Sep 3, 2018 · 1 min read
This is so very true. Realized this when I was asked to implement the state of the art Facial Recognition system during my last internship. The model that we were trying to implement worked fabulously in it’s own dataset but when it came to real world production of this model there must be so much rigorous testing before installing/integrating it with any system.
It soon made me realize that “Research” and “Real World Applications” are as different as chalk and cheese. Research on one hand focuses more on different novel approaches towards solving a problem, and Real World Application encourages to lay more emphasis on the robustness of the system.
