Nvidia GEAR Lab's initiative to solve the robotics data problem using world models

MapGlue: Multimodal Remote Sensing Image Matching

[ICCV 2025] ACE-G is an architecture and pre-training scheme to improve generalization for scene coordinate regression-based visual relocalization.

微舆：人人可用的多Agent舆情分析助手，打破信息茧房，还原舆情原貌，预测未来走向，辅助决策！从0实现，不依赖任何框架。

Lidar Localization and Mapping

Skyfall-GS: Synthesizing Immersive 3D Urban Scenes from Satellite Imagery

Native Multimodal Models are World Learners

📖[IEEE Sensors Journal (JSEN) ] SuperVINS: A Real-Time Visual-Inertial SLAM Framework for Challenging Imaging Conditions (integrated deep learning features)

[NeurIPS'25] Official repository of Concerto: Joint 2D-3D Self-Supervised Learning Emerges Spatial Representations

Cupid: A 3D generator that links 2D image with camera

Code for ICCV'2025 (Best student paper honorable mention) "RayZer: A Self-supervised Large View Synthesis Model"

NOF0 - 开源的 AI 交易竞技场

a Fast and Robust Glue for Joint Point-Line Matching

[ICCV 2025] Perspective-Invariant 3D Object Detection

Trace Anything: Representing Any Video in 4D via Trajectory Fields

Public repository for Skills

Real-Time AI Stereo Vision Library

Repository of the paper "AnyUp: Universal Feature Upsampling".

RL gym for vision language models in JAX

Official code for paper "InstantSfM: Fully Sparse and Parallel Structure-from-Motion"

Code for "FlashWorld: High-quality 3D Scene Generation within Seconds"

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

Thinking with Camera: A Unified Multimodal Model for Camera-Centric Understanding and Generation

EvoWorld: Evolving Panoramic World Generation with Explicit 3D Memory

The best ChatGPT that $100 can buy.

Statistical Uncertainty Learning for Robust Visual-Inertial State Estimation

[NeurIPS 2025] Pixel-Perfect Depth

Visual Odometry with Transformers