Saved in:
Bibliographic Details
Main Authors: Bae, Jeongmin, Kim, Seoha, Yun, Youngsik, Lee, Hahyun, Bang, Gun, Uh, Youngjung
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2404.03613
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911968084885504
author Bae, Jeongmin
Kim, Seoha
Yun, Youngsik
Lee, Hahyun
Bang, Gun
Uh, Youngjung
author_facet Bae, Jeongmin
Kim, Seoha
Yun, Youngsik
Lee, Hahyun
Bang, Gun
Uh, Youngjung
contents As 3D Gaussian Splatting (3DGS) provides fast and high-quality novel view synthesis, it is a natural extension to deform a canonical 3DGS to multiple frames for representing a dynamic scene. However, previous works fail to accurately reconstruct complex dynamic scenes. We attribute the failure to the design of the deformation field, which is built as a coordinate-based function. This approach is problematic because 3DGS is a mixture of multiple fields centered at the Gaussians, not just a single coordinate-based framework. To resolve this problem, we define the deformation as a function of per-Gaussian embeddings and temporal embeddings. Moreover, we decompose deformations as coarse and fine deformations to model slow and fast movements, respectively. Also, we introduce a local smoothness regularization for per-Gaussian embedding to improve the details in dynamic regions. Project page: https://jeongminb.github.io/e-d3dgs/
format Preprint
id arxiv_https___arxiv_org_abs_2404_03613
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Per-Gaussian Embedding-Based Deformation for Deformable 3D Gaussian Splatting
Bae, Jeongmin
Kim, Seoha
Yun, Youngsik
Lee, Hahyun
Bang, Gun
Uh, Youngjung
Computer Vision and Pattern Recognition
As 3D Gaussian Splatting (3DGS) provides fast and high-quality novel view synthesis, it is a natural extension to deform a canonical 3DGS to multiple frames for representing a dynamic scene. However, previous works fail to accurately reconstruct complex dynamic scenes. We attribute the failure to the design of the deformation field, which is built as a coordinate-based function. This approach is problematic because 3DGS is a mixture of multiple fields centered at the Gaussians, not just a single coordinate-based framework. To resolve this problem, we define the deformation as a function of per-Gaussian embeddings and temporal embeddings. Moreover, we decompose deformations as coarse and fine deformations to model slow and fast movements, respectively. Also, we introduce a local smoothness regularization for per-Gaussian embedding to improve the details in dynamic regions. Project page: https://jeongminb.github.io/e-d3dgs/
title Per-Gaussian Embedding-Based Deformation for Deformable 3D Gaussian Splatting
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2404.03613