What is the best way to learn augmented reality? – alltolearn.com
Well, that depends on what you want to do with augmented reality.
If you want to make an app for it, look into existing hardware prototypes like Meta and Hololens, and buy what you need (i.e., Meta is better for arms-length work, Hololens is better for viewing things far away; both have pros and cons in terms of field-of-view and registration/tracking; look into it).
Also, look over the various APIs to see whether they’re appropriate for your app.
After you’ve received the prototype, put it up and start coding/learning the API thoroughly.
Look into research publications on places like Google Scholar, the Web of Knowledge, and ResearchGate (you can also find a medley of useful information and tutorials with plain-old Google searches and on Wikipedia). If you’re more interested in the research aspects behind the headsets, developing better headsets or technologies that are useful for a broad spectrum of AR applications, look into research publications on places like Google Scholar, Web of Knowledge, and ResearchGate (you can also find a medley of useful information/tutorials).
There are a number of possible research topics (which almost always have some overlap).
These can be broken down even further into subtopics.
Hardware optics and rendering: vergence-accommodation conflict, latency-related issues, pupil/eye-tracking for AR/VR, foveated rendering, varifocal and multifocal/light-field displays, and so on.
-Tracking/ego-motion/registration: approaches based on vision and infrared, IMU-based methods, hybrid methods, filtering, and so on.
3D reconstruction includes tracking and ego-motion but is much more focused on SLAM and reconstructing the geometry, moving geometry, concurrent segmentation to aid reconstruction, kinematic structure of the geometry, material characteristics, scene lighting, and so on. –
Networking: for AR/VR applications, distributed content/sharing platforms, coordination between headsets and other wearable computing devices, and so on. –
Semantics/higher-level computer vision challenges, such as object segmentation, detection/recognition, tracking, classification, and logic, as they apply to AR applications –
Augmented reality interface and applications: haptics, higher-level interaction and GUI based on computer vision inputs, teaching, training, and simulation
All you have to do now is figure out what interests you the most and start learning about it. I hope this information is useful!
This content was originally published here.