Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

C++ code is out! The Signed Heat Method has been added to geometry-central. In 3D, compute generalized SDFs to point clouds, triangle meshes, and polygon meshes. The method should "just work" on geometry with holes, intersections, nonmanifold-ness, etc. Links below 🔽

Nicole Feng

1,385 subscribers

56,658 views • 1 year ago •via X (Twitter)

Education Science & Technology

Anya Rossi• Live Now

Private livecam show

10 Comments

Nicole Feng1 year ago

Demo projects here: The demo projects let you try things out right away: load your own data, play around with constraints, export solutions. I also included a Blender file with the shader I used to make our paper figures. Enjoy!

Nicole Feng1 year ago

Main X thread here: Paper and project here:

Nicole Feng1 year ago

FYI shoutout to Chris Yu and Nick Sharp (@nmwsharp) for developing the awesome volume grid/tet visualization features in Polyscope.

Hayden1 year ago

sick

Hugo Elias1 year ago

very cool, I love the idea & the paper! SDFs 💞

Dan Brickley1 year ago

Do the point cloud methods make sense for 3D Gaussian splats?

Nicole Feng1 year ago

Haven't tried gaussian splats, but in principle the method should work as you have normals for the geometry - which for gaussians might be done by taking a gradient where you've deemed your zero level set surface to be. Might be some finicky details to figure out.

Ronal Dragon1 year ago

incredible! I'll see if I can replace CGAL's AlphaWrap with it (for offset surfaces)

Kiaran Ritchie1 year ago

Amazing. I could use this.

木十1 year ago

wow

Related Videos

Nvidia presents EdgeRunner! The method can generate high-quality 3D meshes with up to 4,000 faces at a spatial resolution of 512 from images and point-clouds.

Nvidia presents EdgeRunner! The method can generate high-quality 3D meshes with up to 4,000 faces at a spatial resolution of 512 from images and point-clouds.

Dreaming Tulpa 🥓👑

173,103 views • 1 year ago

Super clean and efficient meshes by an AI? YES! The typical 3D Generative AI solutions produce lots of artifacts and usually way to many polygons due to volumetric approaches. In comparison “MeshGPT creates triangle meshes by autoregressively sampling from a transformer model that has been trained to produce tokens from a learned geometric vocabulary. These tokens can then be decoded into the faces of a triangle mesh. This method generates clean, coherent, and compact meshes, characterized by sharp edges and high fidelity.” Surely it is limited by the trained vocabulary but various versions can be trained for specific sets to create generative model libraries for certain object groups. Very promising approach with the high quality.

Super clean and efficient meshes by an AI? YES! The typical 3D Generative AI solutions produce lots of artifacts and usually way to many polygons due to volumetric approaches. In comparison “MeshGPT creates triangle meshes by autoregressively sampling from a transformer model that has been trained to produce tokens from a learned geometric vocabulary. These tokens can then be decoded into the faces of a triangle mesh. This method generates clean, coherent, and compact meshes, characterized by sharp edges and high fidelity.” Surely it is limited by the trained vocabulary but various versions can be trained for specific sets to create generative model libraries for certain object groups. Very promising approach with the high quality.

René Schulte

20,772 views • 2 years ago

Adobe is entering the image-to-3D game! Their new method, LRM, can create high-fidelity 3D meshes from a single image in just 5 seconds 🔥 It's trained on 1 million objects and is able to generate objects from real-world images and generative AI models.

Adobe is entering the image-to-3D game! Their new method, LRM, can create high-fidelity 3D meshes from a single image in just 5 seconds 🔥 It's trained on 1 million objects and is able to generate objects from real-world images and generative AI models.

Dreaming Tulpa 🥓👑

270,942 views • 2 years ago

KIRI Engine 4.2 is here. The biggest upgrade is 3DGS to Mesh 3.0🤩 Cleaner meshes. Better thin-structure retention. Less broken geometry. 20% faster processing. We also added a new Measuring & Scaling tool for 3D printing workflows😆

KIRI Engine 4.2 is here. The biggest upgrade is 3DGS to Mesh 3.0🤩 Cleaner meshes. Better thin-structure retention. Less broken geometry. 20% faster processing. We also added a new Measuring & Scaling tool for 3D printing workflows😆

KIRI Engine - 3D Scanner App

20,001 views • 2 months ago

Nesting insert meshes feels illegal 🚓 100% exportable geometry #normalMagic #b3d #geometrynodes

Nesting insert meshes feels illegal 🚓 100% exportable geometry #normalMagic #b3d #geometrynodes

SpaghetMeNot

399,552 views • 9 months ago

Excited to share 🎨🖌️ 3D Paintbrush - a method for generating local stylized textures on meshes using text as input! Our method predicts a localization map & a highly detailed texture map which conforms to it (1/3)

Excited to share 🎨🖌️ 3D Paintbrush - a method for generating local stylized textures on meshes using text as input! Our method predicts a localization map & a highly detailed texture map which conforms to it (1/3)

Rana Hanocka

48,987 views • 2 years ago

🔥 HUGE: Today, we are thrilled to unveil a massive breakthrough in AI and 3D: ✨Introducing Primfusion v0 💫 For the first time ever in 3D, easily convert your 3D meshes, scans, and high-poly models into easy-to-edit, goopy primitives in Womp. Trained solely on geometry principles, never your style. Join the waitlist ➡️ Questions/Concerns? Check out our FAQ ➡️ #womp3d #3dmodeling

🔥 HUGE: Today, we are thrilled to unveil a massive breakthrough in AI and 3D: ✨Introducing Primfusion v0 💫 For the first time ever in 3D, easily convert your 3D meshes, scans, and high-poly models into easy-to-edit, goopy primitives in Womp. Trained solely on geometry principles, never your style. Join the waitlist ➡️ Questions/Concerns? Check out our FAQ ➡️ #womp3d #3dmodeling

Womp

136,108 views • 2 years ago

It’s really cool to see “legacy” players in geospatial like ESRI adopt 3D gaussian splatting in ArcGIS. Radiance fields are v. complementary to point clouds and textured 3D meshes — exceedingly human readable and much closer to reality than anything else.

It’s really cool to see “legacy” players in geospatial like ESRI adopt 3D gaussian splatting in ArcGIS. Radiance fields are v. complementary to point clouds and textured 3D meshes — exceedingly human readable and much closer to reality than anything else.

Bilawal Sidhu

26,954 views • 4 months ago

DeepMesh is out on Hugging Face Auto-Regressive Artist-mesh Creation with Reinforcement Learning Conditioned on point clouds and images, DeepMesh generates meshes with intricate details and precise topology, outperforming state-of-the-art methods in both precision and quality.

DeepMesh is out on Hugging Face Auto-Regressive Artist-mesh Creation with Reinforcement Learning Conditioned on point clouds and images, DeepMesh generates meshes with intricate details and precise topology, outperforming state-of-the-art methods in both precision and quality.

AK

109,157 views • 1 year ago

Oh, Hyeri used the triangle method on Subin. 🫣 Psychology tricks: if you want someone to fall in love with you, try the triangle method. 😆 🎥: mahcmach #FriendlyRivalry #HyeBin

Oh, Hyeri used the triangle method on Subin. 🫣 Psychology tricks: if you want someone to fall in love with you, try the triangle method. 😆 🎥: mahcmach #FriendlyRivalry #HyeBin

𝕤♡𝕟𝕖_𝕜𝕪

29,721 views • 1 year ago

The subject of Sacred Geometry is FAR MORE IMPORTANT and RELEVANT than many people realize. Sacred Geometry holds the key to what has been called free/unlimited zero point energy technology. The cabal knows this and keeps it hidden and to themselves so they can keep us enslaved Sacred Geometry is like the signature and fingerprint of God.

The subject of Sacred Geometry is FAR MORE IMPORTANT and RELEVANT than many people realize. Sacred Geometry holds the key to what has been called free/unlimited zero point energy technology. The cabal knows this and keeps it hidden and to themselves so they can keep us enslaved Sacred Geometry is like the signature and fingerprint of God.

ĐⱤØ₲Ø🇺🇸

87,063 views • 2 years ago

Learn how to optimize Unreal Engine 5 scenes with Instanced Static Meshes and a clever method for setting up material variations in PolySphere Studio's tutorial. Watch here:

Learn how to optimize Unreal Engine 5 scenes with Instanced Static Meshes and a clever method for setting up material variations in PolySphere Studio's tutorial. Watch here:

80 LEVEL

30,947 views • 1 year ago

To support massive, high-fidelity worlds, our engine implements an efficient LOD strategy with a cloud transcoding system that streams meshes and textures, even on lower end mobile devices. We are expanding this framework to AI upsampling of 3D worlds from prompts. In this preview video below, with a single user prompt, our 3D upsampling technology updates the geometry and texture to transform the classic Roblox game Crossroads into a fantasy world with rich organic detail. 3/4

To support massive, high-fidelity worlds, our engine implements an efficient LOD strategy with a cloud transcoding system that streams meshes and textures, even on lower end mobile devices. We are expanding this framework to AI upsampling of 3D worlds from prompts. In this preview video below, with a single user prompt, our 3D upsampling technology updates the geometry and texture to transform the classic Roblox game Crossroads into a fantasy world with rich organic detail. 3/4

Roblox

1,418,373 views • 4 months ago

📢Happy to present Convex Splatting, a novel way for 3D reconstruction based on 3D smooth convexes. For the first time, a splatting-based method reaches the quality of NeRF sota methods but with real-time rendering and few primitives!! I expect this to replace Gaussian splatting for 3D in the coming months. CODE RELEASED TODAY! joint work with collaborators from Université de Liège Visual Geometry Group (VGG) , KAUST Computer Vision Lab (IVUL) a thread 🧵 1/n

📢Happy to present Convex Splatting, a novel way for 3D reconstruction based on 3D smooth convexes. For the first time, a splatting-based method reaches the quality of NeRF sota methods but with real-time rendering and few primitives!! I expect this to replace Gaussian splatting for 3D in the coming months. CODE RELEASED TODAY! joint work with collaborators from Université de Liège Visual Geometry Group (VGG) , KAUST Computer Vision Lab (IVUL) a thread 🧵 1/n

Abdullah Hamdi

71,661 views • 1 year ago

Curve hair, but you can set different meshes per curve strand! And have options to adjust them. And its using the flat wrapping method so its easy to work with! This is the hair workflow i've wanted for over 7 years, and it can finally be built #b3d

Curve hair, but you can set different meshes per curve strand! And have options to adjust them. And its using the flat wrapping method so its easy to work with! This is the hair workflow i've wanted for over 7 years, and it can finally be built #b3d

aVersionOfReality

147,907 views • 3 years ago

I needed better accuracy in my Blender vertex attribute smoothing, so I built a Laplacian Smooth in Geometry Nodes. Notice the difference compared to blur: it preserves the geometry spacing on triangulated meshes. Basically, it smooths complex topologies as if they were quads. Blender has a modifier version, but it doesn't give as much control. And of course, the next step is to adapt it for smoothing arbitrary attributes, like Normals!

I needed better accuracy in my Blender vertex attribute smoothing, so I built a Laplacian Smooth in Geometry Nodes. Notice the difference compared to blur: it preserves the geometry spacing on triangulated meshes. Basically, it smooths complex topologies as if they were quads. Blender has a modifier version, but it doesn't give as much control. And of course, the next step is to adapt it for smoothing arbitrary attributes, like Normals!

aVersionOfReality

46,069 views • 1 year ago

⚡ Maybe you can't see it now, but the implications of this are massive. Nvidia has found a way to create PERFECT meshes from just point clouds. Let me introduce you to Meshtron 🧵👇

⚡ Maybe you can't see it now, but the implications of this are massive. Nvidia has found a way to create PERFECT meshes from just point clouds. Let me introduce you to Meshtron 🧵👇

Javi Lopez ⛩️

347,211 views • 1 year ago

✏️Tutorial on how to achieve the eyeball effect: This effect is achieved by adding 2 meshes that have inverted normals on them! One is for the whites of the eyes and one is for the iris itself. You can see how the meshes look like in the video below.

✏️Tutorial on how to achieve the eyeball effect: This effect is achieved by adding 2 meshes that have inverted normals on them! One is for the whites of the eyes and one is for the iris itself. You can see how the meshes look like in the video below.

刻矛 MAO 🔆🪱🔆

591,926 views • 1 year ago

New open-source 3D world-generation model. I'm rendering a couple of worlds in the video, so check it out. You'll find the GitHub and the Hugging Face links to the model below. This is a multi-modal world model that you can use for a bunch of things: • To generate new worlds • To reconstruct worlds • To simulate 3D interactive worlds from a prompt, images, or a video You can edit the 3D outputs in Unity and Unreal Engine (they export as meshes, 3DGS files, and point clouds). You can also generate 3D characters in the world and walk around. Pretty fun stuff!

New open-source 3D world-generation model. I'm rendering a couple of worlds in the video, so check it out. You'll find the GitHub and the Hugging Face links to the model below. This is a multi-modal world model that you can use for a bunch of things: • To generate new worlds • To reconstruct worlds • To simulate 3D interactive worlds from a prompt, images, or a video You can edit the 3D outputs in Unity and Unreal Engine (they export as meshes, 3DGS files, and point clouds). You can also generate 3D characters in the world and walk around. Pretty fun stuff!

Santiago

65,308 views • 1 month ago

Check this out. No Booleans, SDFs, or anything fancy. Just setting the normals via geometry nodes. And the setup is right in the 4.5 release notes! The new "Free Normals" option helps this calculate nice and fast.

Check this out. No Booleans, SDFs, or anything fancy. Just setting the normals via geometry nodes. And the setup is right in the 4.5 release notes! The new "Free Normals" option helps this calculate nice and fast.

Jonathan Lampel

134,955 views • 1 year ago