A beginner-friendly, from-zero tutorial on Large Language Models with slides, scripts, and hands-on labs.

🚀 An open-source, hands-on curriculum bridging the gap from basic RL concepts to LLM alignment, RLVR, and advanced Agentic systems.

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

Baking Gaussian Splatting into Diffusion Denoiser for Fast and Scalable Single-stage Image-to-3D Generation and Reconstruction (ICCV 2025)

本项目提供一个从零开始构建多模态大模型（MLLM）的完整实践路径，使用纯 PyTorch 手写 Transformer、Vision Transformer 和 GPT-style LLM，并通过 Connector 融合成可“看图说话”的多模态模型。同时，项目支持基于 SCST 的强化学习微调，实现从监督训练到奖励优化的端到端扩展。

Access Google Colab compute from your local VSCode

🗺️ Generate City 3D map with R3F. Include building and road information.

[IROS 2024] I2EKF-LO: A Dual-Iteration Extended Kalman Filter based LiDAR Odometry

A lightweight library for instance-level visual road marking extraction, parameterization, mapping, etc.

单目VIO+单目稠密三维重建

add a new type of input for gaussian-splatting

Real-time Depth Estimation for Jetson Orin

Gaussian-SLAM: Photo-realistic Dense SLAM with Gaussian Splatting

Effectively Detecting Loop Closures using Point Cloud Density Maps

NVIDIA TensorRT deployment of Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data.

Fast C++ logging library.

[ICRA 2024] Towards Large-Scale Incremental Dense Mapping using Robot-centric Implicit Neural Representation

a Android demo of depth anything

语义slam，物体级别特征，动态点剔除，目标检测与分割

MD-SLAM: Multi-cue Direct SLAM. Implements the first photometric LiDAR SLAM pipeline, that works withouth any explicit geometrical assumption. Universal approach, working independently for RGB-D an…

Photometric SLAM and Bundle Adjustment for RGB-D and LiDAR in CUDA.

A Modular and Multi-Modal Mapping Framework

A repository for storing models that have been inter-converted between various frameworks. Supported frameworks are TensorFlow, PyTorch, ONNX, OpenVINO, TFJS, TFTRT, TensorFlowLite (Float32/16/INT8…

PyTorch Lightning Optical Flow models, scripts, and pretrained weights.

A PyTorch implementation of NeRF (Neural Radiance Fields) that reproduces the results.

Hello AI World guide to deploying deep-learning inference networks and deep vision primitives with TensorRT and NVIDIA Jetson.

Online Monocular Lane Mapping Using Catmull-Rom Spline (IROS 2023)