1 of 7

CSE 163

Convolutions

��Hunter Schafer

​

​

​

Questions During Class?

sli.do (Code: 163)

​

​

🎵Music: Charly Bliss

2 of 7

Convolution

When wanting to use “local” information, we commonly use a sliding window approach (i.e. a convolution)

​

Move the sliding window across the image, and compute the sum of the element wise product of the window (kernel) and image

2

3

3

2

1

0

0

0

1

3

1

3

1

2

2

3

2

0

0

2

2

2

0

0

0

1

Image

0

1

2

2

2

0

0

1

2

Kernel

3 of 7

Convolution Example

3

4 of 7

Common Kernels

What do the numbers in the kernel do?

4

Identity

Edge Detection

Sharpen

Box Blur

5 of 7

Image Classification

For a really long time, image classification was done by painstakingly crafting these features (like edge detectors), by hand.

​

This kind of worked, but we quickly hit our peak using this method.

​

Then came the buzz-word… deep learning

5

6 of 7

Image Classification

  • Is this a solved problem?
    • We get pretty decent error rates on challenges like ImageNet
  • What we can’t do
    • Sometimes can’t generalize to other real-world datasets
    • Adversarial attacks

6

7 of 7

Group Work:

Best Practices

When you first working with this group:

  • Turn on your mic / camera or introduce yourself in chat
    • We prefer mic/camera if available to encourage sense of human interaction :)
  • Share your name + where in the world you’re calling in from!
  • Elect one person to “drive” and share their screen for reference

​

Tips:

  • Starts with making sure everyone agrees to work on the same problem
  • Make sure everyone gets a chance to contribute!
  • Ask if everyone agrees and periodically ask each other questions!
  • Call TAs over for help if you need any!

7