Peter Kocsis @Peter4AI

📢Code release for 3DGS-LM (#ICCV2025)📢 Gaussian-Splatting made faster by changing the underlying optimizer. It's compatible with other changes that accelerate the optimization along different axes. lukashoel.github.io/3DGS-LM/

Matthias Niessner @MattNiessner

12 months ago

📢Code release for 3DGS-LM (#ICCV2025)📢 Gaussian-Splatting made faster by changing the underlying optimizer. It's compatible with other changes that accelerate the optimization along different axes. lukashoel.github.io/3DGS-LM/

2 54 195 29K 81

Download Image

ScaffoldAvatar: High-Fidelity Gaussian Avatars with Patch Expressions (#SIGGRAPH) We reconstruct ultra-high fidelity photorealistic 3D avatars capable of generating realistic and high-quality animations including freckles and other fine facial details. We operate on patch-based…

📢 LiteReality: Graphics-Ready 3D Scene Reconstruction from RGB-D Scans🏠✨ -> converts RGB-D scans into compact, realistic, and interactive 3D scenes — featuring high-quality meshes, PBR materials, and articulated objects. 📷youtu.be/ecK9m3LXg2c 🌍litereality.github.io

Check out our #ICCV2025 work on functional 3d scan editing, learning to optimize, multi-level 3d captioning, interactive mesh editing, audio-driven avatars, & shape matching! Congrats @ElBoudjogh24002, @liuyuehcheng, @chandan__yes, @hcxrli, @shivangi2201, Emery for amazing work!

Amazing in-the-wild video relighting results! I like the quality of the cast shadows, showing that current diffusion models have understanding about light propagation.

Zian Wang @zianwang97

3 months ago

Amazing in-the-wild video relighting results! I like the quality of the cast shadows, showing that current diffusion models have understanding about light propagation.

9 122 518 53K 341

Download Video

Presenting PrEditor3D at #CVPR2025 📢📢 If you'd like to learn more about our work and discuss 3D generation/editing, come visit our poster on Friday, June 13th, in ExHall D between 10:30-12:30 (Poster #44). Project Page: ziyaerkoc.com/preditor3d

Check out the EnvGS code. Now open-source! Recently added: high-quality real-time GUI rendering to view base color, reflection color, normal map, and depth map. EnvGS performs exceptionally well on highly specular indoor scene reconstruction. Link in the comments! 👇

MrNeRF @janusch_patas

9 months ago

Check out the EnvGS code. Now open-source! Recently added: high-quality real-time GUI rendering to view base color, reflection color, normal map, and depth map. EnvGS performs exceptionally well on highly specular indoor scene reconstruction. Link in the comments! 👇 https://t.co/rqcXBqXEpQ

2 47 341 44K 207

Download Video

📢BecomingLit: Relightable Gaussian Avatars with Hybrid Neural Shading We propose a hybrid neural shading scheme for creating intrinsically decomposed 3DGS head avatars, that allow real-time relighting and animation. 🌍jonathsch.github.io/becominglit/ 📷youtu.be/xPyeIqKdszA

📢Code Release of Pixel3DMM 📢 Looking for a robust and accurate face tracker? Our state-of-the-art tracker handles challenging in-the-wild settings, such as extreme lighting conditions, fast movements, and occlusions. 👨‍💻github.com/SimonGiebenhai… 🌍simongiebenhain.github.io/pixel3dmm/

📢PBR-SR: Mesh PBR Texture Super Resolution from 2D Image Priors📢 We propose a new optimization to up-sample textures of 3D assets (albedo, roughness, metallic, and normal maps) by leveraging 2D super-resolution models. 📝arxiv.org/abs/2506.02846 📽️youtu.be/eaM5S3Mt1RM

Today we're releasing FLUX.1 Kontext - a suite of generative flow matching models that allow you to generate and edit images. Unlike traditional text-to-image models, Kontext understands both text AND images as input, enabling true in-context generation and editing.

📢📢𝐁𝐈𝐆 𝐍𝐄𝐖𝐒: 𝐒𝐮𝐩𝐞𝐫 𝐞𝐱𝐜𝐢𝐭𝐞𝐝 𝐭𝐨 𝐚𝐧𝐧𝐨𝐮𝐧𝐜𝐞 𝐒𝐩𝐀𝐈𝐭𝐢𝐚𝐥 𝐀𝐈 📢📢 We’re building Spatial Foundation Models — a new paradigm of generative AI that reasons about space and time! Really stoked about our world-class team – it’s gonna be mind-boggling!

SpAItial AI @SpAItial_AI

3 months ago

📢📢𝐁𝐈𝐆 𝐍𝐄𝐖𝐒: 𝐒𝐮𝐩𝐞𝐫 𝐞𝐱𝐜𝐢𝐭𝐞𝐝 𝐭𝐨 𝐚𝐧𝐧𝐨𝐮𝐧𝐜𝐞 𝐒𝐩𝐀𝐈𝐭𝐢𝐚𝐥 𝐀𝐈 📢📢 We’re building Spatial Foundation Models — a new paradigm of generative AI that reasons about space and time! Really stoked about our world-class team – it’s gonna be mind-boggling!

38 113 738 141K 397

Download Video

📢 QuickSplat: Fast 3D Surface Reconstruction via Learned Gaussian Initialization @liuyuehcheng learns 2DGS initialization, densification, and optimization priors from ScanNet++ => fast & accurate reconstruction! Project: liu115.github.io/quicksplat

📢GeomHair: Reconstruction of Hair Strands from Colorless 3D Scans📢 We present a novel method to reconstruct hair strands from colorless 3D scans by extracting orientation cues directly from the mesh surface geometry by finding local characteristic lines and from shaded…

📢Pixel3DMM: Versatile Screen-Space Priors for Single-Image 3D Face Reconstruction📢 -> highly accurate face reconstruction by training powerful VITs via surface normals and UV-coordinates estimation. The geometric cues from our 2D foundation model backbone constrain the 3DMM…

📢 LinPrim: Linear Primitives for Differentiable Volumetric Rendering 📢 We use octahedra or tetrahedra as explicit as volumetric building blocks for gradient-based novel view synthesis - as an alternative to 3D Gaussians with discrete, bounded geometry. We show how it can be…

📢ScanEdit: Hierarchically-Guided Functional 3D Scan Editing Edit complex, real-world 3D scans with text -- @ElBoudjogh24002 combines LLM reasoning with geometric optimization to produce physically plausible, instruction-aligned scene edits Check it out: aminebdj.github.io/scanedit/

📢 SHeaP: Self-Supervised Head Predictor Learned via 2D Gaussians 📢 Given a single input image, we predict accurate 3D head geometry, pose, and expression. Previous works (e.g. DECA, EMOCA) use differentiable mesh rasterization to learn a self-supervised head geometry…

Happy to share our latest work: IntrinsiX 🥳 peter-kocsis.github.io/IntrinsiX/ We generate intrinsic components directly from text! Using SDS, our model can generate full room-scale PBR textures, ready for photo-realistic rendering. Big shout-out to @LukasHollein and @MattNiessner!

Matthias Niessner @MattNiessner

5 months ago

Happy to share our latest work: IntrinsiX 🥳 peter-kocsis.github.io/IntrinsiX/ We generate intrinsic components directly from text! Using SDS, our model can generate full room-scale PBR textures, ready for photo-realistic rendering. Big shout-out to @LukasHollein and @MattNiessner!

1 56 249 20K 112

Download Video

📢Announcing our 3D head avatar benchmark📢 Two tasks with hidden test sets: - Dynamic Novel View Synthesis on Heads - Monocular FLAME-driven Head Avatar Reconstruction Our goal is to make research on 3D head avatars more comparable and ultimately increase the realism of…