DL in the field of CV

Posted 2021-03-01 imperfect-tattoo

Stepping into the field of Computer Vision, the common problems are Image Classification and Object Detection.

Due to the features of this sepecific field, Convolution is an important concept that we should be aware of.

 

Having seen how the Edge Detection works, we come to single object detection on the image, and spread all the way to multi-feature detection.

Putting different types of layers together, we get to know the whole CNN.

Begin with the classic CNNs, we‘ll see why the special technics are proposed as well as what they are trying to solve.

 

Going back to the task of object detection, what we trying to do is put a bounding box around the target object, and detect for some information about this object. Here are some detection algorithms to make this work.

Actually, there‘re some related literatures worth reading, I‘ll try to go through some of those outstanding works and share with you in the following weeks.