Neural Photo Editing Andrew Brock Introduction Background: VAEs - - PowerPoint PPT Presentation

neural photo editing
SMART_READER_LITE
LIVE PREVIEW

Neural Photo Editing Andrew Brock Introduction Background: VAEs - - PowerPoint PPT Presentation

Neural Photo Editing Andrew Brock Introduction Background: VAEs Background: VAEs Background: GANs Background: GANs The IAN Model Multiscale Dilated Conv Blocks Image credit: http://liangchiehchen.com/fg/deeplab_aspp.jpg Multiscale Dilated


slide-1
SLIDE 1

Neural Photo Editing

Andrew Brock

slide-2
SLIDE 2

Introduction

slide-3
SLIDE 3

Background: VAEs

slide-4
SLIDE 4

Background: VAEs

slide-5
SLIDE 5

Background: GANs

slide-6
SLIDE 6

Background: GANs

slide-7
SLIDE 7

The IAN Model

slide-8
SLIDE 8

Multiscale Dilated Conv Blocks

Image credit: http://liangchiehchen.com/fg/deeplab_aspp.jpg

slide-9
SLIDE 9

Multiscale Dilated Conv Blocks

Expressivity = range of functions a block can represent / number of parameters

slide-10
SLIDE 10

Faster Dilation through Reshapes

Image credit: Real-Time Single Image and Video Super- Resolution Using an Efcient Sub-Pixel Convolutional Neural Network, Twitter Cortex

slide-11
SLIDE 11

Orthogonality

Initializing weights with orthogonal matrices works well…so why not keep them

  • rthogonal?
slide-12
SLIDE 12

Orthogonal Regularization

Initializing weights with orthognal matrices works well…so why not keep them

  • rthogonal?
slide-13
SLIDE 13

Adding Modifcations

slide-14
SLIDE 14

Photo Editing

slide-15
SLIDE 15

Photo Editing

slide-16
SLIDE 16

Thanks!

slide-17
SLIDE 17