On the Design of Mixture-of-Experts for Dynamic Gaussian Splatting
- 문서 유형
- 연구출판물 · 학술지 논문
- 등재
- SCIEScopus
- 저널 순위
- JCR Q1Scopus Q1
- 저널
- IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE48(11) · pp.14333 · 2026
- 피인용
- Web of Science 0Scopus 0
- FWCI (Scopus)
- 0
- JCR IF
- JCR 2024 18.6
- ISSN
- 0162-8828
저자
- 진인환제1저자Pusan Natl Univ, Dept Elect & Elect Engn, Busan 46241, South Korea
- 문형주Pusan Natl Univ, Dept Elect & Elect Engn, Busan 46241, South Korea
- Kim, JoonsooElect & Telecommun Res Inst, Immers Media Res Sect, Daejeon 34129, South Korea
- Yun, KugjinElect & Telecommun Res Inst, Immers Media Res Sect, Daejeon 34129, South Korea
- 공경보교신저자Pusan Natl Univ, Dept Elect & Elect Engn, Busan 46241, South Korea
초록
Dynamic scene reconstruction remains challenging due to the heterogeneous and spatially varying nature of real-world motion. Although recent 3D Gaussian Splatting methods have introduced diverse deformation formulations for dynamic novel view synthesis, each method typically relies on a single deformation model within its representation, which limits robustness across diverse dynamic scenarios. In this work, we study a fundamental problem—multi-deformation modeling for dynamic 3D Gaussian representations—under two distinct integration constraints that differ in when and how multiple deformation experts interact during training. From a Mixture-of-Experts (MoE) perspective, we view multi-deformation modeling as the problem of combining multiple specialized deformation models within a unified 3D representation. We first introduce Mixture of Deformation Experts (MoDE), which integrates multiple deformation experts directly into the deformable Gaussian Splatting pipeline through joint optimization. In MoDE, experts operate on a shared canonical Gaussian representation, enabling multi-deformation modeling without introducing additional training stages or modifying the original optimization schedule. In contrast, we further present Mixture of Experts for Dynamic Gaussian Splatting (MoE-GS) under a different integration constraint, where deformation experts are optimized independently and combined through a separate routing stage. As a result, expert interaction occurs over non-canonical Gaussian representations after individual optimization. Together, these two approaches provide alternative strategies for multi-deformation modeling, clarifying how integration constraints shape the design and behavior of deformation experts in dynamic 3D Gaussian representations.
키워드
3D Gaussian Splattingdeformation modelingdynamic novel view synthesisDynamic scene reconstructionMixture of Experts
식별자
- Scopus EID
- 2-s2.0-105045256510
- WoS UID
- 001897133900001



로그인
연구물 목록으로