2 papers
cs.CV2025
LeanGaussian: Breaking Pixel or Point Cloud Correspondence in Modeling 3D Gaussians
Jiamin Wu, Kenkun Liu, Han Gao +3
Recently, Gaussian splatting has demonstrated significant success in novel view synthesis. Current methods often regress Gaussians with pixel or point cloud correspondence, linking…
cs.CV2025
UniGS: Modeling Unitary 3D Gaussians for Novel View Synthesis from Sparse-view Images
Jiamin Wu, Kenkun Liu, Yukai Shi +3
In this work, we introduce UniGS, a novel 3D Gaussian reconstruction and novel view synthesis model that predicts a high-fidelity representation of 3D Gaussians from arbitrary numb…