CVPR'20 iMLCV tutorial: Exploring and Exploiting Interpretable Semantics in GANs by Bolei Zhou

preview_player
Показать описание
Bolei Zhou: Exploring and Exploiting Interpretable Semantics in GANs.
  1. Bolei Zhou
Рекомендации по теме
CVPR'20 iMLCV tutorial: 0:46:47

CVPR'20 iMLCV tutorial: Exploring and Exploiting Interpretable Semantics in GANs by Bolei Zhou

CVPR'20 iMLCV tutorial: 0:40:53

CVPR'20 iMLCV tutorial: Understanding Deep Neural Networks by Ruth C. Fong

CVPR'20 iMLCV tutorial: 0:44:54

CVPR'20 iMLCV tutorial: Introduction to Circuits in CNNs by Chris Olah

CVPR'20 iMLCV tutorial: 0:31:05

CVPR'20 iMLCV tutorial: Modeling Conceptual Understanding in Image Reference Games by Zeynep Ak...

CVPR'20 Tutorial on 0:03:51

CVPR'20 Tutorial on Interpretable Machine Learning: Opening Remark

CVPR'20 tutorial: AutoML 0:39:35

CVPR'20 tutorial: AutoML for TinyML with Once-for-All Network

Disentangled Image Generation 0:04:58

Disentangled Image Generation Through Structured Noise Injection

CycleISP: Real Image 0:05:00

CycleISP: Real Image Restoration via Improved Data Synthesis (CVPR 2020 -- Oral)

[CVPR'20 Workshop on 1:15:18

[CVPR'20 Workshop on Scalability in Autonomous Driving] Waymo Open Dataset Challenge

A U-Net Based 0:08:52

A U-Net Based Discriminator for Generative Adversarial Networks (CVPR 2020)

[CVPR 2020] GAN 0:06:11

[CVPR 2020] GAN Compression: Efficient Architectures for Interactive Conditional GANs

CVPR 2020 Recap 0:56:47

CVPR 2020 Recap

[CVPR 2020] RANSAC 0:43:14

[CVPR 2020] RANSAC in 2020 tutorial: Introduction

1282 - Temporal 0:05:01

1282 - Temporal GAN for Large Scale Video Generation

GCPR 2020: Day 0:31:39

GCPR 2020: Day 1 - Sabine Suesstrunk - Editing in Style: Uncovering the Local Semantics of GANs

MineGAN: Effective Knowledge 0:01:01

MineGAN: Effective Knowledge Transfer From GANs to Target Domains With Few Images

CVPR 2020: Efficient 0:05:00

CVPR 2020: Efficient Derivative Computation for Cumulative B-Splines on Lie Groups

CVPR 2020 Invited 0:29:08

CVPR 2020 Invited speaker: Racheal Jack

Deep Implicit Volume 0:14:26

Deep Implicit Volume Compression

Bird Size Vector 0:00:05

Bird Size Vector (discovered using GANspace)

Object Detection as 0:24:55

Object Detection as a Machine Learning Problem - Ross Girshick

CVPR 2018 Spotlight: 0:04:03

CVPR 2018 Spotlight: Disentangled Person Image Generation

1309 - FACEGAN: 0:04:59

1309 - FACEGAN: Facial Attribute Controllable rEenactment GAN

Image Processing Using 0:01:00

Image Processing Using Multi-Code GAN Prior

Комментарии
Автор

Very good work in GAN! This space is still rich for innovations.

drpchankh
Автор

I actually cried, seeing the Bedroom to Living Room object change(HiGAN) - keeping the layout constant! I think those are some tears out of excitement! Everything in this world is some distribution. Great work Professor, I have read many research papers recently, but your papers are the most simplified to the point type. I won't miss your work anymore, Huge Fan :)

harshamusunuri
Автор

Great video! I've been trying to understand GAN inversion for awhile now and this really made it click.

arcosin
Автор

Thank you for presenting it! I have some question. Have the smooth intermediate conversion images shown in the video(such as bedroom images) been seen by generator or learned by generator? If you control the layout of the bedroom, for example, do you need a large number of multi-view datasets?

txma
Автор

Great lecture! Can I ask if the latent space is similar to the semantic hashing code? If not, can we also use semantic hashing to do the synthesis?

yapingsun
Автор

Hello and thank you for the very insightful journey into GANs! You mentioned during your lecture the missing metrics to evaluate the disentanglement in the latent space and the need of a benchmark to compare different methods. Do you know if there is any active work on assessing the results beyond a qualitative standpoint? Thank you!

iame
Автор

Great work! Thank you for presenting it!

Dazitu