Generative Densification: Learning to Densify Gaussians for High-Fidelity Generalizable 3D Reconstruction

no code implementations • 9 Dec 2024 • Seungtae Nam, Xiangyu Sun, Gyeongjin Kang, Younggeun Lee, Seungjun Oh, Eunbyung Park

Generalized feed-forward Gaussian models have achieved significant progress in sparse-view 3D reconstruction by leveraging prior knowledge from large multi-view datasets.

3D Reconstruction

SelfSplat: Pose-Free and 3D Prior-Free Generalizable 3D Gaussian Splatting

no code implementations • 26 Nov 2024 • Gyeongjin Kang, Jisang Yoo, Jihyeon Park, Seungtae Nam, Hyeonsoo Im, Sangheon Shin, Sangpil Kim, Eunbyung Park

Our model addresses these challenges by effectively integrating explicit 3D representations with self-supervised depth and pose estimation techniques, resulting in reciprocal improvements in both pose accuracy and 3D reconstruction quality.

3D Reconstruction Pose Estimation

CodecNeRF: Toward Fast Encoding and Decoding, Compact, and High-quality Novel-view Synthesis

no code implementations • 7 Apr 2024 • Gyeongjin Kang, Younggeun Lee, Seungjun Oh, Eunbyung Park

Neural Radiance Fields (NeRF) have achieved huge success in effectively capturing and representing 3D objects and scenes.

Decoder NeRF +1