Skip to content
This repository has been archived by the owner on Oct 2, 2020. It is now read-only.

Character Recognition #9

Open
sr229 opened this issue Dec 22, 2019 · 2 comments
Open

Character Recognition #9

sr229 opened this issue Dec 22, 2019 · 2 comments
Labels
enhancement New feature or request
Projects
FVP
Milestone
FVP

Comments

@sr229
Copy link
Contributor

sr229 commented Dec 22, 2019

Similar to #8. We can use an AI to train multiple models of anime characters we know and love. There are 4 different ways, for example, to draw Mash Kyrielight from Fate, from four different artists. We should be able to compensate for that. Once we know where the character came from, we can also output the franchise/series that they came from.
@sr229 sr229 added the enhancement New feature or request label Dec 22, 2019
@sr229 sr229 added this to the FVP milestone Dec 22, 2019
@sr229 sr229 added this to To do in FVP via automation Dec 22, 2019
@DonaldTsang
Copy link

DonaldTsang commented Apr 24, 2020 •
edited

Firstly, AI is not all-powerful since there will always be errors.
Secondly, are you only classifying or are you doing Semantic Segmentation too?

In the beginning, we should check https://github.com/KichangKim/DeepDanbooru

@sr229
Copy link
Contributor Author

sr229 commented Apr 25, 2020

We're going for classification only, the primary aim here is to have grouping by character in the near future - ala-Pixiv

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Assignees
No one assigned
Labels
enhancement New feature or request
Projects
FVP
  
To do
Milestone
FVP
Development

No branches or pull requests
2 participants
@DonaldTsang @sr229