HUI11126 / Compute-continuous-moments-de-ned-in-a-rectangular-region-using-CUDA-and-some-applications
☆24Updated 2 years ago
Related projects ⓘ
Alternatives and complementary repositories for Compute-continuous-moments-de-ned-in-a-rectangular-region-using-CUDA-and-some-applications
- segmentation algorithm yolact use tensorrt deploy☆13Updated 2 years ago
- SuperGlue MNN C++部署,SuperGlue C++ Inference with MNN☆63Updated 2 years ago
- in this repo, we deployed SOLOv2 to TensorRT with C++.☆33Updated last year
- TensorRT实现BiSeNetV1与BiSeNetV2部署☆18Updated 2 years ago
- c++ detail implementation of camera calibration☆50Updated 4 years ago
- cuda 加速3D点云算法库,持续更新(含cudaicp,glfw点云可视化等)☆13Updated 2 years ago
- BevDet_TensorRT☆4Updated last year
- 1.相机模型标定(鱼眼相机,普通相机);2.立体相机标定与极线矫正;3.RGB(/灰度)相机与深度相机(tof)标定,数据融合;4.外参标定☆37Updated 6 months ago
- 深度学习特征点匹配算法SuperGluePretrainedNetwork https://github.com/magicleap/SuperGluePretrainedNetwork C++部署☆38Updated 3 years ago
- TensorRT of DDRNet for real-time segmentation☆38Updated 2 years ago
- 张正友内参标定,内含详细推导☆80Updated last year
- LightGlue-OnnxRunner is a repository hosts the C++ inference code of LightGlue in ONNX format,supporting end-to-end/decouple model infere…☆15Updated 11 months ago
- ☆112Updated last year
- A plugin to make view transformer from perspective view to bird-eye-view, it is used in bevdet☆19Updated last year
- ☆24Updated 2 years ago
- 实现Ransac随机采样算法分割地面点云☆30Updated 2 years ago
- A fork of the BEVDet series .☆20Updated last year
- Fisheye correction using distortion table☆24Updated 7 years ago
- ☆15Updated 9 months ago
- ☆13Updated 2 years ago
- The repo is modified from the source code of CannyLine. It provides support for pybind to make it available for python.☆10Updated 3 years ago
- 本仓库在OpenVINO推理框架下部署Nanodet检测算法,并重写预处理和后处理部分,具有超高性能!让你在Intel CPU平台上的检测速度起飞! 并基于NNCF和PPQ工具将模型量化(PTQ)至int8精度,推理速度更快!☆14Updated last year
- ☆27Updated 2 years ago
- PointPillars TensorRT version pretrained on MMDetection3d with WaymoOpenDataset☆16Updated 2 years ago
- ☆24Updated last year
- 这是一个基于CUDA加速的SGM立体匹配代码,它的核心是SemiglobalMatching(SGM)算法,它不仅在时间效率上要远远优于基于CPU的常规SGM,而且明显占用更少的内存,这意味着它不仅可以在较低分辨率(百万级)图像上达到实时的帧率,且完全具备处理千万级甚至更高…☆53Updated 6 months ago
- ☆22Updated last year