1 of 17

A.F.R.O. LOGOBOT

Automotive Fast Responsible Open source

2 of 17

“Hello! My name is AFRO. I’m here to help you to quickly generate a logo for your company using a few words.”

3 of 17

PROJECT OBJECTIVES

SAVE MONEY

OPTIMIZE TIME

BE VARIOUS

4 of 17

INSIDE THE MODEL

User inputs text

Image denoising and super resolution

StyleGAN2 takes random vector and returns image

Bot returns image blended with text

01

03

02

04

5 of 17

GAN IN ESSENCE

GAN consists of a generator and discriminator. The generator tries to create an image that will be accepted by the discriminator

At the same time, discriminator tries to decline generated images, which do not look real

6 of 17

GAN TRAINING

It hurts...

First attempts

Some progress!

Not bad

7 of 17

StyleGAN2

If it even can generate waifu for you..

StyleGAN2 - https://arxiv.org/abs/1912.04958

8 of 17

OUTPUTS

9 of 17

SUPER RESOLUTION

32x32

128x128

https://arxiv.org/abs/1809.00219

https://pypi.org/project/ISR/

10 of 17

TEXT TRANSFORMS

https://github.com/albumentations-team/albumentations

https://opencv.org/

11 of 17

What didn't work

12 of 17

MONETIZATION

Advertising sales in the bot messages

PHASE 1

PHASE 3

Propose subscription for users with advertisement removal

PHASE 2

Participation in the government orders (tenders)

13 of 17

FUTURE WORK

Decrease the time for image generation

OPTIMIZATION

Share the bot at the tg-channels

COVERAGE

Increase the image reality and resolution

QUALITY

Prepare for the 1st investment presentation

INVESTMENT

Prepare clusters for input text embeddings

CLUSTERING

14 of 17

TRY IT YOURSELF�(t.me/LogoAfroBot)

15 of 17

TEAM

01

03

02

04

05

ALEKSANDR FROLOV

Team management, text to embeddings, image preprocessing

RODION BUKHANEVICH

Research using CycleGAN, clusterization, data preprocessing

KIRILL BUROVIN

Research using iWGAN-GP, StyleGAN2, SRGAN, literature review

BOGDAN TYSHCHENKO

Telegram bot, user interface, model release to production

ALEKSANDR ISAKOV

Image postprocessing, MVP, presentation, literature review

16 of 17

Contacts:

@kupchanski @AllexFrolov @rd_magic @jityshchenko @asisakov

17 of 17

Last slide behind