Tip for 3D artists, never stop at still images.... Add flythroughs, movement, different lighting scenarios. Finishing with only a still image is wasted potential.show more

This is Ray3.14 with Image-to-Video. Now in native 1080p,... transform still images into cinematic motion. Create depth, parallax, and scene structure with detailed movement, transforming images into expressive, dimensional video with visual clarity. Available now in Dream Machine.show more

introducing: interactive 3d image distortion for Framer add a... smooth distortion effect to your images that reacts to cursor movement—perfect for giving depth and interactivity to your designs. comment "DISTORT" and i’ll send you the component :)show more

Serious question for 3D artists: 🎙️ If you’re still... paying thousands a year for software, what is the one feature that keeps you from switching to Blender? #VFX #b3d #3Dmodelingshow more

MJ v8.2 is really good, with a little Kling... for movement. add "--preview" at the end of your MJ prompt.show more

I created this using Claude + 3D animation: 1.... Ask Claude for a scroll-based cinematic animation pla 2. Generate prompts for depth, parallax, and camera movement 3. Install Three.js + GSAP + ScrollTrigger 4. Use scroll to control zoom, scale, and scene transitions 5. Add layered images for 3D depth illusion 6. Apply smooth easing, lighting, and shadows for realism 7. Refine until it feels like one continuous scene Save thisshow more

i created a fully automated pipeline for me and... my team where the only human feedback is to pick up the winner images everything else: research, script, creating the characters, image prompts, cameras and video clips is done by agents... all by just writting a vague prompt i still have to edit the video, but its mostly import to Capcut, add music, render... that's allshow more

Create a 3D model from a single image, set... of images or a text prompt in < 1 minute 😮‍💨 This new AI paper called CAT3D shows us that it’ll keep getting easier to produce 3D models from 2D images — whether it’s a sparser real world 3D scan (a few photos instead of hundreds) or your favorite 2D image generator like Midjourney (just an image). How does this magic work? “This architecture is similar to video diffusion models, but with camera pose embeddings for each image instead of time embeddings. The generated views are passed into a robust 3D reconstruction pipeline to create the 3D representation (Zip-NeRF or 3DGS)”show more

We wanted to get video overviews in your hands... as soon as we possibly could. While we've been LOVING all the excitement, we know there are still some crucial features we need to add. For example, currently the best way to add images to your video overviews is through Docs, Slides, or image-only PDFs. While support for pulling images from all PDFs is our TOP priority, here's a "super easy" and "clear to understand" hack on how to make it work in the meantime ❤️❤️❤️show more

This is not camera footage. It is a Blender... character with 8K skin, detailed wrinkles, wet eyes, facial controls and enough micro-detail to make your brain keep waiting for the person to behave like a person. HumanPro packages that skin workflow into a Blender add-on instead of making artists rebuild it from scratch every time. The interesting AI angle is not “AI made a realistic girl.” A reusable 3D human can keep the same face across thousands of shots, then be relit, reposed, animated and dropped into completely different scenes without the identity drifting every six frames. Add Claude through Blender MCP and the workflow gets stranger: the model can help assemble scenes, adjust cameras and lighting, inspect renders and correct obvious visual problems, while the character system handles the skin and facial structure. It still does not remove the artist. Someone has to control expression, motion, lighting and the exact moment realism quietly turns into a very expensive mannequin. Most AI influencer projects are still fighting prompt consistency one image at a time. A rigged digital human is less magical, but probably much closer to how this becomes an actual production system.show more

In Terminal War, each map will ship with two... weather variations. Still early but here is a closer look at final lighting. Which version do you prefer?show more

#F1 #MiamiGP 🇺🇸 08:36 LOCAL TIME - It is... currently still lightly raining at the circuit. Radar images suggest that the rain will continue for another 40-60 minutes, and then should stop for a dry window. The F2 race may start without rain, but with a wet track.show more

I now managed to make an actual body swap... with Seedance 2 tuning up the prompt, I suppose this tool has a lot of potential for 3D animation. It's still a long way far of making actual "anime" like Sora 2, for example, but for 3D enjoyers it might be worth a try. #bodyswapshow more

"you are still young, child, and the sea is... never still, that is not where your world should be. there's more beyond the round sky you look up at. once you're out of there, it'll be a whole different view" 🥹🥹🥹🥹show more

I've been wanting to make an old school 3D... dungeon crawler game and over the past couple of days, I made a simple 3D dungeon crawler prototype with 2.5D sprites. I'm still playing around with the procedural map placement and getting the lighting and ambient right. Vibe coded with Cursor for Godot engine.show more

By far the best way to generate images at... different angles is this 3D Camera Control node. Instead of describing angles with words, you just click the view you want. Simple UI, and the Qwen Image Edit 2511 multi-angle LoRA keeps things consistent across generations. Workflow below 👇show more

Super useful tip for fellow AI Filmmakers! 🚨 Kling... AI 2.6 is VERY good at making cuts in your generations, similar to SORA. Meaning instead of generating multiple first frame images, simply add in the prompt "Multiple quick cuts" while making sure to describe a bit the content of the different shots you make. This way, all you need is one starting image, run a 10 second generation, and you will have multiple angles to choose from. Major time saver. Follow for more tips!show more

Video diffusion models have strong implicit representations of 3D... shape, material, and lighting, but controlling them with language is cumbersome, and control is critical for artists and animators. GenLit connects these implicit representations with a continuous 5D control signal describing the direction and intensity of a point light source. This enables single-image near-field relighting of an image using a video diffusion model. We use a ControlNet-like approach and show that, with a small amount of synthetic data, GenLit generalizes to complex real-world images. Given a single image and the 5D lighting signal, GenLit creates a video of a moving light source that is inside the scene. It moves around and behind scene objects, producing effects such as shading, cast shadows, secularities, and interreflections with a realism that is hard to obtain with traditional inverse rendering methods. GenLit shows that it is possible to get continuous control over implicit physical processes within a video model. I think this is just the beginning and promises to make such models much more practical for creators. Shrisha Bharadwaj will present today at SIGGRAPH Asia Room: S423/S424, Level 4 @ 13:50 on 15 of Dec.show more