Course Study with CSC321
[2021.02.27]
Topic
: Lecture 12: Image Classification (Chanju) , CNN misc part (이동재)
Notes
:
- https://drive.google.com/file/d/1GYaiD1fUVTdfW4P82oZz8oK5EJTm8xgX/view?usp=sharing (Chanju)
- https://drive.google.com/file/d/1npGyII1IWSrQQN_LGjtjK7RtgibJFvX-/view?usp=sharing (이동재)
Links
:
- https://www.cs.toronto.edu/~rgrosse/courses/csc321_2017/slides/lec12.pdf (Lecture 12 slide)
- https://yjjo.tistory.com/8 (start of CNN , concept of receptive field , how to calculate Conv. net size)
- https://bskyvision.com/421 (all about AlexNet : info-share , ReLU , local response normalization)
- https://bskyvision.com/422 (what is top-5 , top-1 error ?)
Next
: 2021.03.06 9:00 PM KST
- https://m.blog.naver.com/laonple/220710707354 (라온피플)
- Special Lecture : Inception , VGG Net , Res Net (이동재 , Chloe)
Google Net (이동재)
- 1X1 Convolution
- Inception module (different kernel size , naive & advanced)
- global average pooling
- auxilary classifier
VGG Net , Res Net (Chloe)
- important features of vgg net ( diff between 16 & 19 ?)
- concept of skip connection from Res Net
- R-CNN , Fast R-CNN , Faster R-CNN 은 이번 파트 X
***