omg u here??? i know u from your videos. thats so cool!
@AICoffeeBreak
2 жыл бұрын
@@NoahElRhandour Haha, I can only reply with: omg, u recognize me??? That is so cool! Yes, I am here. I have to keep a close eye on the competition! 😆
@NoahElRhandour
2 жыл бұрын
@@AICoffeeBreak i see :D
@alexandterfst6532
2 жыл бұрын
Incredible videos
@baothaiba7099
Жыл бұрын
Great work !!!!
@melisakilic726
2 жыл бұрын
So excited for the next one!
@idealintelligence7009
2 жыл бұрын
Thanks boy :) Please speak louder in the video your voice is low.:)
@mchahhou
2 жыл бұрын
awesome!! More of this please.
@filipequincas1485
2 жыл бұрын
Brilliantly explained
@ehsanakefi689
Ай бұрын
very nice thanks👍👍
@AIwithAniket
Жыл бұрын
great video
@raeeskhan9058
Жыл бұрын
you are truly an outlier!
@yassinesafraoui
Жыл бұрын
Hmm isn't trying to train the whole network ( decoder and encoder) using the discriminator just too complicated and would result in a loss function that's so complex that using the gradient descent to minimize it would be inefficient? I mean wouldn't it take a longer time to train? Hence the following idea, why not use separate discriminators to train the decoder and the encoder separately. Yes it would be quite a lot more complicated than this to design but I guess it's worth giving a shot 😀 If someone knows if something like this is already done( cuz I have a feeling it probably is), may he enlighten me, thanks
@AaronBate-g1r
15 күн бұрын
Tressie Ville
@AlsoppCaroline-x9r
14 күн бұрын
Monserrat Knoll
@AnhelDemriya
24 күн бұрын
Taylor Kimberly Lopez Helen Jones Nancy
@DeQuinceyBerton-d7f
14 күн бұрын
Cummings Neck
@HarrietteMoran-l8v
14 күн бұрын
Adolf Grove
@khan.saqibsarwar
4 ай бұрын
That's a really nice video. You summarised so much information in a concise video. And the explanations were crystal clear. Thanks a lot.
@felixvgs.9840
2 жыл бұрын
What an amazing video. Please keep up the great work! :)
@sourabhpatil9406
2 жыл бұрын
Crisp Explanation! I would request you to talk little bit slower, it would be really helpful. Keep up the good work.
@saulcanoortiz7902
7 ай бұрын
Hey! Really great video:) I have one question. Imagine you want to use a diffusion model to learn image-to-image translation, more specifically, from segmentation masks to synthetic images. Then, you can have a tool to create images from hand-painted segmentation masks, and then, you can augment a dataset and see if state-of-the-art segmentation networks trained with the augmented dataset improve its performance. Do you know a diffusion model for this image-to-image translation task with some explanations and available repos?
@JeffersonHilary
10 күн бұрын
Davis Jeffrey Anderson Michael Jackson Steven
@JeavanCooper
4 ай бұрын
The strange patten in the reconstructed image and the generated image is likely to be caused by the perceptual loss, I have no idea why but the disappears when I take the perceptual loss away.
@TheAero
Жыл бұрын
I can't find the VQGan Paper!
@lynnessmacadam3029
10 күн бұрын
Thomas Amy Smith Michael Anderson Mark
@PatriciaDixon-k7v
21 күн бұрын
Harris Shirley Thompson Jason Gonzalez Michelle
@nexyboye5111
Ай бұрын
wow, thanks! now I will be able to encode data more efficiently :3
@MelvilleAlbert-h2j
15 күн бұрын
Anderson Matthew Williams Gary Davis Donna
@louisdrewainwright5359
18 күн бұрын
Davis Susan Williams Dorothy Perez Thomas
@ShaunSteele-n7b
27 күн бұрын
Robinson Edward White Carol Perez Matthew
@tiln8455
2 жыл бұрын
Thank you for this video, now I can be better
@AnhelDemriya
8 күн бұрын
Jackson Michelle Harris Sandra Walker Daniel
@yendar2806
2 жыл бұрын
Ich liebe dich Mathemann❤️
@EdithHazel-g9o
Ай бұрын
Johnson Eric Thomas Robert Young Laura
@wilhelmgoldman9844
19 күн бұрын
Johnson Dorothy Williams Thomas Anderson Amy
@JeffersonHilary
8 күн бұрын
Hernandez Kimberly Miller Linda Hall Daniel
@JeffersonHilary
Ай бұрын
Smith Sandra Gonzalez Edward Lee Richard
@Paul-wk7rp
2 жыл бұрын
Very cool video
@robbyvikleapman3413
15 күн бұрын
Anderson George Hernandez Brenda White Linda
@RobertLee-c9n
8 күн бұрын
Brown Margaret Thomas Maria Johnson Jason
@mlaopiuayhszalaoiusyh
7 күн бұрын
Young Edward Clark Larry Hall Paul
@samanthaelmers4964
22 күн бұрын
Miller Gary Williams Donald Clark Susan
@GloriaOwens-i3n
Ай бұрын
Clark William Lewis Jennifer Brown Jason
@readbyname
5 ай бұрын
Hey great video. Can you tell me why random sampling of codebook vectors doesn't generate a meaningful images. In Vae we random sample from std gaussian, why the same doesn't work for vq auto encoders.
@outliier
5 ай бұрын
Because in a VAE you only predict mean and standard deviation. Sampling this is easier. Sampling the codebook vectors happens independently and this is why the output doesn‘t give a meaningful output.
@MarkoSanace
27 күн бұрын
Johnson Charles Gonzalez Sharon Brown Frank
@BnmJea-x2s
27 күн бұрын
Anderson Helen Garcia Sharon Hall Steven
@reasoning9273
9 ай бұрын
By far the best video on VQVAE. Great job, outlier!
@NoahElRhandour
2 жыл бұрын
didaktisch, visuell und inhaltlich absolut insane, dicke probs
@RoseWilson-u2t
Ай бұрын
White Donna Johnson Helen Martin Helen
@LouisaWright-n3c
21 күн бұрын
Anderson Eric Garcia Donna Lopez Lisa
@MosesRoberta
19 күн бұрын
607 Parker Islands
@WaiBoshard-v6g
14 күн бұрын
Hahn Turnpike
@code4AI
2 жыл бұрын
Excellent visualization for this smooth transition from VQVAE -> VQGAN (focus on main idea first and details second). 10/10
@aratasaki
Жыл бұрын
Incredible video! Can't tell you how much clearer everything is now. Looking forward to the future of your channel!
@outliier
Жыл бұрын
Thats so nice to hear and motivational. The next video is in the making already about CrossAttention
@joanrodriguez6212
2 жыл бұрын
that made some clicks in understanding! thanks a lot
@SawyerBurnell-q9f
15 күн бұрын
Jacobi Port
@rikki146
Жыл бұрын
Why make 2 loss functions with sg instead of optimizing ||E(x) -z_q||_2^2 directly?
@RoseLiggins-k5z
15 күн бұрын
Jarvis Fork
@rezarawassizadeh4601
2 жыл бұрын
after three days of struggling with the paper, I find this amazing explanation for VQ-GAN.
@RhondaPerez-m3b
14 күн бұрын
Mylene Stravenue
@SarahTerrell-x4k
15 күн бұрын
Rodrigo Rapid
@prabhavkaula9697
2 жыл бұрын
Thank you so much for the explanation Hopefully one can now go ahead with clip and create free version of DALL-E like text-to-image models
@smbonilla
Жыл бұрын
Your videos are great! Super clearly explained :) Thanks!!
@devashishprasad1509
Жыл бұрын
This is such a great channel!!!! Why didn't I find it earlier? Thanks a lot for the great work...
@maralzarvani8154
2 жыл бұрын
cool!
@igorvaz6055
2 жыл бұрын
Nice explanation and visualizations!
@MrArtod
Жыл бұрын
How do we decide on what goes to the codebook? Is it filled with random vectors?
@rikki146
Жыл бұрын
It seems to be the case and they are to be converged over the course of training
@김남형산업공학과한양
2 жыл бұрын
개쩐다
@DollyNipples
Жыл бұрын
Those pictures that were generated with VQGAN are surprisingly coherent. How do you do that?
Пікірлер: 74