Donny Chen

img2threejs. Rebuild the object in a reference image as a code-only, procedural, quality-gated, animation-ready Three.js model. Token-efficient image-to-3D.

Canoverse. CanoVerse: 3D Object Scalable Canonicalization and Dataset for Generation and Pose (ECCV2026)

WorldOlympiad. WorldOlympiad: Can Your World Model Survive a Triathlon?

LatentSpatialMemory. Latent Spatial Memory for Video World Models

TriSplat. TriSplat: Simulation-Ready Feed-Forward 3D Scene Reconstruction

scenesmith. Code for "SceneSmith: Agentic Generation of Simulation-Ready Indoor Scenes", ICML 2026 Spotlight

MineCLIP. Foundation Model for MineDojo

TTM. Official Pytorch Implementation for "Time-to-Move: Training-Free Motion Controlled Video Generation via Dual-Clock Denoising"

World-R1. [ICML 2026] World-R1: Reinforcing 3D Constraints for Text-to-Video Generation

Seed2.0. Jupyter Notebook

nova3r. [ICLR 2026] NOVA3R: Non-pixel-aligned Visual Transformer for Amodal 3D Reconstruction

Awesome-Feed-Forward-3D. An curated list for feed-forward 3D scene modeling, including research directions, datasets, and applications.

genie_sim. Simulation Platform from AgiBot

editor. Create and share 3D architectural projects.

megaflow. MegaFlow: Zero-Shot Large Displacement Optical Flow

Motus. Official code of Motus: A Unified Latent Action World Model

gemini-image-skill. Equip Claude Code with Nano Banana - image generation & editing via Gemini API

resplat. ReSplat: Learning Recurrent Gaussian Splatting

Auto-claude-code-research-in-sleep. ARIS ⚔️ (Auto-Research-In-Sleep) — Lightweight Markdown-only skills for autonomous ML research: cross-model review loops, idea discovery, and experiment automation. No framework, no lock-in — works with Claude Code, Codex, OpenClaw, or any LLM agent.

ironclaw. IronClaw is an Agent OS focused on privacy, security and extensibility

amb3r. [CVPR'26 Highlight] AMB3R: Accurate Feed-forward Metric-scale 3D Reconstruction with Backend

MoVieS. [CVPR 2026] Official implementation of "MoVieS: Motion-Aware 4D Dynamic View Synthesis in One Second".

awesome-openclaw-usecases. A community collection of OpenClaw use cases for making life easier.

YoNoSplat. [ICLR'26] YoNoSplat: You Only Need One Model for Feedforward 3D Gaussian Splatting

SceneWeaver. Official implementation of NeurIPS 2025 paper "SceneWeaver: All-in-One 3D Scene Synthesis with an Extensible and Self-Reflective Agent"

PhysX-Anything. PhysX-Anything: Simulation-Ready Physical 3D Assets from Single Image (CVPR 2026)

CoV. [ACL 2026 Findings] CoV: Chain-of-View Prompting for Spatial Reasoning

g3splat. G3Splat: Geometrically Consistent Generalizable Gaussian Splatting

VolSplat. [ECCV 2026] VolSplat: Rethinking Feed-Forward 3D Gaussian Splatting with Voxel-Aligned Prediction

TRELLIS.2. Native and Compact Structured Latents for 3D Generation

FlexGEMM. A Cross-Platform Backend for High-Performance Sparse Convolutions

Splatent. Splatting Diffusion Latents for Novel View Synthesis [CVPR 2026]

UCPE. 📷 [CVPR'26] Camera-controlled text-to-video generation, now with intrinsics, distortion and orientation control!

gs-texturing. Code for the EGSR 2025 Paper "Content-Aware Texturing for Gaussian Splatting"

Pyramid-Sparse-Attention. Official PyTorch implementation of [PSA: Pyramid Sparse Attention for Efficient Video Understanding and Generation](https://arxiv.org/abs/2512.04025).

mesh-splatting. Python

JiT. PyTorch implementation of JiT https://arxiv.org/abs/2511.13720

NoPo3DFusion. 🏠 [AJCAI'25 Short Oral] NoPo3DFusion: Scaling Two Images to Long Videos via 3D-Aware Iterative Diffusion

PanFlow. 🍳 [AAAI'26] PanFlow: Decoupled Motion Control for Panoramic Video Generation

Depth-Anything-3. Depth Anything 3

VGGT-Long. Official implement of VGGT-Long

BBSplat. BillBoard Splatting (BBSplat): Learnable Textured Primitives for Novel View Synthesis

BrickGPT. [ICCV 2025 Best Paper] Official repository for BrickGPT, the first approach for generating physically stable toy brick models from text prompts.

TraceAnything. [ICLR 2026] Trace Anything: Representing Any Video in 4D via Trajectory Fields

IsaacLab. Unified framework for robot learning built on NVIDIA Isaac Sim

PM-Loss. [3DV 2026] Revisiting Depth Representations for Feed-Forward 3D Gaussian Splatting

ZPressor. [NeurIPS 2025] ZPressor: Bottleneck-Aware Compression for Scalable Feed-Forward 3DGS

lyra. Project Lyra: Open Generative 3D World Models

ReconViaGen. (ICLR2026) ReconViaGen: Towards Accurate Multi-view 3D Object Reconstruction via Generation

slam-handbook-public-release. Release repo for our SLAM Handbook

mvtracker. [ICCV 2025 Oral] MVTracker: Multi-view 3D Point Tracking

LichtFeld-Studio. Train, inspect, edit, automate, and export 3D Gaussian Splatting scenes from a single native application.

pixie. Feed-forward model for predicting 3D physics with 3DGS + NeRF

MeshSplat. [AAAI 2026] MeshSplat: Generalizable Sparse-View Surface Reconstruction via Gaussian Splatting

matchnerf. 🖨️ [TPAMI'25] Explicit Correspondence Matching for Generalizable Neural Radiance Fields

DanceGRPO. An official implementation of DanceGRPO: Unleashing GRPO on Visual Generation

Waver. Industry-level video foundation model for unified Text-to-Video (T2V) and Image-to-Video (I2V) generation.

Puppeteer. [NeurIPS 2025 Spotlight] Official repository for “Puppeteer: Rig and Animate Your 3D Models”

webdataset. A high-performance Python-based I/O system for large (and small) deep learning problems, with strong support for PyTorch.

iLRM. [CVPR 2026] iLRM: An Iterative Large 3D Reconstruction Model

meshtryoshka. Code for "Meshtryoshka: Differentiable Rendering of Real-World Scenes via Mesh Rasterization"

KnapFormer. Python

InteriorGS. InteriorGS: 3D Gaussian Splatting Dataset of Semantically Labeled Indoor Scenes

vfms. Python

ENeRF. SIGGRAPH Asia 2022: Code for "Efficient Neural Radiance Fields for Interactive Free-viewpoint Video"

SurfaceSplat. SurfaceSplat: Connecting Surface Reconstruction and Gaussian Splatting

prope. Cameras as Relative Positional Encoding

SpaTrackerV2. [ICCV 2025] SpatialTrackerV2: 3D Point Tracking Made Easy

Uni3C. Uni3C: Unifying Precisely 3D-Enhanced Camera and Human Motion Controls for Video Generation [Siggraph Asian 2025]

WAFT. [ICLR2026 - Oral] WAFT: Warping-Alone Field Transforms for Optical Flow

JAFAR. [NeurIPS 2025] Official code for JAFAR: Jack up Any Feature at Any Resolution

cosmos-transfer1-diffusion-renderer. Cosmos-Transfer1-DiffusionRenderer: High-quality video de-lighting and re-lighting based on Cosmos video diffusion framework

renderformer. Official Code Release for [SIGGRAPH 2025] RenderFormer: Transformer-based Neural Rendering of Triangle Meshes with Global Illumination

pow3r. Python

Cosmos-Drive-Dreams. Cosmos-Drive-Dreams: Scalable Synthetic Driving Data Generation with World Foundation Models

GEN3C. [CVPR 2025 Highlight] GEN3C: 3D-Informed World-Consistent Video Generation with Precise Camera Control

gen-omnimatte-public. Generative Omnimatte (CVPR 2025)

spark. :sparkles: An advanced 3D Gaussian Splatting renderer for THREE.js

triangle-splatting. Python

GigaTok. [ICCV 2025] Official repo for "GigaTok: Scaling Visual Tokenizers to 3 Billion Parameters for Autoregressive Image Generation"

MASt3R-SLAM. [CVPR 2025] MASt3R-SLAM: Real-Time Dense SLAM with 3D Reconstruction Priors

Hier-SLAM. 🌳 [ICRA'25] Hier-SLAM: Semantic Gaussian Splatting SLAM with Hierarchical Categorical Representation

Stable3DGen. A Modular Framework for 3D Generation and Beyond [WIP]

LVSM. [ICLR 2025 Oral] Official code for "LVSM: A Large View Synthesis Model with Minimal 3D Inductive Bias"

LongAlign. The official PyTorch implementation for Improving Long-Text Alignment for Text-to-Image Diffusion Models (LongAlign)

radfoam. Original implementation of "Radiant Foam: Real-Time Differentiable Ray Tracing"

vggt. [CVPR 2025 Best Paper Award] VGGT: Visual Geometry Grounded Transformer

lmm-r1. Extend OpenRLHF to support LMM RL training for reproduction of DeepSeek-R1 on multimodal tasks.

PCGS. 🪜 [AAAI 2026, Oral] Pytorch implementation of 'PCGS: Progressive Compression of 3D Gaussian Splatting'

mvdust3r. Open source impl of **MV-DUSt3R+ Single-Stage Scene Reconstruction from Sparse Views In 2 Seconds** from Meta Reality Labs. Project page https://mv-dust3rp.github.io/

LTX-Video. Official repository for LTX-Video

NSReg. Official PyTorch implementation of ICLR'25 paper "Open-Set Graph Anomaly Detection via Normal Structure Regularisation"

mega-sam. Code for the project "MegaSaM: Accurate, Fast and Robust Structure and Motion from Casual Dynamic Videos"

FLARE. Python

FeatUp. Official code for "FeatUp: A Model-Agnostic Frameworkfor Features at Any Resolution" ICLR 2024

CUT3R. Official implementation of Continuous 3D Perception Model with Persistent State

PromptDA. [CVPR 2025] Prompt Depth Anything

brush. 3D Reconstruction for all

PanSplat. 🍳 [CVPR'25] PanSplat: 4K Panorama Synthesis with Feed-Forward Gaussian Splatting

Long-LRM. Self-reimplemented version of Long-LRM.

DeepGfM. DeepImplementation of the NeurIPS 2025 paper: Gaussian from Motion: Exploring 3D Geometric Foundation Models for Gaussian Splatting

WonderWorld. Code release for https://kovenyu.com/WonderWorld/

MegaSynth. Code for MegaSynth: Scaling Up 3D Scene Reconstruction with Synthesized Data (CVPR 2025)

genesis-world. Simulation platform for general-purpose robotics & embodied AI learning.

edm2. EDM2 and Autoguidance -- Official PyTorch implementation

See3D. [CVPR'25 Highlight] You See it, You Got it: Learning 3D Creation on Pose-Free Videos at Scale

VAR. [NeurIPS 2024 Best Paper Award][GPT beats diffusion🔥] [scaling laws in visual generation📈] Official impl. of "Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Prediction". An *ultra-simple, user-friendly yet state-of-the-art* codebase for autoregressive image generation!

VideoX-Fun. 📹 A more flexible framework that can generate videos at any resolution and creates videos from images.

dataset.

IC-Light. More relighting!

mvsplat360. 🎞️ [NeurIPS'24] MVSplat360: Feed-Forward 360 Scene Synthesis from Sparse Views

MoGe. [CVPR'25 Oral] MoGe: Unlocking Accurate Monocular Geometry Estimation for Open-Domain Images with Optimal Training Supervision

PF3plat. [ICML'25] Official Implementation of "PF3plat: Pose-Free Feed-Forward 3D Gaussian Splatting"

NoPoSplat. [ICLR'25 Oral] No Pose, No Problem: Surprisingly Simple 3D Gaussian Splats from Sparse Unposed Images

blendify. Lightweight Python framework that provides a high-level API for creating and rendering scenes with Blender.

mochi. The best OSS video generation models, created by Genmo

depthsplat. [CVPR'25] DepthSplat: Connecting Gaussian Splatting and Depth

FCGS. :rocket: [ICLR 2025] Pytorch implementation of 'Fast Feedforward 3D Gaussian Splatting Compression'

diffusion-e2e-ft. [WACV'25 Oral] Fine-Tuning Image-Conditional Diffusion Models is Easier than You Think

spann3r. [3DV'25 Award Candidate] 3D Reconstruction with Spatial Memory

XMem. [ECCV 2022] XMem: Long-Term Video Object Segmentation with an Atkinson-Shiffrin Memory Model

splatt3r. Official repository for Splatt3R: Zero-shot Gaussian Splatting from Uncalibrated Image Pairs

ToonCrafter. [SIGGRAPH Asia 2024, Journal Track] ToonCrafter: Generative Cartoon Interpolation

vggsfm. VGGSfM: Visual Geometry Grounded Deep Structure From Motion

glomap. [DEPRECATED] GLOMAP - Global Structured-from-Motion Revisited

acezero. [ECCV 2024 - Oral] ACE0 is a learning-based structure-from-motion approach that estimates camera parameters of sets of images by learning a multi-view consistent, implicit scene representation.

flash3d. Flash3D: Feed-Forward Generalisable 3D Scene Reconstruction from a Single Image

shape-of-motion. Python

mast3r. Grounding Image Matching in 3D with MASt3R

LaRa. [ECCV 2024] Efficient Large-Baseline Radiance Fields, a feed-forward 2DGS model

croco. Python

RaDe-GS. RaDe-GS: Rasterizing Depth in Gaussian Splatting

MeshAnything. [ICLR 2025] From anything to mesh like human artists. Official impl. of "MeshAnything: Artist-Created Mesh Generation with Autoregressive Transformers"

SuperPoint. Efficient neural feature detector and descriptor

NeuralFusion. Python

SVDiff. Streaming Video Diffusion: Online Video Editing with Diffusion Models

EasyAnimate. 📺 An End-to-End Solution for High-Resolution and Long Video Generation Based on Transformer Diffusion

SpaTracker. [CVPR 2024 Highlight] Official PyTorch implementation of SpatialTracker: Tracking Any 2D Pixels in 3D Space

SEA-RAFT. [ECCV2024 - Oral, Best Paper Award Candidate] SEA-RAFT: Simple, Efficient, Accurate RAFT for Optical Flow

diff-gaussian-rasterization. Differentiable gaussian rasterization with depth, alpha, normal map and extra per-Gaussian attributes, also support camera pose gradient

DeepSeek-V2. DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model

gaussian_surfels. [SIGGRAPH'24] Implementations for "High-quality Surface Reconstruction using Gaussian Surfels".

Metric3D. The repo for "Metric3D: Towards Zero-shot Metric 3D Prediction from A Single Image" and "Metric3Dv2: A Versatile Monocular Geometric Foundation Model..."

feature-3dgs. [CVPR 2024 Highlight] Feature 3DGS: Supercharging 3D Gaussian Splatting to Enable Distilled Feature Fields

PhysDreamer. Code for PhysDreamer

flowmap. [3DV 2025] Code for "FlowMap: High-Quality Camera Poses, Intrinsics, and Depth via Gradient Descent" by Cameron Smith*, David Charatan*, Ayush Tewari, and Vincent Sitzmann

AnimateDiff. Official implementation of AnimateDiff.

latentsplat. [ECCV 2024] Implementation of latentSplat: Autoencoding Variational Gaussians for Fast Generalizable 3D Reconstruction

ParticleNeRF. A particle-based encoding for Neural Radiance Fields

PanFusion. 🍳 [CVPR'24 Highlight] Pytorch implementation of "Taming Stable Diffusion for Text to 360° Panorama Image Generation"

fast-gaussian-rasterization. A geometry-shader-based, global CUDA sorted high-performance 3D Gaussian Splatting rasterizer. Can achieve a 5-10x speedup in rendering compared to the vanialla diff-gaussian-rasterization.

HAC. :house: [ECCV 2024] Pytorch implementation of 'HAC: Hash-grid Assisted Context for 3D Gaussian Splatting Compression'

D-Persona. Official Code for our CVPR 2024 Paper "Diversified and Personalized Multi-rater Medical Image Segmentation" (Highlight)

LEBED. Pytorch implementation for ICLR24:"Online GNN Evaluation Under Test-Time Graph Distribution Shifts"

dust3r. DUSt3R: Geometric 3D Vision Made Easy

T-Stitch. [ICLR 2025] Official PyTorch implmentation of paper "T-Stitch: Accelerating Sampling in Pre-trained Diffusion Models with Trajectory Stitching"

Dataset. News: the 10k dataset is ready for download.

pixelsplat. [CVPR 2024 Oral, Best Paper Runner-Up] Code for "pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction" by David Charatan, Sizhe Lester Li, Andrea Tagliasacchi, and Vincent Sitzmann

GCVD-release. Python

evo. Python package for the evaluation of odometry and SLAM

Depth-Anything. [CVPR 2024] Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data. Foundation Model for Monocular Depth Estimation

Latte. [TMLR 2025] Latte: Latent Diffusion Transformer for Video Generation.

SIGNeRF. SIGNeRF: Scene Integrated Generation for Neural Radiance Fields

NeuralTexture. Unofficial implementation of the paper "Deferred Neural Rendering: Image Synthesis using Neural Textures" in Pytorch.

Deformable-3D-Gaussians. [CVPR 2024] Official implementation of "Deformable 3D Gaussians for High-Fidelity Monocular Dynamic Scene Reconstruction"

Free3D. [CVPR'24] Consistent Novel View Synthesis without 3D Representation

murf. [CVPR'24] MuRF: Multi-Baseline Radiance Fields

warp. A Python framework for GPU-accelerated simulation, robotics, and machine learning.

HumanGaussian. [CVPR 2024 Highlight] Code for "HumanGaussian: Text-Driven 3D Human Generation with Gaussian Splatting"

mip-splatting. [CVPR'24 Best Student Paper] Mip-Splatting: Alias-free 3D Gaussian Splatting

Compact-3DGS. The official repository of Compact 3D Gaussian Representation for Radiance Field

MPVSS. Python

PL-NeRF. NeRF Revisited: Fixing Quadrature Instability in Volume Rendering, Neurips 2023

ZeroNVS. Python

zipnerf-pytorch. Unofficial implementation of ZipNeRF

zero123plus. Code repository for Zero123++: a Single Image to Consistent Multi-view Diffusion Base Model.

gsplat. CUDA accelerated rasterization of gaussian splatting

S3IM-Neural-Fields. [ICCV 2023] Pytorch implementation of "S3IM: Stochastic Structural SIMilarity and Its Unreasonable Effectiveness for Neural Fields".

Dynamic3DGaussians. Python

DFM. Implementation of "Diffusion with Forward Models: Solving Stochastic Inverse Problems Without Direct Supervision"

HyperHuman. [ICLR 2024] Github Repo for "HyperHuman: Hyper-Realistic Human Generation with Latent Structural Diffusion"

SFGC. Pytorch implementation of NeurIPS-23:"Structure-free Graph Condensation (SFGC): From Large-scale Graphs to Condensed Graph-free Data"

efficient-stable-diffusion.

objectsdf_plus. :first_quarter_moon: [ICCV'23] Pytorch implementation of "ObjectSDF++: Improved Object-Compositional Neural Implicit Surfaces"

Awesome-Data-Centric-GraphML. A collection of papers and resources about Data Centric Graph Machine Learning (DC-GML)

CVQ-VAE. [ICCV 2023] Online Clustered Codebook

co-tracker. CoTracker is a model for tracking any point (pixel) on a video.

omnimotion. Python

unimatch. [TPAMI'23] Unifying Flow, Stereo and Depth Estimation

Awesome-Object-Compositional-INR. :book: A collection of object-compositional modeling by implicit neural representation.

mlp_maps. Code for "Representing Volumetric Videos as Dynamic MLP Maps" CVPR 2023

differentiable-blocksworld. [NeurIPS 2023] Code for "Differentiable Blocks World: Qualitative 3D Decomposition by Rendering Primitives"

gaussian-splatting. Original reference implementation of "3D Gaussian Splatting for Real-Time Radiance Field Rendering"

CoactSeg. Official Code for our MICCAI 2023 paper "CoactSeg: Learning from Heterogeneous Data for New Multiple Sclerosis Lesion Segmentation"

SN-Netv2. [ECCV 2024] This is the official implementation of "Stitched ViTs are Flexible Vision Backbones".

NeuSC. A Temporal Voyage: Code for "Neural Scene Chronology" [CVPR 2023]

infinigen. Infinite Photorealistic Worlds using Procedural Generation

dynibar. Implementation of DynIBaR Neural Dynamic Image-Based Rendering (CVPR 2023)

cop3d. Common Pets in 3D

suds. Scalable Urban Dynamic Scenes