greetings, /tg/. Counting up all the fantasy/sci-fi/environment

>>54361714
Makes me REALLY wish for a good tag-based file system. Y'know, instead of just files and folders, everything has a list of tags. >>54361732 could be tagged "knight, sword, medieval, antlers, yellow".

>>54361796
Hydrus

File: Barbarian Hill mainpage.jpg (399KB, 900x624px) Image search: [Google]

399KB, 900x624px

>>54361796
I feel like colours would be unnecessary, no?

File: lizardman_by_obrotowy-d5xlij1.jpg (377KB, 1600x958px) Image search: [Google]

377KB, 1600x958px

>>54361837
shit dude. that looks amazing. I was just thinking of spinning up a program that would have done something similar to this, if someone hadn't already done it.

>>54361837
>Hydrus (software), a software suite for simulating subsurface water flow
?

>>54361839
Maybe, until you get someone asking for a yellow knight picture. Then you're all "BAM I got this!"

File: odin___lotc_by_damascus5-d5txjm3.jpg (317KB, 900x1200px) Image search: [Google]

317KB, 900x1200px

>>54361872
https://hydrusnetwork.github.io/hydrus/

"The hydrus network client is a desktop application written for Anonymous and other internet-fluent media nerds who have large media collections. It organises your files into an internal database and browses them with tags instead of folders, a little like a *booru on your desktop. Tags and files can be anonymously shared through custom servers that any user may run. Everything is free, and the source code is included with the release."

>>54361895
Ok, neat. Have to add "tag" to a Google search to find that, lol

Of course, my next wish is for something to automatically tag all my gorillions of files without me doing it.

File: 1398393369028.jpg (191KB, 600x900px) Image search: [Google]

191KB, 600x900px

>>54361946
get on that neural network, anon.

>>54361959
Aw man, that's like years' more work to figure out how to do it. This all started so simply.

File: 1392745590364.jpg (131KB, 1000x630px) Image search: [Google]

131KB, 1000x630px

>>54362035
the problem is that initial data set that seeds the network. Google already has tags that it thinks are related to your image searches and groups them by colour to how related its guesses are. Training a neural network to recognize one aspect of an image is definitely possible in not that much time (a whole folder full of thousands of different images is different ), you could probably set one up and train it to differentiate single characters (esp with face recognition) vs a "scene".

>>54361946
Hydrus has some support for that kind of thing, but it's mostly older images from boorus.