正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Geometry picking, widget textures, custom materials. :) Flutter Scene 0.18.1 📺

Brandon DeRosier

1,628 subscribers

12,728 次观看 • 3 个月前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

This is rendered by Flutter 3.47 Stable using Flutter GPU via the flutter_scene package. macOS, Linux, Windows, iOS, Android, Web, and custom embedders. Today. :) No forks, no platform views, no external renderers, no secondary graphics context. It's vanilla Flutter. Each pass is a custom material over geometry derived from the model at load time. No pre-baked effect meshes. Just simple Dart APIs AOT compiled to arm64 instructions.

This is rendered by Flutter 3.47 Stable using Flutter GPU via the flutter_scene package. macOS, Linux, Windows, iOS, Android, Web, and custom embedders. Today. :) No forks, no platform views, no external renderers, no secondary graphics context. It's vanilla Flutter. Each pass is a custom material over geometry derived from the model at load time. No pre-baked effect meshes. Just simple Dart APIs AOT compiled to arm64 instructions.

Brandon DeRosier

29,781 次观看 • 1 个月前

Build something real with Flutter 3.47. Scene is powered by Flutter's own graphics stack. Impeller and Flutter GPU. No forks, no platform views, no secondary rendering context. macOS, Linux, Windows, iOS, Android, Web, and custom embedders.

Build something real with Flutter 3.47. Scene is powered by Flutter's own graphics stack. Impeller and Flutter GPU. No forks, no platform views, no secondary rendering context. macOS, Linux, Windows, iOS, Android, Web, and custom embedders.

Brandon DeRosier

19,827 次观看 • 1 个月前

This is Scene Editor. Running on Flutter 3.47 stable. Flutter is excellent for building serious 3D desktop apps. Powered entirely by Flutter's built-in graphics stack, Flutter GPU and Impeller. No forks, no platform views, no external rendering contexts. Scene Editor is just a vanilla Flutter application. Build 3D apps for macOS, Linux, Windows, iOS, Android, Web, and custom embedders. Non-destructive importing, overrides on nested prefabs, live material and shader editing, custom gizmos, and component inspectors you define with property annotations. It watches the filesystem, and custom components you write become instantly available in the editor upon saving. Configure, build, and run your live app without leaving it. Flutter is now 3D.

This is Scene Editor. Running on Flutter 3.47 stable. Flutter is excellent for building serious 3D desktop apps. Powered entirely by Flutter's built-in graphics stack, Flutter GPU and Impeller. No forks, no platform views, no external rendering contexts. Scene Editor is just a vanilla Flutter application. Build 3D apps for macOS, Linux, Windows, iOS, Android, Web, and custom embedders. Non-destructive importing, overrides on nested prefabs, live material and shader editing, custom gizmos, and component inspectors you define with property annotations. It watches the filesystem, and custom components you write become instantly available in the editor upon saving. Configure, build, and run your live app without leaving it. Flutter is now 3D.

Brandon DeRosier

12,184 次观看 • 1 个月前

Been pushing Rive's new 3D and shader support to see what's possible 🚀 Using AI + scripting, I was able to: • Import 3D models • Generate cameras and lights • Create custom shaders (wireframe, AO, sketch, toon) • Build materials • Project animated textures onto 3D geometry • Place 2D components directly in 3D space It's pretty wild to see how quickly Rive is evolving into a true all-in-one platform for interactive design.

Been pushing Rive's new 3D and shader support to see what's possible 🚀 Using AI + scripting, I was able to: • Import 3D models • Generate cameras and lights • Create custom shaders (wireframe, AO, sketch, toon) • Build materials • Project animated textures onto 3D geometry • Place 2D components directly in 3D space It's pretty wild to see how quickly Rive is evolving into a true all-in-one platform for interactive design.

Sanu Sagar

18,218 次观看 • 3 个月前

Issue 039 sacred geometry, a threedimensionsapart moment Scene trendsetter tda delivers emotive guitars & pads, foley textures, and drums both acoustic & digital—all hit-ready sounds with a lot of heart. Cover shot by @elenafortunee Out now on Splice ↓

Issue 039 sacred geometry, a threedimensionsapart moment Scene trendsetter tda delivers emotive guitars & pads, foley textures, and drums both acoustic & digital—all hit-ready sounds with a lot of heart. Cover shot by @elenafortunee Out now on Splice ↓

Moment

11,504 次观看 • 2 年前

UNREAL is 25 years old today so I made a 3D short to celebrate one of the most important FPS (and engines!). Maxon Cinema 4D + OTOY Octane. Lots of materials/textures from Greyscalegorilla and some custom ones. Fullscreen and headphones for the full 1998-nostalgia experience!

UNREAL is 25 years old today so I made a 3D short to celebrate one of the most important FPS (and engines!). Maxon Cinema 4D + OTOY Octane. Lots of materials/textures from Greyscalegorilla and some custom ones. Fullscreen and headphones for the full 1998-nostalgia experience!

Jose

189,097 次观看 • 3 年前

A look at the 5 new completely custom weapons we've added to Runeteria!💀These will either be dropped directly or crafted with materials from our dungeon "Hollow Crypt"☠️We've added custom particles and trails along with emissive textures to give the weapons that extra skeletal finish👌 Also notice the custom signature ability added to our first UNIQUE item💫 Aumo Has been cooking🧑‍🍳 Stay tuned for the first look at our custom made boss fight with some spicy custom mechanics! #Hytale

A look at the 5 new completely custom weapons we've added to Runeteria!💀These will either be dropped directly or crafted with materials from our dungeon "Hollow Crypt"☠️We've added custom particles and trails along with emissive textures to give the weapons that extra skeletal finish👌 Also notice the custom signature ability added to our first UNIQUE item💫 Aumo Has been cooking🧑‍🍳 Stay tuned for the first look at our custom made boss fight with some spicy custom mechanics! #Hytale

Runeteria

16,160 次观看 • 7 个月前

Introducing H3 Max Camera Controls, the world's fastest 3D video model. Turn a single viewpoint into a navigable 3D scene in less than 3 seconds. Set the horizontal and vertical camera angle in degrees, and H3 Max Camera Controls moves through the scene while preserving geometry, position, and materials across perspectives. Fast enough to control the camera in real time.

Introducing H3 Max Camera Controls, the world's fastest 3D video model. Turn a single viewpoint into a navigable 3D scene in less than 3 seconds. Set the horizontal and vertical camera angle in degrees, and H3 Max Camera Controls moves through the scene while preserving geometry, position, and materials across perspectives. Fast enough to control the camera in real time.

fal

211,072 次观看 • 5 天前

Hi3D V3.0 is out today, and it is the most detailed and realistic 3D generator I have run. And its first model to build geometry at 2048³ voxels. The materials were the surprise. PBR output is well ahead of V2.1, and fine surface relief survives where the other frontier models smooth it flat. I compared it against the current top models across a wide spread of objects, textures on and textures off. It lands at number 2 in my ranking. For 3D printing I think it takes first outright. Also, they opened 48 hours of free access for the launch, which almost nobody does. Full breakdown:

Hi3D V3.0 is out today, and it is the most detailed and realistic 3D generator I have run. And its first model to build geometry at 2048³ voxels. The materials were the surprise. PBR output is well ahead of V2.1, and fine surface relief survives where the other frontier models smooth it flat. I compared it against the current top models across a wide spread of objects, textures on and textures off. It lands at number 2 in my ranking. For 3D printing I think it takes first outright. Also, they opened 48 hours of free access for the launch, which almost nobody does. Full breakdown:

Stefan 3D AI

12,503 次观看 • 28 天前

#ByteDance dropped #Seed3D 2.0 and claims an 80% win rate against all major 3D AI tools. After dominating in the AI-video with #Seedance2, they are now entering the 3D generation. And the focus is no longer just textures — it’s about real usability of assets right after generation. I tested: 🔹Geometry (thin & complex structures) 🔹Part-level disassembly for simulations 🔹Unified PBR materials 🔹Compatibility to real game engines Watch the full breakdown — and let’s discuss: can this model actually change the industry?

#ByteDance dropped #Seed3D 2.0 and claims an 80% win rate against all major 3D AI tools. After dominating in the AI-video with #Seedance2, they are now entering the 3D generation. And the focus is no longer just textures — it’s about real usability of assets right after generation. I tested: 🔹Geometry (thin & complex structures) 🔹Part-level disassembly for simulations 🔹Unified PBR materials 🔹Compatibility to real game engines Watch the full breakdown — and let’s discuss: can this model actually change the industry?

Stefan 3D AI

18,827 次观看 • 4 个月前

This is much worse for 3D artists than another AI image generator. Claude takes one architectural reference, connects directly to Blender, and starts rebuilding it as an actual editable scene. It blocks out the buildings, matches the camera, adds materials, vegetation, lighting, and then renders the result. The interesting part is not the first generation. Claude inspects its own render, notices the composition and lighting are wrong, then goes back into Blender and keeps changing the scene. The final result is still rough. The materials look basic, the geometry needs cleanup, and an experienced artist would make it far better. But AI is no longer producing a flat image and pretending it understands 3D. It is operating the software, judging its own work, and iterating inside the real project. The “AI cannot do their actual work” argument is getting harder to defend.

This is much worse for 3D artists than another AI image generator. Claude takes one architectural reference, connects directly to Blender, and starts rebuilding it as an actual editable scene. It blocks out the buildings, matches the camera, adds materials, vegetation, lighting, and then renders the result. The interesting part is not the first generation. Claude inspects its own render, notices the composition and lighting are wrong, then goes back into Blender and keeps changing the scene. The final result is still rough. The materials look basic, the geometry needs cleanup, and an experienced artist would make it far better. But AI is no longer producing a flat image and pretending it understands 3D. It is operating the software, judging its own work, and iterating inside the real project. The “AI cannot do their actual work” argument is getting harder to defend.

Rina

34,431 次观看 • 1 个月前

Meet LingBot-World 2.0.🌍A 1.3B world model, running continuous interactive generation in real time on a single consumer GPU. 🤖 ⚡ It responds to movement, camera control, combat, and text-driven events in real time, while the scene keeps unfolding instead of ending as a fixed video clip. 🌌 Causal pretraining and MoBA help preserve textures, geometry, and scene consistency over long autoregressive rollouts. 🧠 Consistency and distribution-matching distillation compress multi-step generation and train the model on its own rollouts, reducing accumulated drift. 🛠️ The 14B Causal Pretrain backbone and Bidirectional teacher are also open, giving developers a path to post-train, distill, compress, and adapt new world models. 🔒 CC BY-NC-SA 4.0. Non-commercial use only.

Meet LingBot-World 2.0.🌍A 1.3B world model, running continuous interactive generation in real time on a single consumer GPU. 🤖 ⚡ It responds to movement, camera control, combat, and text-driven events in real time, while the scene keeps unfolding instead of ending as a fixed video clip. 🌌 Causal pretraining and MoBA help preserve textures, geometry, and scene consistency over long autoregressive rollouts. 🧠 Consistency and distribution-matching distillation compress multi-step generation and train the model on its own rollouts, reducing accumulated drift. 🛠️ The 14B Causal Pretrain backbone and Bidirectional teacher are also open, giving developers a path to post-train, distill, compress, and adapt new world models. 🔒 CC BY-NC-SA 4.0. Non-commercial use only.

ModelScope

33,630 次观看 • 6 天前

Virtual Textures in the Browser. This is a WebGL 2.0 implementation I did in 2023. The car model is from Sketchfab by Pavel Matoušek. It's a worst-case for virtual textures because the texture is generated from photogrametry software and has pathologically bad UV layout. The texture is 16,384 x 16,384 pixels, which is 180 Mb of PNG data. Loading this in Blender would crash Blender for me back then. The GPU memory requirement for this texture is 1 Gb. The VT displays the model after loading just 34 Kb (one tile). The other tiles are fetched in via a smart priority queue in the background as you move camera around the scene. The physical page texture is 2k, which is 16 times smaller than the source. The solution supports multiple separate virtual textures, and has GLTF integration. Texture filtering as well as anisotropy are fully supported as well. You can try the demo yourself here: Here's a writeup I did while working on this: The video recorded on a very crummy internet connection and I intentionally reload the page in the middle. The delay comes from the geometry download, actual texture part is near-instant.

Virtual Textures in the Browser. This is a WebGL 2.0 implementation I did in 2023. The car model is from Sketchfab by Pavel Matoušek. It's a worst-case for virtual textures because the texture is generated from photogrametry software and has pathologically bad UV layout. The texture is 16,384 x 16,384 pixels, which is 180 Mb of PNG data. Loading this in Blender would crash Blender for me back then. The GPU memory requirement for this texture is 1 Gb. The VT displays the model after loading just 34 Kb (one tile). The other tiles are fetched in via a smart priority queue in the background as you move camera around the scene. The physical page texture is 2k, which is 16 times smaller than the source. The solution supports multiple separate virtual textures, and has GLTF integration. Texture filtering as well as anisotropy are fully supported as well. You can try the demo yourself here: Here's a writeup I did while working on this: The video recorded on a very crummy internet connection and I intentionally reload the page in the middle. The delay comes from the geometry download, actual texture part is near-instant.

Alex Goldring

19,368 次观看 • 8 个月前

America. Europe. Africa. Asia. Oceania. More than 250,000 real-world locations across 120+ countries have already been transformed into ultra high quality 3D Gaussian Splats by Over the Reality. From city centers and museums to beaches, mountains, and historical landmarks, every mapping created by our community contributes to a growing machine-readable representation of the physical world. This scene started as a simple Insta360 X5 recording and was transformed into a digital twin, where geometry, textures, and spatial structure are preserved and can be used by AI, Robotics, VPS, XR, and next-generation world models. This is what city-scale 3D mapping looks like. The link to become an Insta360 X5 mapper is in the first comment. 👇

America. Europe. Africa. Asia. Oceania. More than 250,000 real-world locations across 120+ countries have already been transformed into ultra high quality 3D Gaussian Splats by Over the Reality. From city centers and museums to beaches, mountains, and historical landmarks, every mapping created by our community contributes to a growing machine-readable representation of the physical world. This scene started as a simple Insta360 X5 recording and was transformed into a digital twin, where geometry, textures, and spatial structure are preserved and can be used by AI, Robotics, VPS, XR, and next-generation world models. This is what city-scale 3D mapping looks like. The link to become an Insta360 X5 mapper is in the first comment. 👇

Over the Reality 🌐

19,398 次观看 • 2 个月前

Will Astra kill us in the future? 😱 Maybe Maybe not. But it absolutely killed this by recreating Barcelona’s Camp Nou in 3D with 86,964 individually selectable seats ❤️‍🔥 You can literally pick any seat and see what your view would look like from that exact position inside the stadium. You can also switch between daylight, golden hour, and floodlit night modes to get a much more realistic feel. I very much loved how it optimized the website’s performance by reducing the overview seat geometry from 2,087,136 triangles to 695,712, which is around a 67% reduction, without reducing the number of seats. The stadium itself is built using procedural geometry and custom BufferGeometry, while Astra used InstancedMesh, spatial batching, and custom distance based LOD to efficiently render and manage nearly 87K selectable seats without creating tens of thousands of separate scene objects. For interaction and realism, it also uses optimized raycasting with broad phase batch rejection, TSL and WebGPU materials, shader driven surface detail, and proper seat level camera transitions. A month ago, I built an imaginary stadium using Claude Fable 5. It was great too, but this feels much more realistic. Astra went further by adding proper doorways, staircases, concourses, tier structures, tunnels, railings, and other architectural details you would expect in an actual stadium. Really classy. We should do this for more stadiums and eventually rethink the ticketing experience altogether. Live:

Will Astra kill us in the future? 😱 Maybe Maybe not. But it absolutely killed this by recreating Barcelona’s Camp Nou in 3D with 86,964 individually selectable seats ❤️‍🔥 You can literally pick any seat and see what your view would look like from that exact position inside the stadium. You can also switch between daylight, golden hour, and floodlit night modes to get a much more realistic feel. I very much loved how it optimized the website’s performance by reducing the overview seat geometry from 2,087,136 triangles to 695,712, which is around a 67% reduction, without reducing the number of seats. The stadium itself is built using procedural geometry and custom BufferGeometry, while Astra used InstancedMesh, spatial batching, and custom distance based LOD to efficiently render and manage nearly 87K selectable seats without creating tens of thousands of separate scene objects. For interaction and realism, it also uses optimized raycasting with broad phase batch rejection, TSL and WebGPU materials, shader driven surface detail, and proper seat level camera transitions. A month ago, I built an imaginary stadium using Claude Fable 5. It was great too, but this feels much more realistic. Astra went further by adding proper doorways, staircases, concourses, tier structures, tunnels, railings, and other architectural details you would expect in an actual stadium. Really classy. We should do this for more stadiums and eventually rethink the ticketing experience altogether. Live:

The Bugged Dev

224,274 次观看 • 3 天前

Updated 3D world generated from a reference image with Opus 5. This burned 20% of my weekly quota, but now I can interact with objects in the space and it remembers previous interactions. It started generating indoors but had to stop it (didn't want to burn through my weekly limits). The interesting part was the iterative refinement from Opus 5 subagents. The first pass was not great visually (see the post below), but quality improved dramatically as the agents iterated and validated their own work (really impressed). Its incredible that these agents have subjective assessment on what "visually" appealing looks like. Prompt used with Opus 5 (ultracode) + reference image: Build an explorable 3D world in Three.js based on the attached image. Requirements First-person controls (WASD + mouse look), collision detection, no clipping through geometry AAA visuals: PBR materials, HDRI-based lighting, real shadows, post-processing (tone mapping, bloom, AA) Physics on anything that moves or can be interacted with 60fps on a mid-range GPU Process Split the work into independent streams: terrain, props/geometry, materials/textures, lighting, physics, controls/camera, performance. Assign one sub-agent per stream. Each sub-agent owns its stream end to end and reports what it built plus what is still weak. Integrate, then run a review pass: walk the world, list every visible defect, fix them. Repeat that loop until a full pass produces zero defects, or until 5 loops. Report anything unresolved either way. Deliver a single runnable project plus launch instructions.

Updated 3D world generated from a reference image with Opus 5. This burned 20% of my weekly quota, but now I can interact with objects in the space and it remembers previous interactions. It started generating indoors but had to stop it (didn't want to burn through my weekly limits). The interesting part was the iterative refinement from Opus 5 subagents. The first pass was not great visually (see the post below), but quality improved dramatically as the agents iterated and validated their own work (really impressed). Its incredible that these agents have subjective assessment on what "visually" appealing looks like. Prompt used with Opus 5 (ultracode) + reference image: Build an explorable 3D world in Three.js based on the attached image. Requirements First-person controls (WASD + mouse look), collision detection, no clipping through geometry AAA visuals: PBR materials, HDRI-based lighting, real shadows, post-processing (tone mapping, bloom, AA) Physics on anything that moves or can be interacted with 60fps on a mid-range GPU Process Split the work into independent streams: terrain, props/geometry, materials/textures, lighting, physics, controls/camera, performance. Assign one sub-agent per stream. Each sub-agent owns its stream end to end and reports what it built plus what is still weak. Integrate, then run a review pass: walk the world, list every visible defect, fix them. Repeat that loop until a full pass produces zero defects, or until 5 loops. Report anything unresolved either way. Deliver a single runnable project plus launch instructions.

Prompt

27,549 次观看 • 1 个月前

Robot simulation is getting a big upgrade! 🔥 Normally, building a simulation environment, a kitchen, warehouse, office takes days or weeks. You have to model the geometry, add textures, set lighting, generate collision meshes… all before a robot can even interact with it. Marble by World Labs removes most of that work. You type a text prompt or give it an image, and it generates a full 3D scene with depth, lighting, and a collider mesh you can export into engines like MuJoCo. That means researchers can quickly create lots of different environments to train robots on, which is exactly what domain randomization needs. If we want robots to learn faster, we need to create worlds faster. ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Robot simulation is getting a big upgrade! 🔥 Normally, building a simulation environment, a kitchen, warehouse, office takes days or weeks. You have to model the geometry, add textures, set lighting, generate collision meshes… all before a robot can even interact with it. Marble by World Labs removes most of that work. You type a text prompt or give it an image, and it generates a full 3D scene with depth, lighting, and a collider mesh you can export into engines like MuJoCo. That means researchers can quickly create lots of different environments to train robots on, which is exactly what domain randomization needs. If we want robots to learn faster, we need to create worlds faster. ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

15,498 次观看 • 9 个月前

The WoodLand Pizzeria is a project in a hand-painted art style from EanWinnie to help them explore procedural workflows. Inspired by an illustration from Japanese artist Nonohara, the scene is reinterpreted as an explorable 3D environment. "One thing I found especially useful during this project was using UE’s PCG framework to create the snow-covered version of the environment. I used PCG to create the snow-covered assets needed throughout the entire scene. Compared to manually creating each snow-covered asset individually, this made iteration much faster and saved me a significant amount of time. I also used custom materials and shaders to control colors, values, and surface response, helping maintain a consistent hand-painted look across different lighting and seasonal conditions." - Shuming Zhang Check out the project on ArtStation:

The WoodLand Pizzeria is a project in a hand-painted art style from EanWinnie to help them explore procedural workflows. Inspired by an illustration from Japanese artist Nonohara, the scene is reinterpreted as an explorable 3D environment. "One thing I found especially useful during this project was using UE’s PCG framework to create the snow-covered version of the environment. I used PCG to create the snow-covered assets needed throughout the entire scene. Compared to manually creating each snow-covered asset individually, this made iteration much faster and saved me a significant amount of time. I also used custom materials and shaders to control colors, values, and surface response, helping maintain a consistent hand-painted look across different lighting and seasonal conditions." - Shuming Zhang Check out the project on ArtStation:

Unreal Engine

20,811 次观看 • 1 个月前

🎲 Dreava-Your dream character, ready in just 3 minutes. Introducing the Dreava Customizable Model — nearly 1,000 adjustable parameters packed into one file. 📦 Core model size: 30MB. Runs seamlessly on both PC and mobile. Background apps stay unaffected. 😲 Nearly 1,000 parameters — and expressions stay exceptionally lifelike across all of them. 🎲 Dreava's exclusive randomizer plugin — one click refreshes the entire look. Your OC can debut in under 3 minutes. 🎬 Step-by-step tutorial materials by Artist Jomi (jomi yan ) — from install to advanced customization. 🎨 Bonus for all buyers — Jomi offers a free custom texture replacement service! Just reach out and get your model's textures tailored to your vision, at no extra cost. 💬 Join the official Draw Market Discord — share your creations, get help, and build together with the community. 🏷️ Limited-time price: $118 !!! 👇 Available now on Drawmarket. #Dreava #DreavaCustomizableModel

🎲 Dreava-Your dream character, ready in just 3 minutes. Introducing the Dreava Customizable Model — nearly 1,000 adjustable parameters packed into one file. 📦 Core model size: 30MB. Runs seamlessly on both PC and mobile. Background apps stay unaffected. 😲 Nearly 1,000 parameters — and expressions stay exceptionally lifelike across all of them. 🎲 Dreava's exclusive randomizer plugin — one click refreshes the entire look. Your OC can debut in under 3 minutes. 🎬 Step-by-step tutorial materials by Artist Jomi (jomi yan ) — from install to advanced customization. 🎨 Bonus for all buyers — Jomi offers a free custom texture replacement service! Just reach out and get your model's textures tailored to your vision, at no extra cost. 💬 Join the official Draw Market Discord — share your creations, get help, and build together with the community. 🏷️ Limited-time price: $118 !!! 👇 Available now on Drawmarket. #Dreava #DreavaCustomizableModel

Draw Market Dreava Customizable Model OUT

157,649 次观看 • 5 个月前