videobox.New creates a new videobox.Client that provides helpers for accessing the services that are running locally in the Docker containers we spun up.

Return errors.Wrap(err, 'waiting for results') Results, video, err := waitForVideoboxResults(vb, video.ID) We are going to use the Machine Box Go SDK to help us make requests (but they’re just HTTP requests, so you can use curl if you like).Ĭreate a Videobox client and use it to process the source video file: // TODO: load the source file into src vb := videobox.New(' video, err := vb.Check(src, opts) See the source code for a complete picture. This blog post omits lots of boilerplate stuff, including creating temporary directories etc. The complete source code for the nevernude tool is available to browse directly. In this article we are going to look at a solution using Go, but you can use any language you like (or even a bash script) - after all, Machine Box provides simple RESTful JSON APIs that are easy to consume in any language. In a terminal, run docker-compose up which will spin up two Docker containers, one for Nudebox and one for Videobox.ĭocker compose makes it easy to spin up different Machine Box architectures A little program to process video

You’ll need to have Docker installed and the the MB_KEY variable set. The easiest way to spin up Videobox and Nudebox on your local machine is to use Docker compose (learn more on Docker’s website).Ĭreate a folder called nevernude, and insert the following docker-compose.yml file: version: ‘3’ Machine learning is never 100% accurate, so videos automatically edited using this technique should be double-checked before publication. Use ffmpeg to stitch the segments back into a new video file.

Use ffmpeg to break the original video into appropriate segments.Use that information to generate a list of time spans that therefore do not contain nudity.Send the video to Videobox+Nudebox to get a list of time spans that contain nudity.We have to do four simple steps in order to achieve our goal:

Ffmpeg filters similar image cutout