2019-01-24, 11:34 AM
Valeyard, you're right! Dumb is me 
|
Super resolution
|
Posts: 120
Threads: 5 Joined: 2017 Jun Thanks: 59
Given 59 thank(s) in 44 post(s) Country: 
Posts: 459
Threads: 35 Joined: 2015 Oct Thanks: 1
Given 22 thank(s) in 19 post(s) Country: 
2019-01-24, 02:06 PM
I know I've brought it up before, but I don't suppose it'll hurt to leave this discussion open about using ESRGAN in upscaling.
https://github.com/xinntao/ESRGAN 
Posts: 7,153
Threads: 601 Joined: 2015 Jan Thanks: 1081
Given 1466 thank(s) in 963 post(s) Country: 
2019-01-24, 03:14 PM
(2019-01-24, 02:06 PM)FrankT Wrote: I know I've brought it up before, but I don't suppose it'll hurt to leave this discussion open about using ESRGAN in upscaling. As the images do not work there, for more info go there: https://arxiv.org/pdf/1809.00219.pdf Results are interesting, but how "we" could use it? It does not upscale video, se we should "divide" it in images, and that's right... but, what it takes to upscale a single image (frame)? 
Posts: 360
Threads: 40 Joined: 2015 May Thanks: 124
Given 115 thank(s) in 55 post(s) Country: 
2019-01-24, 08:48 PM
You'll need proper dataset for this to work properly with video and weeks of training. Pretrained stuff does not work well with low resolution video (comparable to bilinear upscale). It just needs more research in that field.
2019-01-25, 02:14 AM
Actually I want to give ESRGAN a try with some SD material. I have a feeling rendering will be very, very slow even with GPU rendering though.
![[Image: sNn6jyF.png]](https://i.imgur.com/sNn6jyF.png) ![[Image: 0sPZMBH.png]](https://i.imgur.com/0sPZMBH.png)
Posts: 7,153
Threads: 601 Joined: 2015 Jan Thanks: 1081
Given 1466 thank(s) in 963 post(s) Country:
2019-01-25, 02:42 AM
Posts: 360
Threads: 40 Joined: 2015 May Thanks: 124
Given 115 thank(s) in 55 post(s) Country:
2019-01-25, 03:26 AM
You can go ahead and just try it out yourself, it's not that hard - they've made program for this ready, but as I said, results are not great, as pretrained models aren't made for movies.
https://github.com/xinntao/BasicSR 
Posts: 46
Threads: 11 Joined: 2019 Jan Thanks: 2
Given 3 thank(s) in 3 post(s) Country: 
2024-06-24, 05:17 PM
(This post was last modified: 2024-06-24, 05:18 PM by CaveDoctor.
Edit Reason: My thoughts on Topaz.
)
I'm going to try out Infognition's software now that it's free.
I want to use AI as well, but I feel that the temporal method might still have some value before I hand it off to an AI. The AI method has so much potential but around a year ago I put an HD version of Jericho's pilot into it, and it just started putting eyes everywhere it could. I kinda feel like it's a beta product still, but version 5.0 of Topaz Video looks very appealing as a tool. Sorry I've been away, but I needed to go away for a while. I've been trying to get the Infognition app working since 2010, but have no idea what I'm doing wrong, so we'll see what all I need to get that in order. 
Posts: 139
Threads: 1 Joined: 2021 Apr Thanks: 270
Given 41 thank(s) in 28 post(s) Country:
2024-10-24, 12:42 PM
(2024-06-24, 05:17 PM)CaveDoctor Wrote: I'm going to try out Infognition's software now that it's free.Did you make it work?
|
|
« Next Oldest | Next Newest »
|
Users browsing this thread: 3 Guest(s)