Software Engineering Courses (SECourses)•3y ago

has anybody done video with gpt4v? can we? extract image, combine RAG or something? can we somehow

has anybody done video with gpt4v? can we? extract image, combine RAG or something?

can we somehow use dall e to create video? create sequence of image and 'stictch' them together?

Iiloveh8 has anybody done video with gpt4v? can we? extract image, combine RAG or somethi...

Furkan Gözükara SECourses•11/16/23, 7:37 AM

i doubt that you can do

iloveh8OP•11/16/23, 7:37 AM

why not? thanks for the response

Iiloveh8 why not? thanks for the response

Furkan Gözükara SECourses•11/16/23, 7:37 AM

it can generetate individual images

Furkan Gözükara SECourses•11/16/23, 7:37 AM

not consisten to make video

iloveh8OP•11/16/23, 7:39 AM

i mean video2text is possible?

workflow is: extract image out of video, image2text using GPT4V, combining text together with prompt engineering, store them in embedding/vector db, finetune a model with those embeddings, then generate such images with DALL-E 3, consume 3rd party API to stitch them together?

iloveh8OP•11/16/23, 7:40 AM

i think openAI has new feature to make DALL-3 more consistent now

iloveh8OP•11/16/23, 7:44 AM

at least a slide show now

Iiloveh8 i mean video2text is possible? workflow is: extract image out of video, image2t...

Furkan Gözükara SECourses•11/16/23, 11:11 AM

yes possible

Furkan Gözükara SECourses•11/16/23, 11:11 AM

easy

Furkan Gözükara SECourses•11/16/23, 11:11 AM

with GPT4 vision API

iloveh8OP•11/16/23, 11:13 AM

thanks, any feedback on my approach? cuz vision api is mostly for images not videos, ive not tried the other way round though, whether image2text (gpt4v) and text2image (dall-3) can be made consistent, with the new feature?

Iiloveh8 thanks, any feedback on my approach? cuz vision api is mostly for images not vid...

Furkan Gözükara SECourses•11/16/23, 11:19 AM

you need to extract

Furkan Gözükara SECourses•11/16/23, 11:19 AM

video frames

iloveh8OP•11/16/23, 11:22 AM

thanks

iloveh8OP•11/16/23, 11:22 AM

will try

iloveh8OP•11/16/23, 11:23 AM

any chance you do a video for this

Iiloveh8 any chance you do a video for this

Furkan Gözükara SECourses•11/16/23, 2:15 PM

i dont have

whtwl•11/16/23, 11:01 PM

holy shit my mind is blown. Check this out https://app.suno.ai/ Music generating AI.

whtwl•11/16/23, 11:02 PM

Example on generated song.

edm.mp3920.63KB

Wwhtwl holy shit my mind is blown. Check this out https://app.suno.ai/ Music generatin...

Furkan Gözükara SECourses•11/16/23, 11:10 PM

for now royalty music is better

FFurkan Gözükara SECourses for now royalty music is better

whtwl•11/16/23, 11:40 PM

Yes, but this is still impressive https://cdn.discordapp.com/attachments/563748094345478154/1174848693984309339/87bfafab-ae29-4652-97bf-ebce3d806a54_1.mp4?ex=6569161c&is=6556a11c&hm=b519ec177759ceff2d753a31d3a8313428418e2367dddc7dedd4a59d6bc4f1c4&

paleolithic•11/18/23, 2:30 AM

For sentimate analysis which one would you prefer ALBERT or DIstilBert? I want to run them in a relatively low resource env and want really good results.
Are there other new models that can are better? i don't want to call an API for this.

liber vir•11/18/23, 8:15 AM

hello
what faceswaper is the best?
and which one works best with pets?
also if i want to generate my kitty into Disney character which just one photo which controllent would u recommend ?

Ppaleolithic For sentimate analysis which one would you prefer ALBERT or DIstilBert? I want t...

Furkan Gözükara SECourses•11/18/23, 6:49 PM

let me check

Furkan Gözükara SECourses•11/18/23, 6:50 PM

message.txt2.04KB

mostafa mostafa•11/21/23, 3:44 AM

i dont know if it is helpful for you guys there is a discord server that make ai aniamtions from text
i dont know i can use discord link so i send website link
https://www.mootion.com/landing

mostafa mostafa•11/21/23, 3:45 AM

its under develop

sexyboi•11/22/23, 1:37 AM

Can't wait to fine tune this thing! https://stability.ai/news/stable-video-diffusion-open-ai-video-model

Stability AI

Introducing Stable Video Diffusion — Stability AI

Stable Video Diffusion is a proud addition to our diverse range of open-source models. Spanning across modalities including image, language, audio, 3D, and code, our portfolio is a testament to Stability AI’s dedication to amplifying human intelligence.

Ssexyboi Can't wait to fine tune this thing! https://stability.ai/news/stable-video-diffu...

Teza•11/24/23, 6:04 AM

did you run it?

TTeza did you run it?

sexyboi•11/24/23, 6:10 AM

No not yet, have you?

iloveh8OP•11/24/23, 9:54 PM

whats your most updated course? is it on udemy? or youtube? or patreon

iloveh8OP•11/24/23, 9:55 PM

quite confused

iloveh8OP•11/24/23, 9:55 PM

cuz you have many channels

iloveh8OP•11/24/23, 10:01 PM

where can i find the google colab version

Iiloveh8 where can i find the google colab version

Furkan Gözükara SECourses•11/24/23, 10:31 PM

youtube + patreon

Furkan Gözükara SECourses•11/24/23, 10:31 PM

you need colab version of what?

iloveh8OP•11/24/23, 11:42 PM

im going through your udemy courses, probably need each one, because i dont have GPU on my local machine

Iiloveh8 im going through your udemy courses, probably need each one, because i dont have...

Furkan Gözükara SECourses•11/25/23, 12:31 AM

sdxl newest video will be added tomorrow hopefully

FFurkan Gözükara SECourses sdxl newest video will be added tomorrow hopefully

iloveh8OP•11/25/23, 12:34 AM

how can i get the colab version of each one of the video on udemy? go through the youtube instead?

Iiloveh8 how can i get the colab version of each one of the video on udemy? go through th...

Furkan Gözükara SECourses•11/25/23, 12:50 AM

sadly we dont have such

Furkan Gözükara SECourses•11/25/23, 12:50 AM

i have colab versions for few videos

Furkan Gözükara SECourses•11/25/23, 12:50 AM

but i have usually runpod for all

Furkan Gözükara SECourses•11/25/23, 12:50 AM

colab banned gradio apps

feber13•11/25/23, 4:53 PM

https://ziplora.github.io/ papers
https://github.com/mkshing/ziplora-pytorch code training

GitHub

GitHub - mkshing/ziplora-pytorch: Implementation of "ZipLoRA: Any S...

Implementation of "ZipLoRA: Any Subject in Any Style by Effectively Merging LoRAs" - GitHub - mkshing/ziplora-pytorch: Implementation of "ZipLoRA: Any Subject in Any Styl...

Ffeber13 https://ziplora.github.io/ papers https://github.com/mkshing/ziplora-pytorch co...

Furkan Gözükara SECourses•11/25/23, 9:38 PM

nice

Deme•11/28/23, 8:50 PM

is rerender a video , working if i get the compile from patreon ?

Deme•11/28/23, 9:15 PM

wnet into patreon got acess

DDeme wnet into patreon got acess

Furkan Gözükara SECourses•11/29/23, 1:02 AM

hello

Furkan Gözükara SECourses•11/29/23, 1:02 AM

rerender a video is working from patreon post auto installer

Furkan Gözükara SECourses•11/29/23, 1:02 AM

https://youtu.be/a8oeCFyM5gA

YouTubeSECourses

Turn Videos Into Animation With Just 1 Click - ReRender A Video Tut...

In this video I share automatic Windows installer for ReRender A Video github repository - open source AI project. The auto installer script is here : https://www.patreon.com/posts/1-click-auto-for-89457537

Auto script is here

https://www.patreon.com/posts/1-click-auto-for-89457537

Official repo is here

https://github.com/williamyang19...

has anybody done video with gpt4v? can we? extract image, combine RAG or something? can we somehow

Similar Threads