This is a great time to get a PhD to work on problems in machine learning and machine perception, especially if you want to go into industry. There is a lot of

This is a great time to get a PhD to work on problems in machine learning and machine perception, especially if you want to go into industry. There is a lot of demand. I finished my PhD about two years ago, and I do R&D at NASA in machine learning and computer vision (although I'm leaving to become a professor). Once you are in a PhD program, be upfront with your advisor about this goal and you will likely have an easier time graduating than those students that claim they want to become professors. Typically, to get a PhD you just need to finish your coursework (fairly easy) and produce about three published papers (can be brutally hard or fairly straightforward depending on a ton of factors). The latter's ease also depends on your tolerance for disappointment, because the rewards and successes in research are somewhat sparse. A lot of the problems people have with the PhD is that there aren't enough academic jobs. You aren't seeking one, so you will be fine as long as you focus on your goal and maintain a good work-life balance most of the time throughout school (this can be very challenging to do, in practice).