Metaliquid applies neutral net based technology to comprehend video images and creates semantic and time dependent context analysis that are meeting the needs of the video and media market industry. Specifically this form of AI is quite different from a broadcast perspective whereas we normally see voice recognition being utilized for captioning or subtitling, now with the onset of advancing AI the range of services that Metaliquid are providing dives deep in the areas of video live feeds and digital video assets. The technology has the capability to process videos and identify in real-time thousands of concepts that are detecting elements associated with the video.
Metaliquid is a computer vision and auditions deep learning solutions specifically designed for the evolving video and media market. The solution is able to recognize people, objects, landscapes, places, categories, mood and analyze the semantic relation between the images. This highly advanced application has core answer questions related to content and understands video as a human being. Some questions include: Who are the people/talents appearing in the scene? Which is the setting? What is happening in the scene? What is the nature of the interaction? What is the sentiment of each scene and what are the viewers related feelings?
With the technology core focused on agility and scalability, Metaliquid captures the neutral nets to recognize new concepts according to the clients needs. The core architecture is composed of 4 primary components. Which are:
Ontology Module: An ontology module managing the taxonomies used in the comprehension and description of the video and the related datasets used for training the neural networks
Deep Learning Processing: A deep learning module dedicated to training the neural networks on the defined datasets.
Core Module: A core processing module responsible for elaborating in real time the contents and for producing the time-dependent semantic graph containing the description of what’s happening in the video.
Services Module: A services module responsible for providing the semantic graph and sharing the data with the clients systems.
The Metaliquid services are leveraging several areas of business value content and are focused in the several areas of Video on Demand and content discovery. The are: The real time analysis of live video certainly supports broadcasters for the delivery of premium experiences to viewership. The video content asset management for the organization and search of digital video content in catalog environments are providing for data extraction and metadata fields. Video and Television advertising that is keyed to achieve the best brand positioning and provides Marketing agencies with deep data for new opportunities.
The future is certainly here when it comes to AI, and we will be seeing more and more of this new technology exploration in the areas of broadcast in the coming years. The Metaliquid’s real time monitoring and video analysis is a empowering service to the the media industry,but also in the machine automation, health and smart cities industries.
For more information on Metaliquid go here: www.meta-liquid.com/
We’re a growing team counting a dozen people. Our founders have many years of experience in the media market and in designing and developing AI solutions, some of us are digital natives. We believe technology and artificial intelligence truly have the transformative power to make everyday life easier and nicer. Every moment is an opportunity to get fresh inspirations to design our services.
Matt and his family currently reside in the Washington, D.C metro area.
Latest posts by Matt Harchick (see all)
- Metaliquid Is Making AI Video Analysis Reality - July 10, 2018
- The 54th SBE National Meeting Oct 2-3 in Danvers, MA - July 9, 2018
- Product Spotlight: Blackmagic Pocket Cinema Camera 4K - July 6, 2018