Software Engineering Courses (SECourses)•11mo ago

Hey I made a music video using veo 2, some of you might find it of interest, anyway its all AI (exce

Hey I made a music video using veo 2, some of you might find it of interest, anyway its all AI (except the music). Hope its ok to post here.
https://www.youtube.com/watch?v=XICNVp7yhg0&t=1s

YouTubeRuairi Robinson

The Blue Stone - Steve Moore

Music video for Steve Moore's "The Blue Stone"
Directed/edited by Ruairi Robinson
Visuals generated using Google Veo 2

.tarkan•2/3/25, 8:38 AM

@Dr. Furkan Gözükara Here a gen from a custom fine tune on large images I’ve tiled and cut into 1024x1024 pixel chunk separate pngs to capture every tiny possible detail from the up to 6K large images. It was an experiment and it went well with your config. Original image count was 15 for those large images and then after filing there were over 400 images. So this seems to be the trick to be able to train on any resolution images without losing any details.

RRuairi Robinson Hey I made a music video using veo 2, some of you might find it of interest, any...

roosterpolice•2/3/25, 1:40 PM

This is absolutely stunning mate!

..tarkan @Dr. Furkan Gözükara Here a gen from a custom fine tune on large images I’ve til...

julius•2/3/25, 7:01 PM

interesting! how does the model know the tiles fit together when training?

julius•2/3/25, 7:03 PM

made a new flux fine tuning model of my friend with 58 images, also gave great and trustworthy results. im wondering how low it can go. my first attempt with 15 images was too glass-skin like. but yea at least now i know that 58 images works well!

Jjulius interesting! how does the model know the tiles fit together when training?

.tarkan•2/3/25, 7:57 PM

I've written a python script with chatGPT that will tile them with overlap and is also including the original full size image. Plus I'm using flux dev de-distilled.

.tarkan•2/3/25, 8:03 PM

@julius so yeah the fine details the model can't train on when only the full image is trained on it's own since the image gets automatically downscaled to 1024x1024 so the training will never see the full details in your dataset. I've taken a step further and have 3 different resolutions for the original image in 2x increments so that it will have 1024x1024 which is the full image downscaled to give the training the full context of the subject then a 2048x2048 version which will serve for 2x Ultimate SD Upscale training on 1024x tiles and then a 4096x4096 size tiled in 1024x1024 tiles to serve as data for training for 4x Ultimate SD Upscale and lastly the original resolution whatever it is to be tiled into 1024x1024 tiles with overlap to not miss any detail from the original dataset.

.tarkan•2/3/25, 8:03 PM

Here some more outputs.

.tarkan•2/3/25, 8:03 PM

.tarkan•2/3/25, 8:04 PM

.tarkan•2/3/25, 8:04 PM

FFurkan Gözükara SECourses Click to see attachment

Neuromasters•2/3/25, 9:22 PM

Wow, there are so many neural nets that you've taken apart and assembled

Darth Teethius•2/3/25, 10:35 PM

Has anyone used the nee deep face similarity app? It works great but I'm not sure what sort of numbers are considered very similar in terms of comparison. I know lower is better but is .3 good or still very off for instance?

Jjulius made a new flux fine tuning model of my friend with 58 images, also gave great a...