诺亚盘古大模型研发背后的真正的心酸与黑暗的故事。

Reading notes about Multimodal Large Language Models, Large Language Models, and Diffusion Models

SGLang is a fast serving framework for large language models and vision language models.

A project demonstrating Lidar related AI solutions, including three GPU accelerated Lidar/camera DL networks (PointPillars, CenterPoint, BEVFusion) and the related libs (cuPCL, 3D SparseConvolution…

TensorRT plugin for 3-dimension grid sample operator

This is 8-bit quantization sample for yolov5. Both PTQ, QAT and Partial Quantization have been implemented, and present the results based on yolov5s.

YOLOv9 TensorRT C++ Implementation

[CVPR 2023 Best Paper Award] Planning-oriented Autonomous Driving

A tutorial for getting started with the Deep Learning Accelerator (DLA) on NVIDIA Jetson

Mirage Persistent Kernel: Compiling LLMs into a MegaKernel

OpenCalib: A Multi-sensor Calibration Toolbox for Autonomous Driving

SuMa++: Efficient LiDAR-based Semantic SLAM (Chen et al IROS 2019)

KITTI Object Visualization (Birdview, Volumetric LiDar point cloud )

Code for a series of work in LiDAR perception, including SST (CVPR 22), FSD (NeurIPS 22), FSD++ (TPAMI 23), FSDv2, and CTRL (ICCV 23, oral).

TensorRT deploy and PTQ/QAT tools development for FastBEV, total time only need 6.9ms!!!

A simple version of "GPU based parallel optimization for real time panoramic video stitching".

Master's thesis research on 3D object detection using LiDAR and Camera data for infrastructure and railway domains, emphasizing inference optimization and utilization of temporal information for di…

🚀🚀🚀 A collection of some awesome public YOLO object detection series projects and the related object detection datasets.

Learn OpenCV : C++ and Python Examples

Papers, code and datasets about deep learning for 3D Object Detection.

BEVFormer inference on TensorRT, including INT8 Quantization and Custom TensorRT Plugins (float/half/half2/int8).

EA-LSS: Edge-aware Lift-splat-shot Framework for 3D BEV Object Detection

Vectornet with tensorrt deploy, less than 1ms/object for inference.

TensorRT Hackathon 2022 Final Competition

A REAL-TIME 3D detection network [Pointpillars] compiled by CUDA/TensorRT/C++.

PPL Quantization Tool (PPQ) is a powerful offline neural network quantization tool.

Official GitHub repository for Argoverse dataset

Lightning fast C++/CUDA neural network framework

The Triton Inference Server provides an optimized cloud and edge inferencing solution.

Tengine is a lite, high performance, modular inference engine for embedded device