المشاركة الشهرية
شهريًامقعد متكرر في محتواي. يدخل المشتركون في قائمة انتظار — أو سحب عشوائي عندما يكون عدد المشتركين أكثر من المواقع — ويظهرون في الفيديو نفسه، أو في الاعتمادات، أو في الوصف.
شهريًا
السعر عند الطلب
التالي
But how do AI images and videos actually work? | Guest video by Welch Labs 38 دقيقة 
Something’s Happening to the Ozone Hole 17 دقيقة 
REQUIEM 4 دقيقة 
I tested every new Apple Product! 15 دقيقة 
“The Biggest Android Update Ever” 13 دقيقة 
ONE OF US 3 دقيقة 
What Bothers Physicists About Black Holes (Interview with Brian Cox) 74 دقيقة 
The Napoleonic Wars - OverSimplified (Part 1) 30 دقيقة 
Greatest Silence 3 دقيقة 
The Hidden Engineering Behind the Falkirk Wheel 16 دقيقة 
What's Under Your Feet in New York City? 23 دقيقة 
Killed By A View Of Mount Fuji: BOAC Flight 911 | Fascinating Horror 11 دقيقة 
He paid pardon brokers $600,000. He never got a pardon #shorts 2 دقيقة 
Amtrak's Subtle, Surprising Success 20 دقيقة 
Hurricane vs. Tiny Houses 23 دقيقة 
Rescued U.S. Air Force officer’s first call home #shorts 1 دقيقة 
One Piece of News From Every Country in the World 43 دقيقة 
Will Smith Surprising My Egyptian Dad! 3 دقيقة 
Abandoned - Jungleland Zoo Florida 20 دقيقة 
Are Beavers Actually Good at Dams? 20 دقيقة 
The Hidden Engineering of Floating Bridges 18 دقيقة 
Practical Magic (1998) - The Witches Fly on Halloween! (11/11) | Movieclips 2 دقيقة 
Man Turns an Empty Space into a Dream Garage | @SmartEasyDIYer 
Name That Infrastructure (Ep. 9) #shorts 1 دقيقة 
How Liquid Dampers in Skyscrapers Work 20 دقيقة 
How to be less neurotic | The Gray Area 38 دقيقة 
The Bizarre Bases of Antenna Towers 19 دقيقة 
The new midlife crisis 2 دقيقة 
Volcanoes Explained: Crash Course Geology #13 11 دقيقة 
Why is getting bitten by a rabid animal so dangerous? - Charles Rupprecht 7 دقيقة 
The Best TINY Tech on the Internet! 20 دقيقة 
What Would Linus Save? 1 دقيقة 
Humanity’s Real Plan to Stop This Asteroid (Feat. Mark Rober) 22 دقيقة 
Why This Robot Kills Weeds With Lasers 15 دقيقة 
Freud was more right than you think | The Gray Area 57 دقيقة 
Why America’s Tank Failed in Ukraine 23 دقيقة 
The Color of Your Crops Can Slow Down Global Warming #scishow #science #stem #shorts #sunflowers 1 دقيقة 
It’s ok if you zone out during this video 1 دقيقة 
Why People Thought "Aliens" Made These Rocks 12 دقيقة 
Name That Infrastructure (Ep. 5) #shorts 1 دقيقة 
What is Plate Tectonics: Crash Course Geology #10 10 دقيقة 
Earthquakes Explained: Crash Course Geology #14 12 دقيقة 
California's Water Problem 20 دقيقة 
Finger Family Song with Street Vehicles | Learn Transportation | Nursery Rhyme & Kids Song | BabyBus 31 دقيقة 
I Built a FREE App That Runs Your Entire Business 28 دقيقة 
Why your best ideas usually start as bad ones | Think Like A Musician 5 دقيقة 
How 3 enemies became one country 35 دقيقة 
Why Only 1% of All Possible Flights Are Direct 12 دقيقة 
WINDY Camping! 🍃🏕️ #PeppaPig #Shorts #summer #LEGODUPLO 1 دقيقة 
We're building a REAL LIFE BATMOBILE | EP.1 32 دقيقة 
Learn Colors with Babies | Colors Song for Kids | Nursery Rhyme & Kids Song | BabyBus 41 دقيقة But how do AI images and videos actually work? | Guest video by Welch Labs
2025 38 دقيقة
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
اعمل مع هذا المبدع
الحزم التي ينشرها المبدع وراء هذا الفيديو.
لم يحدد هذا المبدع أسعاره بعد. أرسل طلبًا وسنمرره — يتم الاتفاق على السعر قبل فرض أي رسوم.
توظيف مضمون — فيديو جديد
مرة واحدةظهور مضمون لمدة ١٠ ثوانٍ على الأقل في فيديوي التالي. يعتمد السعر على مكان الإدراج في مدة العرض.
- أول ١-٣ دقائق
- منتصف الفيديو
- نهاية الفيديو
السعر عند الطلب
- أول ١-٣ دقائق
- منتصف الفيديو
- نهاية الفيديو
توظيف مضمون — مقطع قصير
مرة واحدةظهور مضمون لمدة ١٠ ثوانٍ على الأقل داخل مقطع قصير مقتطع من أحد فيديوهاتي الموجودة.
مرة واحدة
السعر عند الطلب