awesome 3D gaussian splatting 다운로드 - awesome 3D gaussian splatting 소스 코드 다운로드

멋진 3D 가우스 스플래팅 리소스

앞으로 몇 달 동안 예상되는 연구 급증에 보조를 맞추기 위해 3D Gaussian Splatting에 초점을 맞춘 엄선된 논문 및 오픈 소스 리소스 목록입니다. 추가 사항이나 제안 사항이 있으면 언제든지 알려주십시오. 블로그 게시물, 비디오 등과 같은 추가 리소스도 환영합니다.

18개 논문 추가: Z-Splat, Dual-Camera, StylizedGS, Hash3D, Revisiting Densification, Gaussian Pancakes, 3D-aware Deformable Gaussians, SpikeNVS, Zero-shot PC 완성, SplatPose, DreamScene360, RealmDreamer, Gaussian-ILC, GGS를 사용한 강화 학습 , GoMAvatar, OccGaussian, LoopGaussian, 검토

2024년 4월 11일

latentSplat의 코드 릴리스

2024년 4월 9일

1개 논문 추가: EgoLifter

2024년 4월 8일

3개 논문 추가: Robust Gaussian Splatting, SC4D 및 MM-Gaussian

2024년 4월 5일

5개 논문 추가: Surface Reconstruction, TCLC-GS, GaSpCT, OmniGS 및 Per-Gaussian Embedding
수정

2024년 4월 2일

11개 논문 추가: HO, SGD, HGS, Snap-it, InstantSplat, 3DGSR, MM3DGS, HAHA, CityGaussain, Mirror-3DGS 및 Feature Splatting

2024년 3월 30일

8개 논문 추가: 모델링 불확실성, GRM, Gamba, CoherentGS, TOGS, SA-GS 및 GaussianCube

2024년 3월 27일

기타 구현 추가: 360-gaussian-splatting
CVPR '24 라벨이 추가됨
5개 논문 추가: Comp4D, DreamPolisher, DN-Splatter, 2D GS 및 Octree-GS

2024년 3월 26일

13개 논문 추가: latentSplat, GS on the Move, RadSplat, Mini-Splatting, SyncTweedies, HAC, STAG4D, EndoGSLAM, Pixel-GS, Semantic Gaussians, Gaussian in the Wild, CG-SLAM 및 GSDF

2024년 3월 24일 :

추가된 종이: 가우시안 프로스팅

2024년 3월 20일 :

4개 논문 추가: GVGEN, HUGS, RGBD GS-ICP SLAM 및 High-Fidelity SLAM

2024년 3월 19일 :

포인트릭스 추가
원저자가 3DGS 튜토리얼을 추가했습니다.
GauStudio를 추가했습니다.
23개 논문 추가: Touch-GS, GGRt, FDGaussian, SWAG, Den-SOFT, Gaussian-Flow, View-Condependent 3D Editing, BAGS, GeoGaussian, GS-Pose, Analytic-Splatting, Seamless 3D Maps, Texture-GS, Recent Advances 3DGS, Dense Visual SLAM을 위한 Compact 3DGS, BrightDreamer, 3DGS-Reloc, Beyond 불확실성, 모션 인식 3DGS, Fed3DGS, GaussNav, 3DGS-Calib 및 NEDS-SLAM

2024년 3월 17일 :

3DGS.cpp(원래 VulkanSplatting)에 대한 저장소 이름 및 링크 업데이트

2024년 3월 16일 :

스플랫TV
6개 논문 추가: GaussianGrasper, 새로운 분할 알고리즘, 제어 가능한 텍스트-3D 생성, Spring-Mass 3DGS, Hyper-3DGS 및 DreamScene

2024년 3월 14일 :

6개 논문 추가: SemGauss, StyleGaussian, Gaussian Splatting in Style, GaussCtrl, GaussianImage 및 RAIN-GS

2024년 3월 8일 :

튜토리얼: 3DGS용 이미지 캡처 방법
6개 논문 추가: SplattingAvatar, DNGaussian, Radiative Gaussians, BAGS, GSEdit 및 ManiGaussian

2024년 3월 8일 :

3DGStream 뷰어 추가

2024년 3월 6일 :

1개 논문 추가됨: Splat-Nav

2024년 3월 5일 :

1개 논문 추가됨: 3DGStream
코드 릴리스
새로운 뷰어가 추가되었습니다

2024년 3월 2일 :

논문 1개 추가: 애니메이션 및 텍스처링을 위한 3D 가우스 모델
새로운 섹션: 3DGS도 가르치는 코스.

2024년 2월 28일 :

VastGaussian

2024년 2월 27일 :

2개 논문 추가: Spec-Gaussian 및 GEA
SC-GS 코드 출시

2024년 2월 24일 :

논문 2개 추가됨: 불필요한 가우시안 식별 및 Gaussian Pro

2024년 2월 23일 :

EndoGS: Gaussian Splatting을 사용한 변형 가능한 내시경 조직 재구성에 대한 수정된 저자 및 업데이트된 초록

2024년 2월 21일 :

논문 1개 추가: SLAM 재구성: 설문조사

2024년 2월 20일 :

GaussianObject 코드 출시
논문 1개 추가: GaussianHair

2024년 2월 19일 :

블로그 게시물 추가됨: NeRF와 3DGS.

2024년 2월 16일 :

2개 논문 추가: IM-3D 및 GES
GaMeS 코드 출시

2024년 2월 14일 :

추가된 뷰어: VulkanSplatting - C++ 및 Vulkan Compute의 크로스 플랫폼 고성능 3DGS 렌더러

2024년 2월 13일 :

코드 릴리스: (2024년 1월 16일) 4D 가우스 스플래팅을 사용한 실시간 사실적 동적 장면 표현 및 렌더링
3개의 논문이 추가되었습니다: 새로운 비전 시대로서의 3DGala, ImplicitDeepFake 및 3D Gaussians.

2024년 2월 9일 :

논문 1개 추가됨: HeadStudio

2024년 2월 8일 :

3개 논문 추가됨: Rig3DGS, 메시 기반 GS 및 LGM 2024년 2월 6일 :
논문 2개 추가: SGS-SLAM 및 4D Gaussian Splatting

2024년 2월 5일 :

SWAGS를 동적 및 변형 섹션으로 이동했습니다.
논문 2개 추가: GaussianObject 및 GaMeSh
GS++의 이름이 최적 투영으로 변경되었습니다.

2024년 2월 2일 :

6개 논문 추가: VR-GS, Segment Anything, Gaussian Splashing, GS++, 360-GS 및 StopThePop
TRIPS 코드 출시

2024년 1월 30일 :

코드 변경: GaussianAvatars 코드가 비공개로 변경됨

2024년 1월 29일 :

2개의 논문 추가: LIV-GaussMap 및 TIP-Editor

2024년 1월 26일 :

철회된 논문 제거: 인간 동작의 고충실도 합성을 위한 애니메이션 가능한 3D 가우스
3개 논문 추가됨: EndoGaussians, PSAvatar 및 GauU-Scene

2024년 1월 25일 :

추가된 뷰어: Splatapult - C++ 및 OpenGL의 3d 가우스 스플래팅 렌더러, 테더링된 VR용 OpenXR과 함께 작동

2024년 1월 24일 :

유틸리티 추가: SideFX Houdini용 GSOP(Gaussian Splat Operators)
코드 릴리스: GaussianAvatars

2024년 1월 23일 :

3개 논문 추가: Amortized Gen3D, Deformable Endoscopic Tissues, Fast Dynamic 3D Object Generation
코드 릴리스: 애니메이션 가능한 아바타, 압축된 3D 가우스, GaussianAvatar

2024년 1월 13일 :

논문 4개 추가: CoSSegGaussians, TRIPS, 신경 문자를 위한 Gaussian Shadow Casting 및 DISTWAR

2024년 1월 9일 :

논문 1개 추가: 3D 가우스 스플래팅에 관한 조사(1차 조사)

2024년 1월 8일 :

4개의 논문이 추가되었습니다: SWAGS(이전에 추가하는 것을 잊은 2023년에 추가된 논문), 첫 번째 리뷰 논문, 압축된 3DGS 및 위성 기하학 특성화를 위한 응용 논문.

2024년 1월 7일 :

1 오픈 소스 구현: taichi-splatting - 작업은 원래 Taichi 3D Gaussian Splatting에서 파생되었으며 상당한 재구성 및 변경이 이루어졌습니다.

2024년 1월 5일 :

3개의 논문이 추가되었습니다: FMGS, PEGASUS 및 Repaint123.

2024년 1월 2일 :

1개의 논문이 추가되었습니다: Street Gaussians.

2024년 1월 2일 :

디블러링 가우시안 논문 링크가 업데이트되었습니다.
SAGA 코드가 공개되었습니다.
2023년 논문 2개가 추가되었습니다: Text2Immersion 및 2D-Guided 3DG Segmentation.
gsplat lib의 수학적 보충입니다.
카테고리에 연도를 추가하세요.
GSM 코드가 공개되었습니다.

2023년 12월 29일 :

1개의 논문이 추가되었습니다(이전에는 누락된 것으로 보입니다): Gaussian-Head-Avatar.
블로그 게시물 머리 아바타가 추가되었습니다.

2023년 12월 29일 :

3개의 논문이 추가되었습니다: DreamGaussian4D, 4DGen 및 Spacetime Gaussian.

2023년 12월 27일 :

3개의 논문이 추가되었습니다: LangSplat, Deformable 3DGS 및 Human101.
블로그 게시물 추가됨: 3DGS 종합 검토.

2023년 12월 25일 :

단안/다중 뷰 동적 장면을 위한 효율적인 3D 가우스 표현 코드가 출시되었습니다.
GPS-Gaussian 코드가 공개되었습니다.

2023년 12월 24일 :

2개의 논문이 추가되었습니다: Self-Organization Gaussian Grids and Gaussian Splitting.
더 복잡한 장면을 모델링하기 위해 가우시안 렌더링을 향상하기 위한 저장소가 추가되었습니다.

2023년 12월 21일 :

3개의 논문이 추가되었습니다: Splatter Image, pixelSplat 및 가우시안 정렬.
Gaussian Grouping 코드가 공개되었습니다.

2023년 12월 19일 :

2개의 논문이 추가되었습니다: GAvatar 및 GauFRe.

2023년 12월 18일 :

유틸리티 추가: SpectacularAI - 다양한 3DGS 규칙에 대한 변환 스크립트.
SuGaR 코드가 공개되었습니다.

2023년 12월 16일 :

WebGL 뷰어 3: Gauzilla가 추가되었습니다.

2023년 12월 15일 :

4개의 논문이 추가되었습니다: DrivingGaussian, iComMa, Triplane 및 3DGS-Avatar.
Relightable Gaussians 코드가 출시되었습니다.

2023년 12월 13일 :

5개의 논문이 추가되었습니다: Gaussian-SLAM, CoGS, ASH, CF-GS, Photo-SLAM.

2023년 12월 11일 :

2개의 논문이 추가되었습니다: 3D 생성을 위한 Gaussian Splatting SLAM 및 Denoising Scores.
ScaffoldGS 코드가 출시되었습니다.

2023년 12월 8일 :

2개의 논문이 추가되었습니다: EAGLES 및 MonoGaussianAvatar.

2023년 12월 7일 :

LucidDreamer 코드가 공개되었습니다.
9개의 논문이 추가되었습니다: Gaussian-Flow, Feature-3DGS, Gaussian-Avatar, FlashAvatar, Relightable 및 Deblurring Gaussians.

2023년 12월 5일 :

9개의 논문이 추가되었습니다: NeuSG, GaussianHead, GaussianAvatars, GPS-Gaussian, 단안 비강체 객체 재구성을 위한 신경 파라메트릭 가우스, SplaTAM, MANUS, Segment Any 및 언어 내장 3D 가우스.

2023년 12월 4일 :

8개 논문이 추가되었습니다: Gaussian Grouping, MD Splatting, DynMF, Scaffold-GS, SparseGS, FSGS, Control4D 및 SC-GS.

2023년 12월 1일 :

4개의 논문이 추가되었습니다: 효율적인 3D 인간 생성을 위한 Compact3D, GaussianShader, 주기적 진동 가우시안 및 가우시안 쉘 맵.
카테고리별로 목차를 작성하고 줄 바꿈을 추가했습니다.

2023년 11월 30일 :

Unreal 게임 엔진 구현이 추가되었습니다.
5개의 논문이 추가되었습니다: LightGaussian, FisherRF, HUGS, HumanGaussian, CG3D 및 Multi Scale 3DGS.

2023년 11월 29일 :

Point and Move 및 IR-GS라는 두 가지 논문이 추가되었습니다.

2023년 11월 28일 :

GaussinEditor, Relightable Gaussians, GART, Mip-Splatting, HumanGaussian 등 5개 논문을 추가했습니다.

2023년 11월 27일 :

Gaussian Editing 및 Compact 3D Gaussians라는 두 가지 논문이 추가되었습니다.

2023년 11월 25일 :

애니메이션 가능한 가우스 프로젝트가 추가되었습니다(논문은 아직 출시되지 않음).

2023년 11월 22일 :

3개의 새로운 GS 논문이 추가되었습니다: Animatable, Depth-Regularized, Monocular/Multi-view 3DGS.
몇 가지 고전적인 논문을 추가했습니다.
LucidDreamer라는 또 다른 GS 보고서가 추가되었습니다.

2023년 11월 21일 :

3개의 새로운 GS 논문이 추가되었습니다: GaussianDiffusion, LucidDreamer, PhysGaussian.
2개의 GS 논문이 추가되었습니다: SuGaR, PhysGaussian.

2023년 11월 21일 :

GS-SLAM 종이 추가

2023년 11월 17일 :

게임 엔진 섹션에 PlayCanvas 구현을 추가했습니다.

2023년 11월 16일 :

변형 가능한 3D 가우시안 코드가 출시되었습니다.
구동 가능한 3D 가우스 아바타 종이가 추가되었습니다.

2023년 11월 8일 :

3DGS 구현 및 unsive/rsal 형식 논의에 대한 몇 가지 참고 사항입니다.

2023년 11월 4일 :

2D 가우스 스플래팅을 추가했습니다.
3D 가우스 스플래팅을 설명하는 매우 상세한(기술적) 블로그 게시물을 추가했습니다.

2023년 10월 28일 :

유틸리티 섹션이 추가되었습니다.
클라우드 비교 유틸리티에서 3DGS .ply 파일을 편집하기 위한 3DGS 변환기를 추가했습니다.
유틸리티에 대한 변환 지침이 포함된 Kapture(번들러에서 콜맵 모델로 변환용) 및 Kapture 이미지 자르기 스크립트를 추가했습니다.

2023년 10월 23일 :

Python WebGL 뷰어 2가 추가되었습니다.
가우스 스플래팅(및 Unity 뷰어) 비디오 블로그에 소개를 추가했습니다.

2023년 10월 21일 :

Python OpenGL 뷰어가 추가되었습니다.
Typescript WebGPU 뷰어가 추가되었습니다.

2023년 10월 20일 :

초록을 읽을 수 있게 만들었습니다(하이픈 제거).
Windows 튜토리얼을 추가했습니다.
기타 사소한 텍스트 수정.
Jupyter 노트북 뷰어가 추가되었습니다.

2023년 10월 19일 :

실시간 사실적 동적 장면 표현을 위한 Github 페이지 링크가 추가되었습니다.
제목을 재정렬했습니다.
다른 비공식 구현을 추가했습니다.
Nerfstudio gsplat 및 fast: C++/CUDA를 비공식 구현으로 이동했습니다.
Nerfstudio, Blender, WebRTC, iOS 및 Metal 뷰어가 추가되었습니다.

2023년 10월 17일 :

GaussianDreamer 코드가 공개되었습니다.
실시간 사실적인 동적 장면 표현이 추가되었습니다.

2023년 10월 16일 :

변형 가능한 3D 가우스 용지가 추가되었습니다.
동적 3D 가우시안 코드가 출시되었습니다. 2023년 10월 15일 : 처음 6개 논문이 포함된 초기 목록.

3D 가우스 스플래팅을 소개하는 세미나 논문:

실시간 래디언스 필드 렌더링을 위한 3D 가우스 스플래팅

저자 : Bernhard Kerbl, Georgios Kopanas, Thomas Leimkühler, George Drettakis

추상적인

Radiance Field 방법은 최근 여러 사진이나 비디오로 캡처한 장면의 새로운 관점 합성에 혁명을 일으켰습니다. 그러나 높은 시각적 품질을 달성하려면 여전히 훈련 및 렌더링에 비용이 많이 드는 신경망이 필요하며, 최근의 더 빠른 방법은 필연적으로 품질과 속도를 절충합니다. (고립된 객체가 아닌) 무한하고 완전한 장면과 1080p 해상도 렌더링의 경우 현재 어떤 방법으로도 실시간 표시 속도를 달성할 수 없습니다. 경쟁력 있는 훈련 시간을 유지하면서 최첨단 시각적 품질을 달성할 수 있게 하고 중요한 것은 1080p 해상도에서 고품질 실시간(≥ 30fps) 새로운 뷰 합성을 허용하는 세 가지 핵심 요소를 소개합니다. 첫째, 카메라 보정 중에 생성된 희소 점부터 시작하여 빈 공간에서 불필요한 계산을 피하면서 장면 최적화를 위한 연속 체적 복사 필드의 바람직한 속성을 보존하는 3D 가우시안으로 장면을 표현합니다. 둘째, 3D 가우시안의 인터리브 최적화/밀도 제어를 수행하며, 특히 장면의 정확한 표현을 달성하기 위해 이방성 공분산을 최적화합니다. 셋째, 이방성 스플래팅을 지원하고 훈련을 가속화하며 실시간 렌더링을 허용하는 빠른 가시성 인식 렌더링 알고리즘을 개발합니다. 우리는 여러 확립된 데이터 세트에서 최첨단 시각적 품질과 실시간 렌더링을 시연합니다.

3D 객체 감지

2024년

1. 3DGS-DET: 3D 객체 감지를 위한 경계 안내 및 상자 중심 샘플링으로 3D 가우스 스플래팅 강화

저자 : Yang Cao, Yuanliang Jv, Dan Xu

추상적인

NeRF(Neural Radiance Fields)는 새로운 뷰 합성에 널리 사용되며 3DOD(3D 객체 감지)에 적용되어 뷰 합성 표현을 통해 3D 객체 감지에 대한 유망한 접근 방식을 제공합니다. 그러나 NeRF는 본질적인 한계에 직면해 있습니다. (i) 암시적 특성으로 인해 3DOD에 대한 표현 용량이 제한되어 있고 (ii) 렌더링 속도가 느리다는 문제가 있습니다. 최근에는 3DGS(3D Gaussian Splatting)가 더 빠른 렌더링 기능으로 이러한 제한 사항을 해결하는 명시적인 3D 표현으로 등장했습니다. 이러한 장점에 영감을 받아 이 문서에서는 처음으로 3DOD에 3DGS를 도입하여 두 가지 주요 과제를 식별했습니다. (i) 가우시안 얼룩의 모호한 공간 분포 – 3DGS는 주로 2D 픽셀 수준 감독에 의존하므로 가우시안 얼룩의 3D 공간 분포가 불분명합니다. 3DOD를 방해하는 물체와 배경 간의 식별력이 부족합니다. (ii) 과도한 배경 얼룩 – 2D 이미지에는 종종 수많은 배경 픽셀이 포함되어 있어 배경을 나타내는 많은 잡음이 있는 가우스 얼룩이 있는 조밀하게 재구성된 3DGS로 이어져 감지에 부정적인 영향을 미칩니다. 과제(i)를 해결하기 위해 우리는 3DGS 재구성이 2D 이미지에서 파생된다는 사실을 활용하고 2D 경계 안내를 통합하여 가우시안 블롭의 공간 분포를 크게 향상시켜 개체와 개체 간의 명확한 차별화를 제공함으로써 우아하고 효율적인 솔루션을 제안합니다. 그들의 배경(그림 1 참조). 과제 (ii)를 해결하기 위해 우리는 2D 상자를 사용하여 3D 공간에서 객체 확률 분포를 생성하는 상자 중심 샘플링 전략을 제안합니다. 이를 통해 3D에서 효과적인 확률적 샘플링을 통해 더 많은 객체 블롭을 유지하고 노이즈가 있는 배경 블롭을 줄일 수 있습니다. 제안된 경계 안내 및 상자 중심 샘플링의 이점을 활용하는 최종 방법인 3DGS-DET는 추가 학습 가능한 매개변수를 도입하지 않고도 기본 파이프라인 버전에 비해 상당한 개선([email protected]에서 +5.6, [email protected]에서 +3.7)을 달성합니다. . 또한 3DGS-DET는 최첨단 NeRF 기반 방법인 NeRF-Det보다 훨씬 뛰어난 성능을 발휘하여 ScanNet 데이터 세트의 경우 [email protected]에서 +6.6, [email protected]에서 +8.1, 인상적인 +31.5의 개선을 달성했습니다. ARKITScenes 데이터세트의 경우 [email protected]입니다. 코드와 모델은 https://github.com/yangcaoai/3DGS-DET에서 공개적으로 제공됩니다.

? 종이 | 코드(아직 아님)

자율주행:

Despite recent advancements in high-fidelity human reconstruction techniques, the requirements for densely captured images or time-consuming per-instance optimization significantly hinder their applications in broader scenarios. To tackle these issues, we present HumanSplat that predicts the 3D Gaussian Splatting properties of any human from a single input image in a generalizable manner. In particular, HumanSplat comprises a 2D multi-view diffusion model and a latent reconstruction transformer with human structure priors that adeptly integrate geometric priors and semantic features within a unified framework. A hierarchical loss that incorporates human semantic information is further designed to achieve high-fidelity texture modeling and better constrain the estimated multiple views. Comprehensive experiments on standard benchmarks and in-the-wild images demonstrate that HumanSplat surpasses existing state-of-the-art methods in achieving photorealistic novel-view synthesis. Project page: https://humansplat.github.io/.

? 종이 | 프로젝트 페이지

Classic work:

1. A Generalization of Algebraic Surface Drawing

Authors : James F. Blinn

Comment: : First paper rendering 3D gaussians.

추상적인

The mathematical description of three-dimensional surfaces usually falls into one of two classifications: parametric and implicit. An implicit surface is defined to be all points which satisfy some equation F (x, y, z) = 0. This form is ideally suited for image space shaded picture drawing; the pixel coordinates are substituted for x and y, and the equation is solved for z. Algorithms for drawing such objects have been developed primarily for first- and second-order polynomial functions, a subcategory known as algebraic surfaces. This paper presents a new algorithm applicable to other functional forms, in particular to the summation of several Gaussian density distributions. The algorithm was created to model electron density maps of molecular structures, but it can be used for other artistically interesting shapes.

? 종이

2. Approximate Differentiable Rendering with Algebraic Surfaces

Authors : Leonid Keselman and Martial Hebert

Comment: : First paper to do differentiable rendering optimization of 3D gaussians.

추상적인

Differentiable renderers provide a direct mathematical link between an object's 3D representation and images of that object. In this work, we develop an approximate differentiable renderer for a compact, interpretable representation, which we call Fuzzy Metaballs. Our approximate renderer focuses on rendering shapes via depth maps and silhouettes. It sacrifices fidelity for utility, producing fast runtimes and high-quality gradient information that can be used to solve vision tasks. Compared to mesh-based differentiable renderers, our method has forward passes that are 5x faster and backwards passes that are 30x faster. The depth maps and silhouette images generated by our method are smooth and defined everywhere. In our evaluation of differentiable renderers for pose estimation, we show that our method is the only one comparable to classic techniques. In shape from silhouette, our method performs well using only gradient descent and a per-pixel loss, without any surrogate losses or regularization. These reconstructions work well even on natural video sequences with segmentation artifacts.

? 종이 | 프로젝트 페이지 | 코드 | ? Short Presentation

3. Unbiased Gradient Estimation for Differentiable Surface Splatting via Poisson Sampling

Authors : Jan U. Müller, Michael Weinmann, Reinhard Klein

Comment: Builds 2D screen-space gaussians from underlying 3D representations.

추상적인

We propose an efficient and GPU-accelerated sampling framework which enables unbiased gradient approximation for differentiable point cloud rendering based on surface splatting. Our framework models the contribution of a point to the rendered image as a probability distribution. We derive an unbiased approximative gradient for the rendering function within this model. To efficiently evaluate the proposed sample estimate, we introduce a tree-based data-structure which employs multi-pole methods to draw samples in near linear time. Our gradient estimator allows us to avoid regularization required by previous methods, leading to a more faithful shape recovery from images. Furthermore, we validate that these improvements are applicable to real-world applications by refining the camera poses and point cloud obtained from a real-time SLAM system. Finally, employing our framework in a neural rendering setting optimizes both the point cloud and network parameters, highlighting the framework's ability to enhance data driven approaches.

? 종이 코드

4. Generating and Real-Time Rendering of Clouds

Authors : Petr Man

Comment: Splatting of anisotropic gaussians. Basically a non-differentiable implementation of 3DGS.

추상적인

This paper presents a method for generation and real-time rendering of static clouds. Perlin noise function generates three dimensional map of a cloud. We also present a twopass rendering algorithm that performs physically based approximation. In the first preprocessed phase it computes multiple forward scattering. In the second phase first order anisotropic scattering at runtime is evaluated. The generated map is stored as voxels and is unsuitable for the real-time rendering. We introduce a more suitable inner representation of cloud that approximates the original map and contains much less information. The cloud is then represented by a set of metaballs (spheres) with parameters such as center positions, radii and density values. The main contribution of this paper is to propose a method, that transforms the original cloud map to the inner representation. This method uses the Radial Basis Function (RBF) neural network.

? 종이

압축:

3D Gaussian Splatting has recently emerged as a highly promising technique for modeling of static 3D scenes. In contrast to Neural Radiance Fields, it utilizes efficient rasterization allowing for very fast rendering at high-quality. However, the storage size is significantly higher, which hinders practical deployment, eg on resource constrained devices. In this paper, we introduce a compact scene representation organizing the parameters of 3D Gaussian Splatting (3DGS) into a 2D grid with local homogeneity, ensuring a drastic reduction in storage requirements without compromising visual quality during rendering. Central to our idea is the explicit exploitation of perceptual redundancies present in natural scenes. In essence, the inherent nature of a scene allows for numerous permutations of Gaussian parameters to equivalently represent it. To this end, we propose a novel highly parallel algorithm that regularly arranges the high-dimensional Gaussian parameters into a 2D grid while preserving their neighborhood structure. During training, we further enforce local smoothness between the sorted parameters in the grid. The uncompressed Gaussians use the same structure as 3DGS, ensuring a seamless integration with established renderers. Our method achieves a reduction factor of 17x to 42x in size for complex scenes with no increase in training time, marking a substantial leap forward in the domain of 3D scene distribution and consumption.

? 종이 | 프로젝트 페이지 | 암호

확산:

2024년:

1. AGG: Amortized Generative 3D Gaussians for Single Image to 3D

Authors : Dejia Xu, Ye Yuan, Morteza Mardani, Sifei Liu, Jiaming Song, Zhangyang Wang, Arash Vahdat

추상적인

Given the growing need for automatic 3D content creation pipelines, various 3D representations have been studied to generate 3D objects from a single image. Due to its superior rendering efficiency, 3D Gaussian splatting-based models have recently excelled in both 3D reconstruction and generation. 3D Gaussian splatting approaches for image to 3D generation are often optimization-based, requiring many computationally expensive score-distillation steps. To overcome these challenges, we introduce an Amortized Generative 3D Gaussian framework (AGG) that instantly produces 3D Gaussians from a single image, eliminating the need for per-instance optimization. Utilizing an intermediate hybrid representation, AGG decomposes the generation of 3D Gaussian locations and other appearance attributes for joint optimization. Moreover, we propose a cascaded pipeline that first generates a coarse representation of the 3D data and later upsamples it with a 3D Gaussian super-resolution module. Our method is evaluated against existing optimization-based 3D Gaussian frameworks and sampling-based pipelines utilizing other 3D representations, where AGG showcases competitive generation abilities both qualitatively and quantitatively while being several orders of magnitude faster.

? 종이 | Project Page| ? Short Presentation

2. Fast Dynamic 3D Object Generation from a Single-view Video

Authors : Zijie Pan, Zeyu Yang, Xiatian Zhu, Li Zhang

추상적인

Generating dynamic three-dimensional (3D) object from a single-view video is challenging due to the lack of 4D labeled data. Existing methods extend text-to-3D pipelines by transferring off-the-shelf image generation models such as score distillation sampling, but they are slow and expensive to scale (eg, 150 minutes per object) due to the need for back-propagating the information-limited supervision signals through a large pretrained model. To address this limitation, we propose an efficient video-to-4D object generation framework called Efficient4D. It generates high-quality spacetime-consistent images under different camera views, and then uses them as labeled data to directly train a novel 4D Gaussian splatting model with explicit point cloud geometry, enabling real-time rendering under continuous camera trajectories. Extensive experiments on synthetic and real videos show that Efficient4D offers a remarkable 10-fold increase in speed when compared to prior art alternatives while preserving the same level of innovative view synthesis quality. For example, Efficient4D takes only 14 minutes to model a dynamic object.

? 종이 | 프로젝트 페이지 | 코드 | ? Short Presentation

3. GaussianObject: Just Taking Four Images to Get A High-Quality 3D Object with Gaussian Splatting

Authors : Chen Yang, Sikuang Li, Jiemin Fang, Ruofan Liang, Lingxi Xie, Xiaopeng Zhang, Wei Shen, Qi Tian

추상적인

Reconstructing and rendering 3D objects from highly sparse views is of critical importance for promoting applications of 3D vision techniques and improving user experience. However, images from sparse views only contain very limited 3D information, leading to two significant challenges: 1) Difficulty in building multi-view consistency as images for matching are too few; 2) Partially omitted or highly compressed object information as view coverage is insufficient. To tackle these challenges, we propose GaussianObject, a framework to represent and render the 3D object with Gaussian splatting, that achieves high rendering quality with only 4 input images. We first introduce techniques of visual hull and floater elimination which explicitly inject structure priors into the initial optimization process for helping build multi-view consistency, yielding a coarse 3D Gaussian representation. Then we construct a Gaussian repair model based on diffusion models to supplement the omitted object information, where Gaussians are further refined. We design a self-generating strategy to obtain image pairs for training the repair model. Our GaussianObject is evaluated on several challenging datasets, including MipNeRF360, OmniObject3D, and OpenIllumination, achieving strong reconstruction results from only 4 views and significantly outperforming previous state-of-the-art methods.

? 종이 | 프로젝트 페이지 | 코드 | ? Short Presentation

Authors : Heng Yu, Chaoyang Wang, Peiye Zhuang, Willi Menapace, Aliaksandr Siarohin, Junli Cao, Laszlo A Jeni, Sergey Tulyakov, Hsin-Ying Lee

Authors : Junwu Zhang, Zhenyu Tang, Yatian Pang, Xinhua Cheng, Peng Jin, Yida Wei, Munan Ning, Li Yuan

추상적인

Recent one image to 3D generation methods commonly adopt Score Distillation Sampling (SDS). Despite the impressive results, there are multiple deficiencies including multi-view inconsistency, over-saturated and over-smoothed textures, as well as the slow generation speed. To address these deficiencies, we present Repaint123 to alleviate multi-view bias as well as texture degradation and speed up the generation process. The core idea is to combine the powerful image generation capability of the 2D diffusion model and the texture alignment ability of the repainting strategy for generating high-quality multi-view images with consistency. We further propose visibility-aware adaptive repainting strength for overlap regions to enhance the generated image quality in the repainting process. The generated high-quality and multi-view consistent images enable the use of simple Mean Square Error (MSE) loss for fast 3D content generation. We conduct extensive experiments and show that our method has a superior ability to generate high-quality 3D content with multi-view consistency and fine textures in 2 minutes from scratch.

? Paper | Project Page | Code (not yet)

Dynamics and Deformation:

Recently, 3D Gaussian, as an explicit 3D representation method, has demonstrated strong competitiveness over NeRF (Neural Radiance Fields) in terms of expressing complex scenes and training duration. These advantages signal a wide range of applications for 3D Gaussians in 3D understanding and editing. Meanwhile, the segmentation of 3D Gaussians is still in its infancy. The existing segmentation methods are not only cumbersome but also incapable of segmenting multiple objects simultaneously in a short amount of time. In response, this paper introduces a 3D Gaussian segmentation method implemented with 2D segmentation as supervision. This approach uses input 2D segmentation maps to guide the learning of the added 3D Gaussian semantic information, while nearest neighbor clustering and statistical filtering refine the segmentation results. Experiments show that our concise method can achieve comparable performances on mIOU and mAcc for multi-object segmentation as previous single-object segmentation methods.

? 종이

Language Embedding:

3D Gaussian Splatting (3DGS) has recently revolutionized radiance field reconstruction, achieving high quality novel view synthesis and fast rendering speed without baking. However, 3DGS fails to accurately represent surfaces due to the multi-view inconsistent nature of 3D Gaussians. We present 2D Gaussian Splatting (2DGS), a novel approach to model and reconstruct geometrically accurate radiance fields from multi-view images. Our key idea is to collapse the 3D volume into a set of 2D oriented planar Gaussian disks. Unlike 3D Gaussians, 2D Gaussians provide view-consistent geometry while modeling surfaces intrinsically. To accurately recover thin surfaces and achieve stable optimization, we introduce a perspective-accurate 2D splatting process utilizing ray-splat intersection and rasterization. Additionally, we incorporate depth distortion and normal consistency terms to further enhance the quality of the reconstructions. We demonstrate that our differentiable renderer allows for noise-free and detailed geometry reconstruction while maintaining competitive appearance quality, fast training speed, and real-time rendering.

1. [CVPR '24] PhysGaussian: Physics-Integrated 3D Gaussians for Generative Dynamics

Authors : Tianyi Xie, Zeshun Zong, Yuxin Qiu, Xuan Li, Yutao Feng, Yin Yang, Chenfanfu Jiang

추상적인

We introduce PhysGaussian, a new method that seamlessly integrates physically grounded Newtonian dynamics within 3D Gaussians to achieve high-quality novel motion synthesis. Employing a custom Material Point Method (MPM), our approach enriches 3D Gaussian kernels with physically meaningful kinematic deformation and mechanical stress attributes, all evolved in line with continuum mechanics principles. A defining characteristic of our method is the seamless integration between physical simulation and visual rendering: both components utilize the same 3D Gaussian kernels as their discrete representations. This negates the necessity for triangle/tetrahedron meshing, marching cubes, "cage meshes," or any other geometry embedding, highlighting the principle of "what you see is what you simulate (WS2)." Our method demonstrates exceptional versatility across a wide variety of materials--including elastic entities, metals, non-Newtonian fluids, and granular materials--showcasing its strong capabilities in creating diverse visual content with novel viewpoints and movements.

? Paper | Project Page | 코드 | ? Short Presentation

2. [CVPR '24] SuGaR: Surface-Aligned Gaussian Splatting for Efficient 3D Mesh Reconstruction and High-Quality Mesh Rendering

Authors : Antoine Guédon, Vincent Lepetit

추상적인

We propose a method to allow precise and extremely fast mesh extraction from 3D Gaussian Splatting. Gaussian Splatting has recently become very popular as it yields realistic rendering while being significantly faster to train than NeRFs. It is however challenging to extract a mesh from the millions of tiny 3D gaussians as these gaussians tend to be unorganized after optimization and no method has been proposed so far. Our first key contribution is a regularization term that encourages the gaussians to align well with the surface of the scene. We then introduce a method that exploits this alignment to sample points on the real surface of the scene and extract a mesh from the Gaussians using Poisson reconstruction, which is fast, scalable, and preserves details, in contrast to the Marching Cubes algorithm usually applied to extract meshes from Neural SDFs. Finally, we introduce an optional refinement strategy that binds gaussians to the surface of the mesh, and jointly optimizes these Gaussians and the mesh through Gaussian splatting rendering. This enables easy editing, sculpting, rigging, animating, compositing and relighting of the Gaussians using traditional softwares by manipulating the mesh instead of the gaussians themselves. Retrieving such an editable mesh for realistic rendering is done within minutes with our method, compared to hours with the state-of-the-art methods on neural SDFs, while providing a better rendering quality.

? Paper | Project Page | 코드 | ? Short Presentation

3. NeuSG: Neural Implicit Surface Reconstruction with 3D Gaussian Splatting Guidance

Authors : Hanlin Chen, Chen Li, Gim Hee Lee

추상적인

Existing neural implicit surface reconstruction methods have achieved impressive performance in multi-view 3D reconstruction by leveraging explicit geometry priors such as depth maps or point clouds as regularization. However, the reconstruction results still lack fine details because of the over-smoothed depth map or sparse point cloud. In this work, we propose a neural implicit surface reconstruction pipeline with guidance from 3D Gaussian Splatting to recover highly detailed surfaces. The advantage of 3D Gaussian Splatting is that it can generate dense point clouds with detailed structure. Nonetheless, a naive adoption of 3D Gaussian Splatting can fail since the generated points are the centers of 3D Gaussians that do not necessarily lie on the surface. We thus introduce a scale regularizer to pull the centers close to the surface by enforcing the 3D Gaussians to be extremely thin. Moreover, we propose to refine the point cloud from 3D Gaussians Splatting with the normal priors from the surface predicted by neural implicit models instead of using a fixed set of points as guidance. Consequently, the quality of surface reconstruction improves from the guidance of the more accurate 3D Gaussian splatting. By jointly optimizing the 3D Gaussian Splatting and the neural implicit model, our approach benefits from both representations and generates complete surfaces with intricate details. Experiments on Tanks and Temples verify the effectiveness of our proposed method.

? 종이

기타:

In this paper, we address the limitations of Adaptive Density Control (ADC) in 3D Gaussian Splatting (3DGS), a scene representation method achieving high-quality, photorealistic results for novel view synthesis. ADC has been introduced for automatic 3D point primitive management, controlling densification and pruning, however, with certain limitations in the densification logic. Our main contribution is a more principled, pixel-error driven formulation for density control in 3DGS, leveraging an auxiliary, per-pixel error function as the criterion for densification. We further introduce a mechanism to control the total number of primitives generated per scene and correct a bias in the current opacity handling strategy of ADC during cloning operations. Our approach leads to consistent quality improvements across a variety of benchmark scenes, without sacrificing the method's efficiency.

? 종이

2023년:

1. [CVPRW '24] Depth-Regularized Optimization for 3D Gaussian Splatting in Few-Shot Images

Authors : Jaeyoung Chung, Jeongtaek Oh, Kyoung Mu Lee

추상적인

In this paper, we present a method to optimize Gaussian splatting with a limited number of images while avoiding overfitting. Representing a 3D scene by combining numerous Gaussian splats has yielded outstanding visual quality. However, it tends to overfit the training views when only a small number of images are available. To address this issue, we introduce a dense depth map as a geometry guide to mitigate overfitting. We obtained the depth map using a pre-trained monocular depth estimation model and aligning the scale and offset using sparse COLMAP feature points. The adjusted depth aids in the color-based optimization of 3D Gaussian splatting, mitigating floating artifacts, and ensuring adherence to geometric constraints. We verify the proposed method on the NeRF-LLFF dataset with varying numbers of few images. Our approach demonstrates robust geometry compared to the original method that relies solely on images.

? 종이 | Project Page | 암호

2. EAGLES: Efficient Accelerated 3D Gaussians with Lightweight EncodingS

Authors : Sharath Girish, Kamal Gupta, Abhinav Shrivastava

추상적인

Recently, 3D Gaussian splatting (3D-GS) has gained popularity in novel-view scene synthesis. It addresses the challenges of lengthy training times and slow rendering speeds associated with Neural Radiance Fields (NeRFs). Through rapid, differentiable rasterization of 3D Gaussians, 3D-GS achieves real-time rendering and accelerated training. They, however, demand substantial memory resources for both training and storage, as they require millions of Gaussians in their point cloud representation for each scene. We present a technique utilizing quantized embeddings to significantly reduce memory storage requirements and a coarse-to-fine training strategy for a faster and more stable optimization of the Gaussian point clouds. Our approach results in scene representations with fewer Gaussians and quantized representations, leading to faster training times and rendering speeds for real-time rendering of high resolution scenes. We reduce memory by more than an order of magnitude all while maintaining the reconstruction quality. We validate the effectiveness of our approach on a variety of datasets and scenes preserving the visual quality while consuming 10-20x less memory and faster training/inference speed.

? Paper | Project Page | 암호

3. [CVPR '24] COLMAP-Free 3D Gaussian Splatting

Authors : Yang Fu, Sifei Liu, Amey Kulkarni, Jan Kautz, Alexei A. Efros, Xiaolong Wang

추상적인

While neural rendering has led to impressive advances in scene reconstruction and novel view synthesis, it relies heavily on accurately pre-computed camera poses. To relax this constraint, multiple efforts have been made to train Neural Radiance Fields (NeRFs) without pre-processed camera poses. However, the implicit representations of NeRFs provide extra challenges to optimize the 3D structure and camera poses at the same time. On the other hand, the recently proposed 3D Gaussian Splatting provides new opportunities given its explicit point cloud representations. This paper leverages both the explicit geometric representation and the continuity of the input video stream to perform novel view synthesis without any SfM preprocessing. We process the input frames in a sequential manner and progressively grow the 3D Gaussians set by taking one input frame at a time, without the need to pre-compute the camera poses. Our method significantly improves over previous approaches in view synthesis and camera pose estimation under large motion changes.

? Paper | Project Page | Code (not yet) | ? Short Presentation

4. iComMa: Inverting 3D Gaussians Splatting for Camera Pose Estimation via Comparing and Matching

Authors : Yuan Sun, Xuan Wang, Yunfan Zhang, Jie Zhang, Caigui Jiang, Yu Guo, Fei Wang

추상적인

We present a method named iComMa to address the 6D pose estimation problem in computer vision. The conventional pose estimation methods typically rely on the target's CAD model or necessitate specific network training tailored to particular object classes. Some existing methods address mesh-free 6D pose estimation by employing the inversion of a Neural Radiance Field (NeRF), aiming to overcome the aforementioned constraints. However, it still suffers from adverse initializations. By contrast, we model the pose estimation as the problem of inverting the 3D Gaussian Splatting (3DGS) with both the comparing and matching loss. In detail, a render-and-compare strategy is adopted for the precise estimation of poses. Additionally, a matching module is designed to enhance the model's robustness against adverse initializations by minimizing the distances between 2D keypoints. This framework systematically incorporates the distinctive characteristics and inherent rationale of render-and-compare and matching-based approaches. This comprehensive consideration equips the framework to effectively address a broader range of intricate and challenging scenarios, including instances with substantial angular deviations, all while maintaining a high level of prediction accuracy. Experimental results demonstrate the superior precision and robustness of our proposed jointly optimized framework when evaluated on synthetic and complex real-world data in challenging scenarios.

? Paper | 암호

표현:

Neural Radiance Fields (NeRFs) have demonstrated the remarkable potential of neural networks to capture the intricacies of 3D objects. By encoding the shape and color information within neural network weights, NeRFs excel at producing strikingly sharp novel views of 3D objects. Recently, numerous generalizations of NeRFs utilizing generative models have emerged, expanding its versatility. In contrast, Gaussian Splatting (GS) offers a similar renders quality with faster training and inference as it does not need neural networks to work. We encode information about the 3D objects in the set of Gaussian distributions that can be rendered in 3D similarly to classical meshes. Unfortunately, GS are difficult to condition since they usually require circa hundred thousand Gaussian components. To mitigate the caveats of both models, we propose a hybrid model that uses GS representation of the 3D object's shape and NeRF-based encoding of color and opacity. Our model uses Gaussian distributions with trainable positions (ie means of Gaussian), shape (ie covariance of Gaussian), color and opacity, and neural network, which takes parameters of Gaussian and viewing direction to produce changes in color and opacity. Consequently, our model better describes shadows, light reflections, and transparency of 3D objects.

? Paper | 암호

리뷰:

? 종이

쾅:

The integration of neural rendering and the SLAM system recently showed promising results in joint localization and photorealistic view reconstruction. However, existing methods, fully relying on implicit representations, are so resource-hungry that they cannot run on portable devices, which deviates from the original intention of SLAM. In this paper, we present Photo-SLAM, a novel SLAM framework with a hyper primitives map. Specifically, we simultaneously exploit explicit geometric features for localization and learn implicit photometric features to represent the texture information of the observed environment. In addition to actively densifying hyper primitives based on geometric features, we further introduce a Gaussian-Pyramid-based training method to progressively learn multi-level features, enhancing photorealistic mapping performance. The extensive experiments with monocular, stereo, and RGB-D datasets prove that our proposed system Photo-SLAM significantly outperforms current state-of-the-art SLAM systems for online photorealistic mapping, eg, PSNR is 30% higher and rendering speed is hundreds of times faster in the Replica dataset. Moreover, the Photo-SLAM can run at real-time speed using an embedded platform such as Jetson AGX Orin, showing the potential of robotics applications.

? 종이 | Project Page | 암호

부족한:

We introduce the Splatter Image, an ultra-fast approach for monocular 3D object reconstruction which operates at 38 FPS. Splatter Image is based on Gaussian Splatting, which has recently brought real-time rendering, fast training, and excellent scaling to multi-view reconstruction. For the first time, we apply Gaussian Splatting in a monocular reconstruction setting. Our approach is learning-based, and, at test time, reconstruction only requires the feed-forward evaluation of a neural network. The main innovation of Splatter Image is the surprisingly straightforward design: it uses a 2D image-to-image network to map the input image to one 3D Gaussian per pixel. The resulting Gaussians thus have the form of an image, the Splatter Image. We further extend the method to incorporate more than one image as input, which we do by adding cross-view attention. Owning to the speed of the renderer (588 FPS), we can use a single GPU for training while generating entire images at each iteration in order to optimize perceptual metrics like LPIPS. On standard benchmarks, we demonstrate not only fast reconstruction but also better results than recent and much more expensive baselines in terms of PSNR, LPIPS, and other metrics.

? Paper | Project Page | 코드 | ? Short Presentation

항해:

확장하다

awesome 3D gaussian splatting

멋진 3D 가우스 스플래팅 리소스

목차

3D 가우스 스플래팅을 소개하는 세미나 논문:

실시간 래디언스 필드 렌더링을 위한 3D 가우스 스플래팅

3D 객체 감지

2024년

1. 3DGS-DET: 3D 객체 감지를 위한 경계 안내 및 상자 중심 샘플링으로 3D 가우스 스플래팅 강화

자율주행:

2024년:

1. 역동적인 도시 장면 모델링을 위한 거리 가우시안

2. TCLC-GS: 주변 자율 주행 장면을 위한 긴밀하게 결합된 LiDAR-카메라 가우스 스플래팅

3. OmniRe: 옴니 도시 장면 재구성

2023년:

1. [CVPR '24] DrivingGaussian: 주변의 역동적인 자율 주행 장면을 위한 복합 가우시안 스플래팅

2. [CVPR '24] HUGS: 가우시안 스플래팅을 통한 전체적인 도시 3D 장면 이해

아바타:

2024년:

1. GaussianBody: 3D 가우스 스플래팅을 통한 옷을 입은 인간 재구성

2. PSAvatar: 3D 가우스 스플래팅을 사용한 실시간 머리 아바타 생성을 위한 점 기반 변형 가능한 모양 모델

3. Rig3DGS: 캐주얼 단안 비디오에서 제어 가능한 인물 사진 만들기

4. HeadStudio: 3D 가우스 스플래팅을 사용하여 텍스트를 애니메이션 가능한 머리 아바타로 변환

5. ImplicitDeepfake: NeRF 및 Gaussian Splatting을 사용한 암시적 Deepfake 생성을 통한 그럴듯한 얼굴 교환

6. GaussianHair: 빛 인식 가우시안을 사용한 헤어 모델링 및 렌더링

7. GVA : 단안 비디오에서 생생한 3D 가우시안 아바타 재구성

8.

9. Splatface : Gaussian Splat Face Rebonstruction 최적화 가능한 표면을 활용

10. 하하 : 텍스처 메쉬가있는 고도로 조음 된 가우스 인간 아바타

11.

12. Gomavatar : Gaussians-on-Mesh를 사용한 단안 비디오의 효율적인 애니메이션 인간 모델링

13. Occgaussian : 폐색 된 인간 렌더링을위한 3D 가우시안 스플릿

14. [CVPR '24] 보이지 않는 것 추측 : 부분 2D glimpses의 동적 3D 장면 재구성

15. [Neurips '24] 일반화 가능하고 애니메이션 가우스 헤드 아바타

16.

17.

2023년:

1. 운전 가능한 3D 가우스 아바타

2. Splatarmor : 단안 RGB 비디오에서 애니메이션 가능한 인간을위한 관절 가우시안 스플릿

3. [CVPR '24] 애니메이션 가우스 : 고 충실한 인간 아바타 모델링을위한 학습 자세 의존 가우스지도 학습

4. [CVPR '24] GART : GAUSSIAN 관절 템플릿 모델

5.

6. [CVPR '24] 포옹 : 인간 가우시안 splats

7. [CVPR '24] 효율적인 3D 인간 생성을위한 가우스 쉘 맵

8. Gaussianhead : 학습 가능한 가우스 파생물이있는 고 충실도 헤드 아바타

9. [CVPR '24] Gaussianavatars : 조작 된 3D 가우스가있는 광자적 헤드 아바타

10. [CVPR '24] GPS-Gaussian : 실시간 인간 소설 뷰 합성을위한 일반화 가능한 픽셀-3D 가우시안 스플릿

11. Gauhuman : 단안 인간 비디오에서 나온 가우시안 분열

12. HeadGaS: Real-Time Animatable Head Avatars via 3D Gaussian Splatting

13. [CVPR '24] HiFi4G: High-Fidelity Human Performance Rendering via Compact Gaussian Splatting

14. [CVPR '24] GaussianAvatar: Towards Realistic Human Avatar Modeling from a Single Video via Animatable 3D Gaussians

15. [CVPR '24] FlashAvatar: High-fidelity Head Avatar with Efficient Gaussian Embedding

16. [CVPR '24] Relightable Gaussian Codec Avatars

17. MonoGaussianAvatar: Monocular Gaussian Point-based Head Avatar

18. [CVPR '24] ASH: Animatable Gaussian Splats for Efficient and Photoreal Human Rendering

19. [CVPR '24] 3DGS-Avatar: Animatable Avatars via Deformable 3D Gaussian Splatting

20. [CVPR '24] GAvatar: Animatable 3D Gaussian Avatars with Implicit Mesh Learning

21. Deformable 3D Gaussian Splatting for Animatable Human Avatars

22. Human101: Training 100+FPS Human Gaussians in 100s from 1 View

23. [CVPR '24] Gaussian Head Avatar: Ultra High-fidelity Head Avatar via Dynamic Gaussians

24. HumanSplat: Generalizable Single-Image Human Gaussian Splatting with Structure Priors

Classic work:

1. A Generalization of Algebraic Surface Drawing

2. Approximate Differentiable Rendering with Algebraic Surfaces

3. Unbiased Gradient Estimation for Differentiable Surface Splatting via Poisson Sampling

4. Generating and Real-Time Rendering of Clouds

압축:

2024년:

1. [I3D '24] Reducing the Memory Footprint of 3D Gaussian Splatting

2. [CVPR '24] Compressed 3D Gaussian Splatting for Accelerated Novel View Synthesis

3. HAC: Hash-grid Assisted Context for 3D Gaussian Splatting Compression

4. [ECCV '24] End-to-End Rate-Distortion Optimized 3D Gaussian Representation

5. 3DGS.zip: A survey on 3D Gaussian Splatting Compression Methods

6. LapisGS: Layered Progressive 3D Gaussian Splatting for Adaptive Streaming

7. Implicit Gaussian Splatting with Efficient Multi-Level Tri-Plane Representation

2023년:

1. LightGaussian: Unbounded 3D Gaussian Compression with 15x Reduction and 200+ FPS

2. Compact3D: Compressing Gaussian Splat Radiance Field Models with Vector Quantization

3. [CVPR '24] Compact 3D Gaussian Representation for Radiance Field

4. [ECCV '24] Compact 3D Scene Representation via Self-Organizing Gaussian Grids

확산: