awesome-gaussian-skills

Awesome Gaussian Skills

The Most Comprehensive 3D Gaussian Splatting Catalog — 900 methods, 23 Categories, Interactive Explorer

You shouldn't search 20 repos for 3DGS papers. This is the only one you need.

English | 中文

Why This Repo?

Other awesome lists give you paper titles. We give you paper titles + an AI toolkit that makes you faster.

What You Need	Other Lists	This Repo
Browse papers	Static markdown table	Interactive explorer: search, filter, sort
Compare methods	Open 2 papers side by side	10+ dimension auto-comparison
Avoid code bugs	Discover after submission	104 known bug pattern detection
Design experiments	Guess baselines & ablations	Venue-tailored experiment plan
NeRF → 3DGS	Trial-and-error porting	Step-by-step migration guide
CAD → 3DGS	No coverage	40+ method conversion pipeline
Patent filing	Manual from scratch	Auto-generated claims & specs

Live Demo

Try the Interactive Method Explorer →

Search 900 Methods instantly, filter by category, sort by citations, click any method card for details.

Integrations

This repo ships as both AI agent skills and an MCP server, so it works across agent runtimes and MCP clients:

Client	How to use
DSH (DeepSeek Harness)	dsh plugin --profile web add jaccen/Awesome-Gaussian-Skills — the 15 skills + MCP renderer auto-mount via cordis.patch.yml
Claude Code / OpenClaw	skills load directly from skills/ (see CLAUDE.md); MCP server via the config below
Claude Desktop / Cursor / Cline	copy the MCP stdio config from .marketplace/mcp-servers/client-configs.md
Any MCP client	command: node, args: [/path/to/mcp-server/dist/index.js], transport: stdio (21 core tools)

The MCP server exposes 21 tools (import_scene, set_camera, modify_gaussians, render_frame, sculpt_pipeline, SLAT latent editing, …) — see mcp-server/README.md and .marketplace/mcp-servers/mcp-metadata.json.

📖 Online Book: Spatial & Embodied Intelligence (New!)

** NEW (Jul 2026)** —A full open-source technical book, built around 3D Gaussian Splatting as the spine and weaving together spatial intelligence and embodied intelligence into one closed loop: representation → perception → planning → action.

📖 Read the Book →

Core formula (echoing Agent = LLM + Context + Tools):

Embodied Agent = Spatial Representation × Perception × Planning × Action

What's inside — 13 chapters + hands-on prelude & capture SOP, every method name anchored to this repo's real data (900 Methods, 23 categories, 15 skills), zero fabrication:

#	Chapter	Focus
引言	Why this book	Why 3DGS is the key puzzle piece of Physical AI
前	Landing prelude: should you, can you	Tech-selection matrix, eight cases where you should NOT use 3DGS, three reading paths
CH 01	NeRF → 3DGS: A paradigm leap	Explicit vs implicit, the three innovations, the alpha-compositing formula
CH 02	The math & engineering core	Anisotropic Gaussians, differentiable rasterization, adaptive density control, CUDA
操	Data-capture SOP	Device selection, trajectory design, seven exposure hard rules, scale calibration —the first step that decides your quality ceiling
CH 03	From scene to world	Large-scale, dynamic/4D, GS-SLAM, compression & deployment
CH 04	Semantic Gaussians	CLIP/DINO feature distillation, open-vocabulary 3D segmentation
CH 05	Editing · Generation · Asset-ization	Feed-forward reconstruction, SDS generation, animatable assets, PBR relighting
CH 06	Embodied intelligence basics	VLA lineage (RT/π0/GR00T/ReconVLA), simulation, Sim2Real
CH 07	3DGS as robot spatial memory	GS-SLAM, map-as-renderer, three tiers of spatial memory
CH 08	Object-level & articulated understanding	Part-level Gaussians, URDF bridging, the CAD·Mesh·3DGS triangle
CH 09	Agent-driven digital twins	MCP rendering pipeline, gesture interaction, the perception-action loop
CH 10	World models & the future	Six schools of world models, 3DGS×World Model, spatial foundation models, Physical AI
CH 11	Security, provenance & copyright	3DGS watermarking, GaussTrace forensics, adversarial attacks, IP management
CH 12	Capture-to-delivery: the end-to-end pipeline	Ten-stage tool selection & pitfalls, formats & coordinate systems, cost/duration estimates, project failure modes
CH 13	Industry playbooks: six verticals	Museums & tourism, e-commerce, industrial inspection, AD simulation, embodied robots, surveying & mapping
附	Appendix	Datasets, evaluation metrics, delivery-metric conversion & acceptance-report templates
实	Practice index: from book to engineering	15 engineering skills · 34 MCP tools · 900+ method explorer, chapter-to-runnable-resource mapping
后记	Will 3DGS be eaten?	Why explicit representations will be compressed, not consumed

Each chapter ends with hands-on exercises and links back to the repo's method tables, references/, and skills —so reading the book and doing the engineering are one seamless flow.

Highlights you won't find in a paper list:

• The six schools of world models (2026 taxonomy) and where 3DGS sits as the only representation that is simultaneously renderable, differentiable, and editable.
• How GS-World, ManiGaussian, and OrbiSim turn 3DGS into a differentiable simulation engine.
• A three-tier model of robot spatial memory (geometric → appearance → semantic) and where current GS-SLAM actually stands.

Why we wrote it (and how it relates to this repo)

This book is the narrative layer over the repo's data layer. The repo gives you 900 method names, abstracts, and 15 engineering skills —but not the through-line that connects them. The book supplies that through-line: it argues why 3DGS became the bridge between spatial intelligence and embodied intelligence, and walks every chapter back to concrete methods and skills you can use today. Read the book to understand the map; use the repo to ship the territory.

What's New (Oct 2026)

Latest Update (Oct 20, 2026): v0.9.8 — World Models Category Reinforcement (896 → 900) ⭐

Targeted reinforcement of the thinnest category. 4 new arXiv-verified 3DGS-based World Model / Spatial Intelligence methods admitted, taking the library from 896 to 900. The World Models & Spatial Intelligence category grew from 8 → 12 methods: HY-World 2.0, F3DGS (federated multi-agent world modeling), 4DGS-WAM (object-centric world action model), Cloth-Splatting (3DGS cloth state estimation). Two missing arXiv IDs backfilled (Holi-Spatial → 2603.07660, Spatial-TTT → 2603.12255); two entries flagged unverified pending source check (APEIRIA, S²AM3D). GSMem found already in library under Embodied AI & Robotics — not duplicated. See changelog.

Previous Update (Oct 12, 2026): v0.9.7 — Manual Admission: 15 New Methods (881 → 896) ⭐

Manual admission of 15 new arXiv-verified 3DGS methods, taking the library from 881 to 896 (+15, +1.7%). Admissions span 9 categories: Acceleration (H2-Ray, TileSkipper, SplatQ), Feed-Forward (CompactSplat, DeltaSplat, DensiTok), Embodied AI & Robotics (Splat2Act, 2DGS-Planner, NovaSplat), plus OuroWorld (Dynamic & 4D), PAM-ToD (HDR & Relighting), FlyMark (Security), GSCV (Compression & Streaming), OpenSplatGraph (Language & Semantic), PCAsplat (Surface & Rendering). Every admitted method name is self-stated in its own paper's title/abstract (R9 protocol) and arXiv-verified. 12 candidate papers were rejected for lacking a self-stated method name, being benchmark-only, NeRF (not 3DGS), or using 3DGS merely as a tool. All data carriers (methods.json / CSV / methods.html / abstracts.js / reference doc) re-verified at 896, name sets compared entry by entry. No method name was fabricated.

See changelog/2026-10-12.md for the full admission list.

Update (Oct 7, 2026): v0.9.6 — Manual Admission: 9 New Methods (872 → 881) ⭐

Manual admission of 9 new arXiv-verified 3DGS methods, taking the library from 872 to 881 (+9, +1.0%). Admission includes Speedy-Splat (CVPR 2025, precise tile intersection + efficient Hessian-based pruning, 6.71× render speedup) and 8 recent September 2026 papers: AESplat (pose-free feed-forward), AGILE-GS (anchor-guided NBV), OC-GS (irregular turntable), ClearGS (handheld video), ChronoFuseGS (multi-temporal fusion, PG 2026), TangoGS (cross-scale compact), RRTO-CF3DGS (COLMAP-free), Observation-Gram (SH compression). All data carriers (methods.json / CSV / methods.html / abstracts.js / reference doc) re-verified at 881, name sets compared entry by entry. No method name was fabricated.

See changelog/2026-10-07.md for the full admission list.

Update (Sep 21, 2026): v0.9.5 — Daily arXiv Scan: 13 New Methods (859 → 872) ⭐

Routine daily arXiv scan admitted 13 new arXiv-verified 3DGS methods, taking the library from 859 to 872 (+13, +1.5%). Admissions span Surface & Rendering, Feed-Forward, Large-Scale, Compression, Sparse-View, Cross-Domain, Embodied AI & Robotics, Dynamic, and Language & Semantic categories (incl. LiteTex-GS, GrapeSplat, VDGS, CRP-GS, NormLift). 7 candidate papers were rejected for failing the verification protocol — benchmarks without a self-stated method name, descriptive-only paper titles, or 3DGS used merely as a tool. All data carriers (methods.json / CSV / methods.html / abstracts.js / reference doc) re-verified at 872, name sets compared entry by entry. No method name was fabricated; the library's "every entry arXiv-verifiable" guarantee holds.

See changelog/2026-09-21-v3.md for the full admission list.

Earlier updates (v0.9.4 and before)

Latest Update (Sep 21, 2026): v0.9.4 — Multi-Channel Harvest: 700 → 859 Methods + Accuracy Overhaul ⭐

Knowledge base expanded from 700 to 859 verified methods (+159, +22.7%) through a multi-channel arXiv harvest over the last 150 days. Six query families (cs.CV / cs.GR / cs.RO / full-corpus / abs:"3D Gaussian Splatting" / ti:"Gaussian Splatting") returned 395 candidates; after verification 162 were admitted. Also in this release:

• Verification protocol (R1–R6): every admission requires the paper's self-stated method name, reproduced in its own abstract; multi-word names must contain a proper-noun token — this alone rejected 67 descriptions masquerading as method names (Capacity-Controlled, Scene-Level). 114 borderline candidates were read individually; 25 adjudicated in, the rest dropped.
• Accuracy repairs: 57 reference links had a displayed arXiv ID differing from the hyperlinked one (all had the correct ID displayed, wrong link). 13 mis-attributed entries deleted — e.g. Dynamic 3D Gaussians pointed at 2309.13114, which is actually a magnetic spin-glass physics paper, and GS-LRM-v2 pointed at DOF-GS. All now verified against the arXiv API.
• AIGC traces removed from 29 text files; 21 historically corrupted JPEGs restored and re-stripped safely.
• README_CN.md recovered from double-encoding corruption that had mangled every Chinese string.
• Full carrier sync — methods.json / CSV / methods.html / abstracts.js / reference doc all verified at 881, name sets compared entry by entry rather than by count alone.

Adjudication wave (same day). A second pass resolved every name carrying a derived suffix or colliding with another entry: Scaffold-GS+ and GaussianSplatting-SLAM-v2 deleted (no such paper exists), Luminance-GS++ renamed to its self-declared base name, and Spark 2.0's description corrected — it is World Labs' open-source WebGL2 splat renderer, not an NVIDIA robotics simulator. Two independent papers both self-name GeoGS-SLAM (2607.07452 and 2607.11184, disjoint author lists); they are two methods, not a sequel, and are now disambiguated. 862 → 859.

See changelog/2026-09-21.md and changelog/2026-09-21-v2.md for the full audit trail.

Latest update (Sep 20): v0.9.2 —Scene Agent + Mapmost Engineering Wave: 5 New Methods + 6 Skill Enhancements. Knowledge base expanded from 692→ 97 methods. Added 5 scene-generation Agent papers: SceneSmith (ICML 2026 Spotlight, MIT CSAIL —hierarchical designer-critic-orchestrator VLM agents for simulation-ready indoor scenes), Scenethesis (ICLR 2026, NVIDIA —training-free LLM+vision agentic framework), SceneGenAgent (ACL 2025 —LLM coding agent for industrial scenes), SceneConductor (arXiv —multi-agent orchestration from single image), SceneAssistant (arXiv —visual-feedback agent for open-vocabulary scene generation). Injected Mapmost production engineering knowledge into 4 skill files: 3DGS individualization (单体化) in best-practices.md, LOD streaming + seam-fix + adaptive partition training in best-practices.md, 3DGS+3DTiles multi-source fusion in gis-toolchain.md, LUT color grading in tech-stack.md, Mapmost as Digital Twin player with 100km² benchmark in industry-landscape.md. Categories updated: Generation +4, Embodied AI & Robotics +1.

Previous (Sep 20): v0.9.1 —ECCV 2026 Wave: 14 New Methods. Knowledge base expanded from 678→ 92 methods. All 14 new entries arXiv-verified or ECCV-official: Wat3R (underwater 3D geometry without annotations), GeoNVS (geometry-grounded video diffusion with GS-Adapter), Edit3r (instant 3D editing from unposed images), SA-ResGS (first residual learning for 3DGS in next-best-view selection), GaussianLens (on-demand localized high-res reconstruction), Deformable Triangle Splatting (non-convex primitive with per-edge control points), Neural Harmonic Textures (per-primitive harmonic feature decoding), CubicSplat (error-bounded differentiable vector graphics), TetraSDF (analytic isosurface extraction), Heat Kernel Textures (ECCV 2026 Best Paper —geodesic Gaussians), **GR