(1), a deep learning model of computer vision.

Because of computer vision, the data we process are all pictures, so there are two main deep learning models, one is CNN convolutional neural network, and the other is Transformers such as swin transformer. The recommended learning models are as follows: Alexnet, Googlenet, Resnet,

Unet (image segmentation), CycleGAN (image style transfer), Vit (visual transformation model), swin transformer (Mar prize paper).

(2), artificial intelligence foundation

Scene understanding and analysis, pattern recognition, image search, data mining, deep learning, etc.

(3) Code

For computer vision, it is a subject that pays equal attention to theory and practice. Only in the process of writing your own code can you better understand the training process of the model.

(4) Knowledge of image processing

Image processing generally includes: optical imaging basis, color, filter, image local features, image texture, image matching and so on.

(5), related disciplines

Other disciplines related to computer vision include machine vision, digital image processing, medical imaging, photogrammetry, sensors and so on.