The project has indirectly received funding from the European Commission’s Horizon 2020 Framework Programme through the STADIEM project (Grant Agreement 957321)
DeepVA is a visual data mining platform that provides everything you need to use AI in the most simple and effective way. We help you obtain training data automatically, customize your AI and then use it to analyze your images and videos. We have developed the finest API for that which works best with your technology.
Discover the full potential of your images, videos and live streams with a variety of fast and accurate recognition services. Employ state-of-the-art AI technology to adapt to a changing industry.
You only need one sample image to train your own AI model within seconds. Professional and reliable results without any expertise. It’s built for the users.
While working with AI, training data is one of the most important elements. Our dataset management offers you the best tools, such as quality feedback and a fully automated data collection process.
Want to stay informed?
Sign up for our newsletter!
Don’t worry, we reserve our newsletter for important news, so we only send a few updates once in a while.
You are in good company
Intelligently organize your images and videos and gain meaningful insights without manual effort.
Experience the entire AI process from managing and training to deploying AI for your media workflows.
Use our platform to operationalize AI and start driving your business value.
The first AI platform that will always stay GDPR-compliant, no matter how and where you use it. Your data will always be your data!
Media companies are constantly facing changes in consumer behavior. With the development of social media platforms, direct-to-consumer models and OTT services, the demands on broadcasters and media companies have never been greater. In a competitive environment with increasing cost pressure, flexible and intelligent technology solutions are therefore indispensable.
DeepVA is a promising partner for AI solutions from Europe. Through our experience in the media environment and close cooperation with well-known broadcasters, we see DeepVA software as the expert solution for qibb's current and future challenges.
We strive for the best developer experience; therefore, we provide a complete, accurate and easy to digest API documentation. See for yourself!
FLEXIBLE & TROUBLE-FREE
Your workflows don’t change, there is no need for new software. Whether you use DeepVA on-premise or in the cloud, on GPU or CPU servers or you decide for hybrid solutions, we guarantee minimum effort!
Identify more than 21.000 personalities in a variety of categories such as politics, sports, business, religion or entertainment in real-time with the highest precision!
Precisely segment your videos to skip intros and detect recaps, title sequences, contributors or breakbumpers. Find the perfect scene for your use case like ad placement!
Easily and reliably categorize and archive your content with more than 1.500 visual concepts specifically designed for media & entertainment industry.
Explicitly detect names from lower thirds. The lower thirds recognition detects the names displayed on the screen and assigns them to the appropriate person.
Analyze emotions like “happy”, “sad”, “angry”, “confused” and many more as well as several other facial characteristics and demographic information.
Identify more than 30.000 major landmarks, architectural structures and natural monuments throughout Europe and North America.
Establish gender equality and make sure your content lives up to today’s diversity standards.
Monitor your brand in ads and social media and monetize your visual content! Track logo appearance and measure sponsorship performance!
Detect and read out QR-codes in images and videos.
Recognize not only text inserts, but also any form of written information on billboards, signs, documents and other surfaces.
Identify the format of your images and videos and crop them intelligently to your requirements.
We deliver a selection of the best thumbnails for you and all you have to do is choose the one you like most. We even format them to your requirements.
With just three clicks you can add new people to your own AI model and raise the number of recognized faces from 21.000 to potentially unlimited.
Create your own AI model and identify all kinds of buildings! Whether it’s your own city, historical buildings or even your own house, everything is possible!
This state-of-the-art method of face fingerprinting allows you to analyze media data followed by assignment of unique IDs.
Completely skip the effort of providing training data and let the AI automate this process for you!
We go the extra mile and evaluate every single sample image for you and tell you whether it’s suitable for your AI model.
DeepVA provides features that make it easy for you to efficiently manage very large datasets while maintaining data privacy.