Щомісячна участь
на місяцьРегулярне місце в моєму контенті. Підписники потрапляють у чергу — або у випадковий розіграш, коли підписників більше, ніж слотів, — і з'являються в самому відео, у титрах або в описі.
на місяць
Ціна за запитом
Далі
But how do AI images and videos actually work? | Guest video by Welch Labs 38 хв 
I Investigated India’s Biggest Smartphone Controversy 42 хв 
Игра Без Сохранений 2 (Мото VLOG) 58 хв 
Sam Altman Shows Me GPT 5... And What's Next 66 хв 
Google Pixel 11 Hands on - The Phone Crisis Begins. 14 хв 
SEXTANT 2 хв 
UNDEAD DOGS 1 хв 
I hit rock bottom 17 хв 
I Tested YouTubers' Favourite Products 30 хв 
Buying a 1 euro House in Italy - is it a scam? 42 хв 
I'm Actually Short 1 хв 
Fixing the Most Dangerous Dam in the World 23 хв 
Seas #TeamSeas 3 хв 
India’s internet war 9 хв 
EXTREME HIDE & SEEK IN FOOTBALL STADIUM 46 хв 
Spreading Poop Everywhere Situation is Crazy 12 хв 
How the Iran War Spiked Oil Prices 18 хв 
What toys have kids played with throughout history? 6 хв 
How Taiwan is Preparing to Get Invaded 17 хв 
Hurricane vs. Tiny Houses 23 хв 
How Earthquake Bearings Work 19 хв 
Name That Infrastructure (Ep. 10) #shorts 1 хв 
Practical Magic (1998) - The Exorcism of Nicole Kidman! (10/11) | Movieclips 6 хв 
Name That Infrastructure (Ep. 13) #shorts 1 хв 
They Call It A "Drowning Machine" | Fascinating Horror Shorts 1 хв 
Stopping the Unstoppable 16 хв 
The Hidden Engineering of Niagara Falls 16 хв 
China’s Plan to Block All Submarines 21 хв 
There’s something wrong with Hallmark’s youtube channel 19 хв 
How Disney Built A Realistic Lightsaber 17 хв 
How to undo the damage of sitting all day | The Gray Area 42 хв 
Why straight dating is harder than ever | The Gray Area 56 хв 
Why avalanches are most deadly when they stop - Simon Trautman 6 хв 
My Favorite Twilight Knockoff 101 хв 
How We Built the ISS - Part 2 29 хв 
How Self Storage Consumed America 19 хв 
How has technology influenced literature?: Crash Course Latin American Literature #11 11 хв 
The Failed Logistics of Hurricane Katrina 21 хв 
How did ocean invertebrates end up on Mt. Everest?! 1 хв 
What Happens When Science Clashes with the Public?: Crash Course Scientific Thinking #7 13 хв 
How Private Equity Ruined American Youth Sports 19 хв 
Why Fast Food Got So Expensive 20 хв 
Name That Infrastructure (Ep. 3) #shorts 1 хв 
You've Never Seen A Wheelchair Like This 17 хв 
What’s Ruining Our Ruins? 12 хв 
Do Penguins Control the Weather? 8 хв 
Peppa Pig Tales 2026 🤔 Can Peppa Find Out if BABY EVIE Has a TWIN?! 🍼 BRAND NEW Peppa Pig Episodes 121 хв 
Peppa's OUTDOOR Cinema! #PeppaPig #Shorts 1 хв 
Pizza in Italy 🍕 #shorts #peppapig 1 хв 
I almost quit YouTube.... 24 хв 
These Rocks Are Older Than the Sun 13 хв But how do AI images and videos actually work? | Guest video by Welch Labs
2025 38 хв
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
Працюйте з цим творцем
Пакети, опубліковані творцем цього відео.
Цей творець ще не встановив свої тарифи. Надішліть запит, і ми передамо його — ціна узгоджується до будь-яких списань.
Гарантоване розміщення — нове відео
разовоГарантована поява щонайменше 10 секунд у моєму наступному відео. Ціна залежить від того, куди у часі потрапляє вставка.
- Перші 1–3 хвилини
- Середина відео
- Кінець відео
Ціна за запитом
- Перші 1–3 хвилини
- Середина відео
- Кінець відео
Гарантоване розміщення — Short
разовоГарантована поява щонайменше 10 секунд у Short, нарізаному з одного з моїх існуючих відео.
разово
Ціна за запитом