Natalia StopMo NFTNYC2024 60Alpha

I finally got my hands on Seedance 2.0 from @dreamina_ai Get ready for even more crazy creativity. It's incredible! P.S. The video shows how my wife and I met. And yes, as you can see, Seedance supports reference photos of real people.

We obsess over realistic skin, 8K resolution, and textures, yet we often end up with a boring image. Why? Because we forget to step into the director’s chair. AI (whether it’s Nano Banana or Seedream) is a notoriously lazy camera operator. By default, it always takes the safe route: a "medium shot at eye level." It’s technically correct, but emotionally flat—like dry documentary footage. Here is how to control viewer emotion using camera angles: 🔻 High Angle (From Above). Visually pins the subject to the ground. The horizon line sits low. This makes the character appear small, weak, or vulnerable. Perfect for depicting defeat or—interestingly—building trust. (This is exactly how you should be selling your courses! 😉) 🔺 Low Angle (From Below). The classic "Hero Shot." Raise the horizon line, and suddenly you’re not looking at a mere mortal, but an epic hero. This angle adds weight, status, and gravitas. 🐛 Extreme Low / Worm’s Eye (Ground Level). Use this when you need to convey total dominance, a looming threat, or the massive scale of an object. ⚙️ Pro Tip: AI often ignores camera descriptions if they are buried at the end of the prompt (the model’s attention tends to drift in the "tail"). For consistent results, lock in your camera settings right at the start—before describing the action or anything else. 📝 Working Syntax:SHOT: • Composition: [camera proximity], [camera angle ], [Detailed Focus: focal plane, DoF falloff, bokeh character], [Framing] Also, don’t leave the composition up to the algorithm’s random chance. You are the one in charge here, not the neural network. Save this post so you don’t have to hunt for the formula later, and let me know—which angle should we break down next? 👇 photo made in @higgsfield_ai video in @Kling_ai

Google Mixboard - A Free AI-Powered Moodboard 🤩⭐ A new video is out on my ukraine channel (youtube.com/watch?v=K_koT9YQ…) reviewing Google’s fresh experimental service – Mixboard. What it is: It’s a canvas (similar to Canva) and an alternative to Adobe moodboards, but infused with AI magic. You can gather references, but the main feature is generation. You can write a prompt like "Make me 10 frames" and specify camera angles (Wide shot, High angle, View from behind). Pros: 💍Price: It’s free. 💍Speed: Generates very quickly; no waiting in queues like with paid alternatives. 💍Convenience: You can move images, add text, and throw together a storyboard on the fly in just a couple of minutes. Nuances (because there are always some): 💍Access: Not available in Ukraine yet; you need to turn on a VPN (any will do, even Urban). 💍Quality: Only 1K resolution. It’s strictly for drafting. 💍Models: It feels like there’s some kind of "Nano Banana Fast" model under the hood. It often makes mistakes, and the quality isn’t perfect. 💍Limitations: You can’t upload videos; it only works with images. Conclusion: For quick storyboards, when you need to sketch out an idea for $0, it’s top-notch. While Google is letting us use it for free, take advantage of it. Check out the video for more details on how to work with it and the bugs to watch out for ☝️