Zihan Wang

sage. Official Code Release of SAGE: Scalable Agentic 3D Scene Generation for Embodied AI

AnyRecon. [SIGGRAPH Asia 2026] AnyRecon: Arbitrary-View 3D Reconstruction with Video Diffusion Model

GeoWorld. [ECCV 2026] GeoWorld: Providing Full-frame Geometry Features to Facilitate 3D Scene Generation

worldmesh. [ECCV 2026] WorldMesh: Generating Navigable Multi-Room 3D Scenes via Mesh-Conditioned Image Diffusion

Spatia. [CVPR2026] Long-horizon, spatially consistent video generation enabled by persistent 3D scene point clouds and dynamic-static disentanglement.

PixWorld. PixWorld: Unifying 3D Scene Generation and Reconstruction in Pixel Space

FlexWorld. Official PyTorch implementation for "FlexWorld: Progressively Expanding 3D Scenes for Flexiable-View Synthesis".

LatentSpatialMemory. Latent Spatial Memory for Video World Models

GaussianGPT. [ECCV'26] Our method creates 3D Gaussian scenes completely autoregressively, allowing for generation, completion, and outpainting with the same model.

G4Splat. Official implementation of ICLR26 paper "G4Splat: Geometry-Guided Gaussian Splatting with Generative Prior"

robot-PointAct. Official implementation of PointACT: Vision-Language-Action Models with Multi-Scale Point-Action Interaction (RSS'26).

Moebius. [ECCV 2026] Moebius: 0.2B Lightweight Image Inpainting Framework with 10B-Level Performance

Prompt-Inpaint. An end-to-end pipeline for text-prompt detection, SAM3 segmentation, and background inpainting. 文本提示检测 + SAM3 分割 + 背景修复（inpaint）的端到端管线

drifting-model. Personal PyTorch implementation of "Generative Modeling via Drifting" with Claude

DriftingModel. PyTorch implementation of Drifting Models by Kaiming He et al.

MV-SAM3D. SAM 3D Objects with Multi-view Images

motrixsim-docs. A high-performance physics simulation engine designed for multibody dynamics and robotics simulation.

gs_playground. Jupyter Notebook

lingbot-vision. Self-supervised learning for spatial perception

Argus. [ECCV 2026] Argus: Metric Panoramic 3D Reconstruction for Indoor Scenes

ArtiFixer. Python

robocasa. RoboCasa: Large-Scale Simulation of Everyday Tasks for Generalist Robots

genie_sim. Simulation Platform from AgiBot

recgen. Jupyter Notebook

stretch_mujoco. This library provides a simulation stack for Stretch, built on MuJoCo.

HomeWorld.

GN0. The official Implementation of GN0: Toward a Unified Paradigm for Generation, Evaluation, and Policy Learning in Visual-Language Navigation

image-blaster. An image-to-world skillset for Claude.

lingbot-map. A feed-forward 3D foundation model for reconstructing scenes from streaming data

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/

AnySplat. [SIGGRAPH Asia 2025 (ACM TOG)] AnySplat: Feed-forward 3D Gaussian Splatting from Unconstrained Views

splatter360. [CVPR 2025] Splatter-360: Generalizable 360 Gaussian Splatting for Wide-baseline Panoramic Images

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

molmospaces. An end-to-end open ecosystem for robot learning

AgentVLN.

PointWorld. PointWorld: Scaling 3D World Models for In-The-Wild Robotic Manipulation

pMF. Official Implementation of pMF https://arxiv.org/abs/2601.22158

lingbot-vla. A Pragmatic VLA Foundation Model

imeanflow. Official Implementation of iMF https://arxiv.org/abs/2512.02012

meanflow. JAX implementation of MeanFlow

linkerhand-python-sdk. Linkerhand Python SDK

diff-gaussian-feature-rasterization. Differentiable Gaussians' Features Rasterization Submodule

vdpm. Official implementation of Video-DPM

DexTeleGym. After modifications on OpenTelevision, Quest 3 can be used for teleoperation of the Franka Panda robotic arm and Inspire Hand in Isaac Gym.

inspire_hands. working with the inspire robot hands (RH56 series)

inspire_demos. Python

Action100M. A Large-scale Video Action Dataset

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

lightning-grasp. The Lightning Grasp System

AstraNav-World. Official implementation of [AstraNav-World: World Model for Foresight Control and Consistency]

OmniNav. 【ICLR 2026】 Official implementation of [OmniNav: A Unified Framework for Prospective Exploration and Visual-Language Navigation]

docker-selkies-egl-desktop. KDE Plasma Desktop container designed for Kubernetes, supporting OpenGL EGL and GLX, Vulkan, and Wine/Proton for NVIDIA GPUs through WebRTC and HTML5, providing an open-source remote cloud/HPC graphics or game streaming platform.

IAmGoodNavigator. Python

RealSee3D. RealSee3D: A multi-view RGB-D dataset combining real-world captures and procedurally generated scenes, with extensible annotations for diverse 3D vision research.

SID-VLN. [ECCV 2026] Learning Goal-Oriented Language-Guided Navigation with Self-Improving Demonstrations at Scale

Depth-Anything-3. Depth Anything 3

DiT360. [CVPR 2026] Official implementation of "DiT360: High-Fidelity Panoramic Image Generation via Hybrid Training".

JanusVLN. [ICLR2026] Official implementation for "JanusVLN: Decoupling Semantics and Spatiality with Dual Implicit Memory for Vision-Language Navigation"

se3ds. This repository hosts the code for our paper, "Simple and Effective Synthesis of Indoor 3D Scenes".

world-in-world. "World Models in a Closed-Loop World" (ICLR'26 Oral)

Qwen3-VL. Qwen3-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.

grounded-rl. Python

TTT3R. [ICLR 2026] A simple state update rule to enhance length generalization for CUT3R

MTU3D. Python

Nav-R1. Nav-R1: Reasoning and Navigation in Embodied Scenes

3DLLM-Mem.

GraspMolmo. Code and website for "GraspMolmo: Generalizable Task-Oriented Grasping via Large-Scale Synthetic Data Generation"

Holodeck. CVPR 2024: Language Guided Generation of 3D Embodied AI Environments.

Vid2Sim. [CVPR 25] Vid2Sim: Realistic and Interactive Simulation from Video for Urban Navigation

SceneCompleter. SceneCompleter: Dense 3D Scene Completion for Generative Novel View Synthesis

thortils. Utility functions when working with Ai2-THOR. Try to do one thing once.

A0. Python

RynnEC. RynnEC: Bringing MLLMs into Embodied World

scaling_on_scales. When do we not need larger vision models?

dinov3. Reference PyTorch implementation and models for DINOv3

VLN_CLASH. This is the official repository for VLN-CLASH.

DifNav. This is the source code to paper “DAgger Diffusion Navigation: DAgger Boosted Diffusion Policy for Vision-Language Navigation”.

Being-H. Being-H is BeingBeyond's family of human-centric embodied foundation models.

VPN. Official implementation of "VPN: Visual Prompt Navigation"

SmartCLIP. SmartCLIP: A training method to improve CLIP with both short and long texts

habitat-sim. A flexible, high-performance 3D simulator for Embodied AI research.

InternNav. InternRobotics' open platform for building generalized navigation foundation models.

stretch_isaacsim.

LOVMM. LOVMM

Pi3. [ICLR 2026] π^3: Permutation-Equivariant Visual Geometry Learning

objaverse-xl. 🪐 Objaverse-XL is a Universe of 10M+ 3D Objects. Contains API Scripts for Downloading and Processing!

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

MoMaKitchen. [ICCV 2025] MoMa-Kitchen: A 100K+ Benchmark for Affordance-Grounded Last-Mile Navigation in Mobile Manipulation

StreamVLN. [ICRA 2026] Official implementation of the paper: "StreamVLN: Streaming Vision-and-Language Navigation via SlowFast Context Modeling"

NaVILA. [RSS'25] This repository is the implementation of "NaVILA: Legged Robot Vision-Language-Action Model for Navigation"

YOTO. [TPAMI2026 / RSS2025] Code for my paper "You Only Teach Once: Learn One-Shot Bimanual Robotic Manipulation from Video Demonstrations"

LAPA. [ICLR 2025] LAPA: Latent Action Pretraining from Videos

mobipi. Official implementation for Mobi-π.

NavMorph. Official implementation of NavMorph: A Self-Evolving World Model for Vision-and-Language Navigation in Continuous Environments (ICCV'25).

TrackVLA. [CoRL 2025] Repository relating to "TrackVLA: Embodied Visual Tracking in the Wild"

NavDP. [ICRA 2026] NavDP: Learning Sim-to-Real Navigation Diffusion Policy with Privileged Information Guidance

Object2HabitatMap. Awesome habitat top down map work 🤩

Uni-NaVid. [RSS 2025] Uni-NaVid: A Video-based Vision-Language-Action Model for Unifying Embodied Navigation Tasks.

instant_policy.

brs-algo. Official Algorithm Codebase for the Paper "BEHAVIOR Robot Suite: Streamlining Real-World Whole-Body Manipulation for Everyday Household Activities"

transformers. 🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.

Isaac-GR00T. NVIDIA Isaac GR00T N1.7 - A Foundation Model for Generalist Robots.

HA-VLN. Official implementation for "HA-VLN 2.0: An Open Benchmark and Leaderboard for Human-Aware Navigation in Discrete and Continuous Environments with Dynamic Multi-Human Interactions".

stretch_community. Portal for resources for the Stretch community

TinySAM. [AAAI 2025] Official PyTorch implementation of "TinySAM: Pushing the Envelope for Efficient Segment Anything Model"

EMMOE. EMMOE: A Comprehensive Benchmark for Embodied Mobile Manipulation in Open Environments

MSR3D. [NeurIPS 2024] MSR3D: Multimodal Situated Reasoning in 3D Scenes

procthor. 🏘️ Scaling Embodied AI by Procedurally Generating Interactive 3D Houses

multiscan. [NeurIPS 2022] MultiScan: Scalable RGBD scanning for 3D environments with articulated objects

SoFar. [NeurIPS 2025 Spotlight] SoFar: Language-Grounded Orientation Bridges Spatial Reasoning and Object Manipulation

OpenFly-Platform. C++

academicpages.github.io. Github Pages template based upon HTML and Markdown for personal, portfolio-based websites.

Uni-NaVid. JavaScript

FLaRe. [ICRA 25] FLaRe: Achieving Masterful and Adaptive Robot Policies with Large-Scale Reinforcement Learning Fine-Tuning

mpao. Python

stretch-open. Python

SegmentAnything3D. [ICCV'23 Workshop] SAM3D: Segment Anything in 3D Scenes

FastSAM. Fast Segment Anything

anygrasp_sdk. Python

ok-robot. An open, modular framework for zero-shot, language conditioned pick-and-drop tasks in arbitrary homes.

spoc-robot-training. SPOC: Imitating Shortest Paths in Simulation Enables Effective Navigation and Manipulation in the Real World

RVT. Official Code for RVT-2 and RVT

ManiSkill. Manipulation Skill Framework, an open source GPU parallelized robotics simulator and benchmark

stretch_urdf. URDFs for the Stretch mobile manipulators from Hello Robot Inc.

EmbodiedScan. [CVPR 2024 & NeurIPS 2024] EmbodiedScan: A Holistic Multi-Modal 3D Perception Suite Towards Embodied AI

openpi. Python

Video-3D-LLM. [CVPR 2025] The code for paper ''Video-3D LLM: Learning Position-Aware Video Representation for 3D Scene Understanding''.

VLN-CE. Vision-and-Language Navigation in Continuous Environments using Habitat

rectified-flow. 从零手搓Flow Matching（Rectified Flow）

ESAM. [ICLR 2025, Oral] EmbodiedSAM: Online Segment Any 3D Thing in Real Time

stretch_visual_servoing. Example code for visual servoing using Stretch 3's gripper camera

dobb-e. Dobb·E: An open-source, general framework for learning household robotic manipulation

harmonic_mobile_manipulation. Python

legged-loco. Low-level locomotion policy training in Isaac Lab

NaVILA-Bench. Vision-Language Navigation Benchmark in Isaac Lab

NaVid-VLN-CE. [RSS 2024 & RSS 2025] VLN-CE evaluation code of NaVid and Uni-NaVid

mshab. A Benchmark for Low-Level Manipulation in Home Rearrangement Tasks

UnSAM. [NeurIPS 2024] Code release for "Segment Anything without Supervision"

LightRAG. [EMNLP2025] "LightRAG: Simple and Fast Retrieval-Augmented Generation"

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

forcesight. Given an RGBD image and a text prompt, ForceSight produces visual-force goals for a robot, enabling mobile manipulation in unseen environments with unseen object instances.

open-eqa. OpenEQA Embodied Question Answering in the Era of Foundation Models

Embodied_RAG. Python

CLIP-SAM. Experiment on combining CLIP with SAM to do open-vocabulary image segmentation.

stretch_ai. Python

InfiniteWorld. Python

VLN-SRDF. Official implementation of: Bootstrapping Language-Guided Navigation Learning with Self-Refining Data Flywheel

SAME. [ICCV 2025] Official implementation of SAME: Learning Generic Language-Guided Visual Navigation with State-Adaptive Mixture of Experts

point_lio_unilidar. Point-LIO algorithm for Unitree LiDAR products.

vrb. Python

MapGPT. [ACL 24] The official implementation of MapGPT: Map-Guided Prompting with Adaptive Path Planning for Vision-and-Language Navigation.

open_x_embodiment. Jupyter Notebook

LLaVA-3D. [ICCV 2025] A Simple yet Effective Pathway to Empowering LLaVA to Understand and Interact with 3D World

OpenING. Official Implementation of OpenING: A Comprehensive Benchmark for Judging Open-ended Interleaved Image-Text Generation

AdaVLN. IsaacSim Extension for Dynamic Objects in Matterport3D Environments for AdaVLN research

feature-splatting-inria. Feature splatting based on INRIA GS rasterizer

concept-graphs. Official code release for ConceptGraphs

VLN-Game. A new zero-shot framework to explore and search for the language descriptive targets in unknown environment based on Large Vision Language Model.

PoliFormer. PoliFormer: Scaling On-Policy RL with Transformers Results in Masterful Navigators

partnr-planner. A repository accompanying the PARTNR benchmark for using Large Planning Models (LPMs) to solve Human-Robot Collaboration or Robot Instruction Following tasks in the Habitat simulator.

SummaryOfLoanSuspension. 全国各省市停贷通知汇总

PIVOT-R. [NeurIPS 2024] PIVOT-R: Primitive-Driven Waypoint-Aware World Model for Robotic Manipulation

graphrag. A modular graph-based Retrieval-Augmented Generation (RAG) system

rsrd. Code for "Robot See Robot Do" presented at CoRL 2024!

learning-language-navigation. Official repository for LeLaN training and inference code

Chat-Scene. [NeurIPS 2024 & TPAMI 2026] Chat-Scene: Bridging 3D Scene and Large Language Models with Object Identifiers

RoboticsDiffusionTransformer. RDT-1B: a Diffusion Foundation Model for Bimanual Manipulation

DigitalTwinArt. Python

ovon. Open Vocabulary Object Navigation

carbonyl. Chromium running inside your terminal

bypy. Python client for Baidu Yun (Personal Cloud Storage) 百度云/百度网盘Python客户端

dinov2. PyTorch code and models for the DINOv2 self-supervised learning method.

openvla. OpenVLA: An open-source vision-language-action model for robotic manipulation.

e2e-NeRF-nav.

InstructNav. Python

open_clip. An open source implementation of CLIP.

habitat_tools. python tools to work with habitat-sim environment.

ScanReason. [ECCV 2024] Empowering 3D Visual Grounding with Reasoning Capabilities

VLA-3D. Python

vil3dref. Official implementation of Language Conditioned Spatial Relation Reasoning for 3D Object Grounding (NeurIPS'22).

SQA3D. [ICLR 2023] SQA3D for embodied scene understanding and reasoning

feature-splatting. Python

Long-CLIP. [ECCV 2024] official code for "Long-CLIP: Unlocking the Long-Text Capability of CLIP"

PQ3D. Official implementation of the paper "Unifying 3D Vision-Language Understanding via Promptable Queries"

Director3D. Code for "Director3D: Real-world Camera Trajectory and 3D Scene Generation from Text" (NeurIPS 2024).

BEVInstructor. [ECCV24] Navigation Instruction Generation with BEV Perception and Large Language Models

GaussNav. PyTorch implementation of paper: GaussNav: Gaussian Splatting for Visual Navigation

robot_sugar. Official implementation of "SUGAR: Pre-training 3D Visual Representations for Robotics" (CVPR'24).

BunnyVisionPro. Bimanual Dexterous Teleoperation with Real-Time Retargeting using VisionPro

TeleVision. [CoRL 2024] Open-TeleVision: Teleoperation with Immersive Active Visual Feedback

VLN-PRET. Jupyter Notebook

InternUtopia. A simulation platform for versatile Embodied AI research and developments.

NavGPT-2. [ECCV 2024] Official implementation of NavGPT-2: Unleashing Navigational Reasoning Capability for Large Vision-Language Models

VLN-MAGIC. This is the official repository for MAGIC: Meta-Ability Guided Interactive Chain-of-Distillation Learning towards Efficient Vision-and-Language Navigation

HA3D_simulator. Official implementation of Human-Aware Vision-and-Language Navigation: Bridging Simulation to Reality with Dynamic Human Interactions (NeurIPS DB Track'24 Spotlight).

LangSplat. Official implementation of the paper "LangSplat: 3D Language Gaussian Splatting" [CVPR2024 Highlight]

DiaLoc. Official implementation of DiaLoc: An Iterative Approach to Embodied Dialog Localization (CVPR 2024)

SceneVerse. Official implementation of ECCV24 paper "SceneVerse: Scaling 3D Vision-Language Learning for Grounded Scene Understanding"

vlfm. The repository provides code associated with the paper VLFM: Vision-Language Frontier Maps for Zero-Shot Semantic Navigation (ICRA 2024)

FreeSplat. Official implementation of NeurIPS 2024 paper: "FreeSplat: Generalizable 3D Gaussian Splatting Towards Free-View Synthesis of Indoor Scenes"