Měsíční účast
za měsícOpakující se místo v mém obsahu. Odběratelé vstupují do fronty — nebo náhodného losu, když je odběratelů více než slotů — a objevují se v samotném videu, v titulkách nebo v popisu.
za měsíc
Cena na vyžádání
Další
But how do AI images and videos actually work? | Guest video by Welch Labs 38 min 
I Tested YouTubers' Favourite Products 30 min 
The tragedy of Cassandra, the princess who predicted the fall of Troy - Iseult Gillespie 5 min 
HYS 2 min 
I Wrote a NEW Sci-Fi with Project Hail Mary Author Andy Weir 60 min 
STUCK IN THE SYSTEM 5 min 
My First FAILURE!! 9 min 
Why NASA Is Sending This Truck To The Moon 24 min 
LEFT SIDE 3 min 
"he's crazy" - We Found a Man Living Alone in the Amazon (39 years isolated) 43 min 
No Way 13 min ![Can we cross Manchester city centre using its forgotten Victorian tunnels? [PART 2]](https://adm.6-9.space/storage/films/posters/yt_can-we-cross-manchester-city-centre-using-its-forgotten-victorian-tunnels-part-2_1789704303.jpg)
Can we cross Manchester city centre using its forgotten Victorian tunnels? [PART 2] 40 min 
Thunderquakes Let Us Find Secrets Underground 1 min 
He's Back Again 9 min 
Members have a think about this DOOR? 7 min 
What Was Pokemon Shock? | Fascinating Horror Shorts 1 min 
The moment rescuers reached stranded U.S. officer in Iran #shorts 2 min 
So You Want to Build a Tunnel... 21 min 
Stopping the Unstoppable 16 min 
Restoring The Weirdest Xbox 360 Controller - Retro Console Restoration & Repair 15 min 
Concrete's Greatest Weakness is Time 18 min 
Name That Infrastructure (Ep. 11) #shorts 1 min 
Evil Ed’s Terrifying Wolf Transformation | Fright Night 5 min 
Woman Rescues Flooded Car And Restores It Back To New | By @Flame-Frame-VN 
The Physics Behind the Thumb Trick 18 min 
California’s Tallest Bridge Has Nothing Underneath 18 min 
Practical Magic (1998) - Gillian & Sally Kill Jimmy (2/11) | Movieclips 2 min 
Why is everyone suddenly neurodivergent? 8 min 
Sawing a Dam in Half (on Purpose) 21 min 
NPCs in Video Games 13 min 
Infrared Touchscreens #technology #electronic #experiment 1 min 
How NASA Discovered a Military Base under Greenland 22 min 
The Los Angeles Aqueduct is Wild 23 min 
Plasma Globe Can Be a Fire Hazard #experiment #technology #science #funny 1 min 
A wake-up call for the Democratic establishment | America, Actually 48 min 
How old is the Earth?: Crash Course Geology #17 12 min 
Animals & Shapeshifters: Crash Course Latin American Literature #10 10 min 
Do you ever feel isolated amongst your peers? #shorts 1 min 
The Rock Episode: Crash Course Geology #6 9 min 
Why People Thought "Aliens" Made These Rocks 12 min 
How to trick your brain to fall asleep instantly 9 min 
Why men can’t get hard 9 min 
How Taiwan is Preparing to Get Invaded 17 min 
Security Room #PeppaPig #Shorts 1 min 
Tales VS Toys: Dollhouse DISASTER! 🏠 #PeppaPig #Shorts #toys #toyplay 1 min 
I almost quit YouTube.... 24 min 
Take a Bath, Dirty Monster! | Good Habits Song | Bath Song | Nursery Rhymes | BabyBus 24 min 
AI News: OpenAI Finally Released What We Asked For 34 min 
Learn Colors with Vending Machine | Learn Fruits & Veggies | Nursery Rhymes | BabyBus 36 min 
AI News: OpenAI Absolutely Cooked This Week! 35 min 
How 3 enemies became one country 35 min But how do AI images and videos actually work? | Guest video by Welch Labs
2025 38 min
Diffusion models, CLIP, and the math of turning text into images Welch Labs Book: https://www.welchlabs.com/resources/imaginary-numbers-book Sections 0:00 - Intro 3:37 - CLIP 6:25 - Shared Embedding Space 8:16 - Diffusion Models & DDPM 11:44 - Learning Vector Fields 22:00 - DDIM 25:25 - Dall E 2 26:37 - Conditioning 30:02 - Guidance 33:39 - Negative Prompts 34:27 - Outro 35:32 - About guest videos Special Thanks to: Jonathan Ho - Jonathan is the Author of the DDPM paper and the Classifier Free Guidance Paper. https://arxiv.org/pdf/2006.11239 https://arxiv.org/pdf/2207.12598 Preetum Nakkiran - Preetum has an excellent introductory diffusion tutorial: https://arxiv.org/pdf/2406.08929 Chenyang Yuan - Many of the animations in this video were implemented using manim and Chenyang’s smalldiffusion library: https://github.com/yuanchenyang/smalldiffusion Cheyang also has a terrific tutorial and MIT course on diffusion models https://www.chenyang.co/diffusion.html https://www.practical-diffusion.org/ Other References All of Sander Dieleman’s diffusion blog posts are fantastic: https://sander.ai/ CLIP Paper: https://arxiv.org/pdf/2103.00020 DDIM Paper: https://arxiv.org/pdf/2010.02502 Score-Based Generative Modeling: https://arxiv.org/pdf/2011.13456 Wan2.1: https://github.com/Wan-Video/Wan2.1 Stable Diffusion: https://huggingface.co/stabilityai/stable-diffusion-2 Midjourney: https://www.midjourney.com/ Veo: https://deepmind.google/models/veo/ DallE 2 paper: https://cdn.openai.com/papers/dall-e-2.pdf Code for this video: https://github.com/stephencwelch/manim_videos/tree/master/_2025/sora Written by: Stephen Welch, with very helpful feedback from Grant Sanderson Produced by: Stephen Welch, Sam Baskin, and Pranav Gundu Technical Notes The noise videos in the opening have been passed through a VAE (actually, diffusion process happens in a compressed “latent” space), which acts very much like a video compressor - this is why the noise videos don’t look like pure salt and pepper. 6:15 CLIP: Although directly minimizing cosine similarity would push our vectors 180 degrees apart on a single batch, overall in practice, we need CLIP to maximize the uniformity of concepts over the hypersphere it's operating on. For this reason, we animated these vectors as orthogonal-ish. See: https://proceedings.mlr.press/v119/wang20k/wang20k.pdf Per Chenyang Yuan: at 10:15, the blurry image that results when removing random noise in DDPM is probably due to a mismatch in noise levels when calling the denoiser. When the denoiser is called on x_{t-1} during DDPM sampling, it is expected to have a certain noise level (let's call it sigma_{t-1}). If you generate x_{t-1} from x_t without adding noise, then the noise present in x_{t-1} is always smaller than sigma_{t-1}. This causes the denoiser to remove too much noise, thus pointing towards the mean of the dataset. The text conditioning input to stable diffusion is not the 512-dim text embedding vector, but the output of the layer before that, [with dimension 77x512](https://stackoverflow.com/a/79243065) For the vectors at 31:40 - Some implementations use f(x, t, cat) + alpha(f(x, t, cat) - f(x, t)), and some that do f(x, t) + alpha(f(x, t, cat) - f(x, t)), where an alpha value of 1 corresponds to no guidance. I chose the second format here to keep things simpler. At 30:30, the unconditional t=1 vector field looks a bit different from what it did at the 17:15 mark. This is the result of different models trained for different parts of the video, and likely a result of different random initializations. Premium Beat Music ID: EEDYZ3FP44YX8OWT
Pracujte s tímto tvůrcem
Balíčky publikované tvůrcem tohoto videa.
Tento tvůrce si ještě nenastavil své sazby. Odešlete žádost a my ji předáme — cena je dohodnuta před jakýmkoli účtováním.
Garantované umístění — nové video
jednorázověGarantované objevení se po dobu nejméně 10 sekund v mém dalším videu. Cena závisí na tom, kam se vložka v průběhu času umístí.
- První 1–3 minuty
- Střed videa
- Konec videa
Cena na vyžádání
- První 1–3 minuty
- Střed videa
- Konec videa
Garantované umístění — Short
jednorázověGarantované objevení se po dobu nejméně 10 sekund v Shortu sestříhaném z jednoho z mých stávajících videí.
jednorázově
Cena na vyžádání