Module 5- Part 2- Deep computer vision, CNN and different convolution operations

Relevant playlists:

Instructor: Pedram Jahangiry

All of the slides and notebooks used in this series are available on my GitHub page, so you can follow along and experiment with the code on your own.

Lecture Outline:
0:00 Roadmap and recap. What is a convolutional layer?
5:30 CNN architecture
8:25 Weight parameters in CNN + nonlinearity
12:50 Pooling layer, size, stride and type
18:50 Putting it together! our first CNN model
25:20 ANN and CNN visualization for handwritten digits
33:30 Standard convolutional (volume base)
37:45 Depthwise convolution
40:35 Pointwise convolution
42:45 Depthwise separable convolution
49:40 Transposed convolution

[skyhappy]

Thank you so much for quality education...you are making the world a better place :)

[MaysamRadpour]

And i’m searching for a channel to super thank Mr. Jahangiri, for his dedication in sharing his knowledge.

[endoumamoru]

I have a question, where does the 3 channels go at 5:54? After convolution we only take number of filters in account but what about rgb channels?