[CrashCourse CS] #35 Computer Vision

Goal of computer vision is to give computers the ability to extract high-level understanding from digital images and video.

• Color marker tracking and similar algorithms are rarely used, unless environment can be tightly controlled.
• Computer vision algorithms have to consider small regions of pixels, called patches.
• Calcalating the difference between the pixels uses matchmatical notation kernel/filter.
• Applying kernel to a patch of pixels is called convolution.
• Prewitt operators - image enhancing kernels
• Viola-Johns face detection
• Convolutional neural networks
  • Convolutional neural networks aren't required to be many layers deep, but they usually are, in order to recognize complex objects and scenes, That's why the technique is considered deep learning.
  • Can be applied to many image recognition problems like recognizing handwriting, spotting tumors in CT scans, and monitoring traffic flow on roads.
• By recognizing the pattern of the face it can read emotions of a face. - conctext sensative: aware of the surroundings.
• self-driving car, superimposing images on pictures.

Vocabulary

• fidelity - 정확도

Thoughts

이제 5화정도 남았다. PBS CS 시리즈는 정말 재미있다. 끝으로 갈수록 최신 기술에 관한 영상이 나온다.