Best practice to track user storage usage on amazon s3

I'm building an web app that lets user upload files. And I want to limit the single file size to say < 100 MB, and track the user's usage to a certain amount (pretty much like dropbox). They won't be able to upload if that reach their limit, or they can pay more to higher the limit.

Is it a good practice to let user upload directly to AWS S3 (or perhaps google cloud storage if any difference...)? I saw people say that direct upload significantly lower server loading, but won't be able to track user usage?

So is there any ways or work around to track user usage if I let user upload directly to S3? If not, how can I stream or redirect the file to upload from the user to the web server, then to S3, and keep track of each user's usage?