Tao Lu

Knows few

[Tao Lu @isaac__lt]

🎉🎉Glad to announce that Scaffold-GS has been selected as a highlight at #CVPR2024. Congrats and thanks to all of my talented collaborators. Project page: https://t.co/XHe7qwBxWf

[AK @_akhaliq]

Scaffold-GS: Structured 3D Gaussians for View-Adaptive Rendering

paper page: https://t.co/H41DrhOFEQ

Neural rendering methods have significantly advanced photo-realistic 3D scene rendering in various academic and industrial applications. The recent 3D Gaussian Splatting method has achieved the state-of-the-art rendering quality and speed combining the benefits of both primitive-based representations and volumetric representations. However, it often leads to heavily redundant Gaussians that try to fit every training view, neglecting the underlying scene geometry. Consequently, the resulting model becomes less robust to significant view changes, texture-less area and lighting effects. We introduce Scaffold-GS, which uses anchor points to distribute local 3D Gaussians, and predicts their attributes on-the-fly based on viewing direction and distance within the view frustum. Anchor growing and pruning strategies are developed based on the importance of neural Gaussians to reliably improve the scene coverage. We show that our method effectively reduces redundant Gaussians while delivering high-quality rendering. We also demonstrates an enhanced capability to accommodate scenes with varying levels-of-detail and view-dependent observations, without sacrificing the rendering speed.

[Hongyu Li @Hongyu_Lii]

Can a world model 🌏 learn to predict how 198 different deformable objects move -- not just one rope or cloth?

That question motivated Deform360: 1,980 real-world interactions captured from 41 synchronized views with bimanual touch sensors, which becomes crucial when vision is occluded.

Accepted at #ECCV2026. 🧵
1/5

[Aashish Rai @aashishrai3799]

𝗪𝗵𝗮𝘁 𝗶𝗳 𝟰𝗗 𝗩𝗼𝗹𝘂𝗺𝗲𝘁𝗿𝗶𝗰 𝗩𝗶𝗱𝗲𝗼 𝗰𝗼𝘂𝗹𝗱 𝗯𝗲 𝘀𝘁𝗿𝗲𝗮𝗺𝗲𝗱 𝘂𝘀𝗶𝗻𝗴 𝘀𝘁𝗮𝗻𝗱𝗮𝗿𝗱 𝘃𝗶𝗱𝗲𝗼 𝗰𝗼𝗱𝗲𝗰𝘀?
𝗜𝗻𝘁𝗿𝗼𝗱𝘂𝗰𝗶𝗻𝗴 𝗣𝗮𝗰𝗸𝗨𝗩 (accepted at @CVPR)!

In 𝗣𝗮𝗰𝗸𝗨𝗩, we make 4D Gaussian scenes video-native.
[1/4]

[Yuanbo Xiangli @ambie_kk]

🌟We are presenting GSDF today at #NeurIPS2024!🌟
🏠GSDF is a dual-branch architecture that combines the benefits of 3DGS with SDF to boost both rendering and reconstruction quality. 🔗 https://t.co/zJkoCWnDGz
Welcome to chat with us!
📍East Exhibit Hall A-C #1506

[Yixuan Li @liyixxxuan]

Excited to introduce Proc-GS🚀! A new pipeline integrating procedural modeling with 3D-GS to accelerate 3D building creation and infinite city generation with high flexibility and photo-realistic visual quality.
-Project: https://t.co/5PrL16VaXE
-Paper: https://t.co/Fr5OMhEJ05

[Aran Komatsuzaki @arankomatsuzaki]

Video2Game: Real-time, Interactive, Realistic and Browser-Compatible Environment from a Single Video

proj: https://t.co/hKbExxvBTm
abs: https://t.co/iT8qE71qmT

[AK @_akhaliq]

SpatialTracker

Tracking Any 2D Pixels in 3D Space

Recovering dense and long-range pixel motion in videos is a challenging problem. Part of the difficulty arises from the 3D-to-2D projection process, leading to occlusions and discontinuities in the 2D motion domain.

[Linning XU @EvenEveno]

check the code and demos at https://t.co/5TlGvPB3Zq 🚀 :)