project has indirectly received funding from the European Commission’s Horizon 2020 Framework Programme through the STADIEM project (Grant Agreement 957321)

Newsletter

latest AI news

Subscribe to our newsletter

Don’t worry, we reserve our newsletter for important news, so we only send a few updates once in a while. No spam!