Deep learning vs Traditional Computer Vision
I need a suggestion. I’m starting Masters next week and I’m deciding on courses. I have a clash between deep learning and computer vision. Do companies still use traditional CV? Or have they migrated to Deep Learning based implementations? What about companies that do AR, VR, Autonomous cars?
Deep learning by itself isn't often enough.
I vote for traditional computer vision. Deep learning you can learn by taking some online course too
Not true
It is true. I learned deep learning in my undergrad by myself. Result? Got into working at top research lab and a paper before graduation.
+1 for both. For real time computer vision applications on constrained devices (what’s hot today), it’s classical computer vision and a subset of deep learning. You NEED both to be hired in this field
Computational Geometry and State Estimation and tracking are the most fulfilling aspects of Computer Vision. If you can wrap your head around these, everything else is child's play.
What?
Take both. Computer vision offers a lot of depth into different facets.
I have a clash this semester and not sure what will happen next year when they run again. Can you give me some examples of what you mean?