Live video analysis with automated reframing

Automated video reframing

Live video analysis with automated reframing

AISpotter is taking the next steps with video analysis – Our intelligent service is providing video clips and highlights automatically in traditional 16×9 format and and uses automated reframing to capture the objects and actions in mobile formats.

The way how we consume sports is changing. Digital native generation are having different demands for sports content. Content providers and broadcasting companies need to produce and deliver video content in mobile formats on top of the traditional 16×9 video format to engage with the audience. 

Floorball analysis for YLE
AISpotter is working with Finnish Broadcasting Company (YLE Sports) and providing live video analysis, from both Men’s and Women’s Floorball World Championship tournaments. Clips and highlights are automatically delivered using 16×9, 9×16 and 1×1 video formats. 

16×9 clips and highlights are automatically published through Yle’s Areena platform using the similar setup put in place for EURO2020. YLE is using the other reframed 9×16 and 1×1 format video clips for social media and for personalized story format articles they are publishing during the championship tournaments.

In addition YLE is can create visualized views from the games based on the event data our advanced data analysis is collecting. 

How do we do it?
We are receiving live video stream(s) directly from the Client (YLE).  Streams are connected to analyzing algorithm. The AI algorithm detects and identifies predefined game events e.g. goals directly from live stream. When a game event is detected video clips are created and shared with the Client (YLE).  The whole process is automated end to end. 

Benefits
The service we have developed helps streaming and broadcasting companies to increase efficiency and provides content meeting their customers’ needs for fast and automated content creation and sharing using different video formats. Our automated video analysis provides accurate content, simplifies highlight creation and sharing processes and saves both time and resources. 

Also, our innovative service can produce and deliver numerical data as a result of video analysis  – time stamps from detected events. Numerical information can be delivered and shared separately or together with the video clips using e.g. JSON file format.

In case video volumes are high or there is a need to analyze multiple simultaneous live video streams we can deploy edge computing solutions and integrate our service seamlessly to the existing infrastructure in place. Within this service setup delays in data transfer are minimized and our flexible solution makes it possible to scale needed computing capacity according to the volumes used.  

About AISpotter

AISpotter is a cloud-based video analysis and intelligent highlight production service. We use our own AI models to analyze any video, locate desired events, generate highlights, and combine them with data streams, statistics, and sensor data. 

The technology enables streaming companies and sports clubs to scale up their video production capabilities using existing resources. AISpotter can analyze large numbers of simultaneous videos saving hours of production time.

Contact me for further information: Timo Rousti

AI is Changing How You Watch Sports

AI is Changing How You Watch Sports

AI is Changing How You Watch Sports

The Age of AI Series – a podcast with Sanpram.

In this episode of the Age of AI Series Aman Y. Agarwal and Anri Kivimäki are talking about how machine learning is transforming sports broadcasting!

When you watch a live football game, the feed is usually coming from dozens of cameras all over the place, and being produced by a TV crew of hundreds of people. There’s a lot of real-time decision-making about which camera’s feed the viewer should be looking at at any given time.

AI is changing how you watch sports and it turns out, we can use machine learning to detect the most important events in the game from ALL camera feeds, and make very good video-cutting suggestions to the production team within just a few seconds. Not only does it save effort, it also improves the viewer’s experience.

AISpotter small logo

About AISpotter

AISpotter enables professionals to turn massive amounts of recorded and streamed video into usable and shareable information that helps them make better decisions, delight fans, and even predict the future.

The technology pinpoints the most important parts of any recorded and even multiple simultaneous live-streamed videos, from any camera, with any setup, and makes production-ready highlights available instantly.

AISpotter was founded in 2018 when a university vision AI project was first used to help sports teams quickly get the best information from game videos. Now the company helps production people, 24/7 live streams, coaches, and even security firms across Europe to process video into usable knowledge and insights.

Contact me for further information: Anri Kivimäki

Great insights from football games

Data analysis for football

How to get interesting insights from football games?

Looking to analyze individual players? Our data analysis tells you how the individual players are performing in the game. 

Imagine that you want to get detailed player data from each game, and you want to share it with your network. Your first thoughts might be to start by going through huge search volumes. Data analysis gives you the details how the players are performing in the games.

Starting from EURO2020 tournament AISpotter has successfully collaborated with the leading Finnish newspaper Helsingin Sanomat (HS). We are producing comprehensive data analysis from all the team Finland “Huuhkajat” football games.  Collaboration is a joint initiative together with MyCoazh

How the process is running?
The process is flowing the following way: Game event data is collected live during the games. Raw data contains all the actions the Finnish players are having with the ball. Also the key actions (e.g. long runs) without the ball are included in. All the information is stored to a database and automated, predefined queries are used to process the data and to put together all the needed parameters and datasets.

The newspaper is using an API to pull the classified data out from our database and then they are processing with data visualization and the content is ready for publishing shortly after the game.

Data visualization
Helsingin Sanomat (c) HS

What is content is visualized?
It is easy to see from the visualization who were the individual Finnish players who succeeded in each game. Also, there are shots on goal, shot end sequences and passing maps included in.

Take a detailed look what insights data analysis found from the previous games.

EURO2020:
12.6.2021 Denmark – Finland (0 – 0)
16.6.2021 Russia – Finland (1 – 0)
21.6.2021 Finland – Belgium (0 – 2)

Friendly game:
1.9.2021 Finland – Wales (0 – 0)

World Cup qualifiers:
4.9.2021 Finland – Kazakhstan (1 – 0)
7.9.2021 France – Finland (2 – 0)
9.10.2021 Finland – Ukraine (1 – 2)
12.10.2021 Kazakhstan – Finland (0 – 2)
13.11.2021 Bosnia and Herzegovina – Finland (1 – 3)
16.11.2021 Finland – France (0 -2)

Summary:
Data analysis gives you detailed content from each player in fast and accurate way. Even more, data can be customized and visualized in many ways. Advanced data analysis can justify player performance over multiple games.

About AISpotter

AISpotter makes it possible to turn massive amounts of recorded and streamed video into usable and shareable information that helps to make better decisions, delight fans, and even predict the future.

Our technology pinpoints the most important parts of any recorded and even multiple simultaneous live-streamed videos, from any camera, with any setup, and makes production-ready highlights available instantly.

AISpotter was founded in 2018 when a university vision AI project was first used to help sports teams quickly get the best information from game videos. Now the company helps production people, 24/7 live streams, coaches, and even security firms across Europe to process video into usable knowledge and insights.

Contact me for further information: Timo Rousti

Future of Live Video Analytics

The Future of Live Stream

The Future of Live Stream Video Recording and Analytics with AISpotter

A podcast with Expert Dojo.

This week, host Brian Mac Mahon speaks with Anri Kivimäki, Founder and CEO of AISpotter.

In this episode, Brian and Anri discuss what is computer vision and how it helps with the future of machine learning, the amazing technology that AISpotter has created, and how this technology can be used in various businesses from sports to security, and so much more. Lastly, we hear how they plan to achieve $100k/month in revenue.

AISpotter small logo

About AISpotter

AISpotter enables professionals to turn massive amounts of recorded and streamed video into usable and shareable information that helps them make better decisions, delight fans, and even predict the future.

The technology pinpoints the most important parts of any recorded and even multiple simultaneous live-streamed videos, from any camera, with any setup, and makes production-ready highlights available instantly.

AISpotter was founded in 2018 when a university vision AI project was first used to help sports teams quickly get the best information from game videos. Now the company helps production people, 24/7 live streams, coaches, and even security firms across Europe to process video into usable knowledge and insights.

Contact me for further information: Anri Kivimäki