A Modular Framework for 3D Gaussian Splatting and Beyond

ReconX: Reconstruct Any Scene from Sparse Views with Video Diffusion Model

[ICLR 2025] Phidias: A Generative Model for Creating 3D Content from Text, Image, and 3D Conditions with Reference-Augmented Diffusion

[ICLR'25 Oral] No Pose, No Problem: Surprisingly Simple 3D Gaussian Splats from Sparse Unposed Images

GenXD: Generating Any 3D and 4D Scenes. ICLR 2025

[ICCV'25] Ross3D: Reconstructive Visual Instruction Tuning with 3D-Awareness