DeepVA 2018-02-12T15:34:52+00:00
By using DeepVA it is possible to analyze the video and image stock of your company automatically, fast and in an intelligent way. Based on the latest research approaches of Deep-Learning (an area of artificial intelligence) DeepVA enables you to perform a profoundly and broad examination of your company’s data by extracting hidden meta-information from videos & images and making it accessible to your company.
The new, intelligent video and image analysis from’ The Chainless’ offers many possibilities for the data acquisition and archiving of a TV station’s archive. It is therefore a real problem solver for the TV industry. The present prototype (with a novel AI) has already convinced me to use this software in the future.”
Jürgen Flettner, Chief Journalist - Content (SWR)



Become a partner of this innovative progress and gain advantage today from the benefits of our video & image analysis. As a young software enterprise from Germany, we are looking for selected companies in the media and advertising industry to further expand and establish our product.
write us
Member of the Inception Program


Download datasheet




in overview


The facial recognition module identifies faces of people. You can use this module to recognize famous individuals such as athletes, politicians and celebrities.


The object recognition module detects all objects in an image or video frame. E.g. objects, buildings and animals.


The text recognition module (OCR – Optical Character Recognition) extracts text passages from actual images


This module is used to determine captions and corner graphics in TV reports and news magazines. Use this module for an automated identification of unknown but important people. We give faces a name.


The emotion recognition module categorizes certain visual attributes of a face and assigns fitting expressions such as pleasure, anger, agony, disgust and surprise to them.


The demographic model analyzes said facial features and adds information on age, gender and ethnicity for each specific face identified.

Technical Analysis

The module for quality analysis determines technical characteristics of an image. It not only reads the codec’s meta data, but also additional visual quality features such as noise, compression-related artifacts, ??), moiré effects, focus areas and many other components.


The color analysis module ascertains the dominant colors of an image. Use all prevailing color schemes for further processing and use of your data.

Brands & Logos

The trademark recognition module recognizes logos of renowned trademarks. Use this module to assign product placements in videos and images.

The solution for enterprises

Broadcaster, Advertiser & Publisher


Cost Reduction

With DeepVA you can save up to 56% of your expenses and reduce your costs permanently.

Real-Time Analysis

DeepVA accelerates the indexing of video and image data by a factor of 8. Effective parallelization and optimized GPU calculations make real-time analysis of your data possible.

Increase in Sales

DeepVA gives you valuable META-information to help you increase your sales in the long term.

Become part of the revolution

Become a partner of this inventive project and benefit today from all mentioned advantages of our video analysis. As a young software company, we are looking for selected businesses and corporations in the media and advertising industry to further develop our product.
Supported by programs of the Federal Government and the EU, we can offer your company exclusive and advantageous conditions to DeepVA as of now.
We look forward to seeing you.
write us
Congratulations on this unique software. DeepVA is an auspicious and promising solution for all TV and film professionals who are looking for quick and credible image information. The young entrepreneurs of this start-up demonstrate the potential of technological innovations in Germany. I wish you continued success!
Dr. Klaus Weidmann, TV presenter, foreign correspondent and film author (ARD, SWR, PHOENIX)

You might have heard from us through the following

press texts