This week at the Cloud Next conference in San Francisco, Google announced their new Cloud Video Intelligence API for videos on Google Cloud Storage being released in private beta. Google announces that their REST API can detect entities in your video, detect scene changes, specify what region processing will take place and then feed all information back into the video metadata. Google supplies 7 client libraries to support developers.
With this API, Google goes head to head with Microsoft and their Vision APIs, Amazon Rekognition (mostly images) and Clarifai. It will be interesting to follow the development of machine learning and video analytics in the coming months and years. Using Clarifai, one of our developers built a MVP where you upload a video via the SYNQ Video API. The video link is sent to Clarifai's API using our programmable webhooks, then the results are fed back into our custom metadata. You can then fire another webhook that notifies other systems of specific data added or a change in the metadata. We see many interesting use-case connected to these capabilites and are working on supplying this as a standard integration in the near future.