During the Cloud Next conference, Google unveiled its new machine learning technology that can do searches on elements present in videos
At the Cloud Next conference in San Francisco, Google announced the launch of a new machine learning that is useful for automatically recognizing objects in videos. This system can also be used to perform a Web search on an element seen in a particular online movie.
Let’s start from the beginning, API stands for application programming interface. This term in computer science refers to any set of procedures available to the programmer, usually grouped together to form a set of specific tools for performing a given task within a program. Google’s new Video Intelligence API will allow developers to create applications capable of extracting elements from a video. This is an important step forward since this system previously could only be used on still images and not in videos.
Machine Learning
Google has also released the codes of the beta version of its API with machine learning for all developers who are interested in developing applications about it. The company underlined how this technology will help a lot the machine learning, also because until now this system was fed almost exclusively by still images.
For those who don’t know the term, with machine learning, we mean a software program able to analyze data, often a huge amount of data, tracing within them recurrent patterns (structures) and from these automatically extract an algorithm necessary to perform a certain task.