I've recently been working on designing a visual workflow to make it easier for the user/developer to understand the toolbox. Here is a draft for the main structure of the MgVideo object. It presents the three main objectives with display, convert and analyze respectively using ffplay, ffmpeg and ffprobe.
(You can edit the workflow using the following link)
At the moment, the toolbox systematically writes video files to disk after the preprocessing has been done. Even if the sum of all preprocessing is kept, it writes the file and delete it each time. I thought it might be cool to keep the preprocessing part loaded on memory and add three functions read, load, and write after the preprocessing part. That way the user/developer could decide what to do with the video frames of the file before writing it to disk.
I've recently been working on designing a visual workflow to make it easier for the user/developer to understand the toolbox. Here is a draft for the main structure of the
MgVideoobject. It presents the three main objectives withdisplay,convertandanalyzerespectively usingffplay,ffmpegandffprobe.At the moment, the toolbox systematically writes video files to disk after the preprocessing has been done. Even if the sum of all preprocessing is kept, it writes the file and delete it each time. I thought it might be cool to keep the preprocessing part loaded on memory and add three functions
read,load, andwriteafter the preprocessing part. That way the user/developer could decide what to do with the video frames of the file before writing it to disk.