Valossa secures funding for real time, AI powered video search technology

Valossa enables you to search through deep video content with your voice and natural language for better movie services and entertainment control.
Written by Eileen Brown, Contributor
Valossa secures funding for real time, AI powered video search technology ZDNet

Oulu, Finland based Start-up Valossa has released its new video search tool. The technology analyses video streams in real-time to identify more than one thousand concepts such as places and objects from any video stream.

Valossa technology understands the contents of video files themselves through a combination of natural language processing and pattern recognition AI.

It has been designed for service and content providers enabling users to can reach down into their video content, identify it and make it searchable.

You can create queries in natural language which will be understood by the Valossa Search engine.

Searching for videos by asking for 'Sean Connery in red pants' , genres such as 'non-violent princess movies' or 'romantic comedy movies involving career issues and family' will deliver the results you require.

The technology uses computer vision, machine learning and NLP (Natural Language Processing) to obtain 'Deep Content Metadata' from several modalities.

It analyses each video frame and extracts multilingual keywords from existing metadata.

The technology platform also has the capability for video stream analysis of live TV and other videos. This produces holistic video content descriptions at scene level as well as descriptive keyword annotation for video overviews.

The Valossa Smart Search can also easily identify conditional and chained voice commands which is not currently possible with Apple's Siri and other voice-powered search assistants.

This means you can ask two separate queries such as: "Find me epic history movies" followed by "only the ones with large battles".

The company is also developing new features. Users will then be able to find the most recent topical movies using queries such as "latest Sci-Fi movies involving time travel".

It integrates with voice platform vendors, such as Amazon, Apple, Google and Microsoft.

The engine is capable of finding the best matching results to elaborate and complex searches due to university research on multimedia information retrieval and AI that is now part of the company toolkit.

The 'Smart Search' is based on Valossa's patent-pending, deep content models that are constructed from various descriptive movie data sources across the Web. These models are then ranked using the company's proprietary natural query engine.

You can test the technology out yourself at Valossa's demo site which contains data on over than 45,000 English language movies. The company claims that the search accuracy exceeds on-demand movie searching by ten-fold performance.

Last month the company announced finalization of a $650k seed round with Butterfly Ventures (Oulu-based Micro VC) and several Finnish angel investors.

The company has plans for new technologies which will integrate into its platform and analyse TV transcripts for emotional sentiment. It also aims to be able to detect emotions from human faces.

The days of wooden acting might soon be over.

Editorial standards