Computer Vision cross-level quality metrics in practice

Computer vision is a very popular field in research and competitions. It is one of the most popular themes for papers. It includes a huge amount of mathematics, statistics etc. But what about the real world? The big gap exists between the mathematician world and enterprise applications.

How to reduce this gap? How to convert paper with delicious ideas and calculations to a real-worked application? How to check it?

Metrics is a good tool to control your model. But this approach is key for application tool.

In this workshop, the relationship between application and model metrics will be observed. Also, we will look at process and pipeline of system building, pitfalls and features of this approach.