In the desktop application for Mendeley, it's possible to import a PDF and have the bibliographic data automatically extracted (or perhaps looked up, I'm not sure). This feature is more or less reliable. However, Mendeley's code is closed source (they once promised to release the source, but then they were bought by Elsevier). I'm looking for an open source tool that takes one or more PDFs as input and returns a bibtex entry for each.
I've found the following, but couldn't get either of them to work:
At present, the fastest alternative I know is to copy/paste the title into Google Scholar, and click the link to bibtex. That's very nice, but I'm wondering if there is something more automated.