My background is mostly in backend systems and simple UIs, so I'm new to graphics. I've done just a little AR with OpenCV. That was good, but I'm wondering if there's something else I should be using if I want the easiest path possible. I see Unity, Google, and Apple have their own offerings. Any strong recommendations?

I want to experiment with a tool that'll watch me play violin with a camera and provide visual cues on my body to correct my posture, along with an audio parser displaying a timeline of notes I've played and how in-tune they are. The theory is that the visual feedback will make the skills stick way faster. Nothing production-level, no special requirements, just looking for something easy.