Fujitsu streamlines AI video recording recognition with high-quality compression technology
Fujitsu Laboratories Ltd says it has developed a good technology for compressing ultra-high-definition, high-volume video info to the minimum amount size necessary for AI video reputation applications. This technology can compress video info to just one single tenth the size of data prepared using conventional compression technology designed for visual confirmation by human beings.
In recent years, there's been a sharp increase in demand for AI analysis of video data in various business areas. The spread of 5th-generation mobile communications system, specifically, is expected to donate to an explosive increase in the number of ultra-high-definition video images captured by cameras, as well as many photos captured on the road and on development lines.
In developing this fresh compression technology, Fujitsu said it centered on an important divergence in the way in which AI and human beings recognize images. Specifically, AI and humans tend to differ in the regions of the image that will be emphasized as very important to judgment when recognizing persons, animals, or items in video data.
Fujitsu's technology automatically analyzes the areas that AI values and to compress info to the minimum amount size that AI may recognize. This can help you analyze a huge amount of video data without compromising recognition reliability, and at the same time significantly reduce operating and info transmission costs. Additionally it is anticipated that the technology allows users to examine more advanced video info by merging multiple video data stored in the cloud, sensor info, and performance info such as for example sales data.
Source: japantoday.com
