watermelooon / nnie_yolo
海思Hi3559移植YOLO
☆13Updated 4 years ago
Alternatives and similar repositories for nnie_yolo:
Users that are interested in nnie_yolo are comparing it to the libraries listed below
- 基于海思3519的YOLOv3例程☆22Updated 4 years ago
- yolov5 nine hi3516 hi3519 object detect real-time☆42Updated 4 years ago
- a tool of converting darknet yolov3 model to caffe☆18Updated 5 years ago
- Intelligent monitoring of escalator.Function including traffic statistics,passenger retention detection and large object retention detect…☆24Updated last year
- 华为海思hi系列芯片使用的NNIE推理框架教程☆60Updated 11 months ago
- MNN MTCNN C++☆50Updated 5 years ago
- CenterTrack_caffe☆24Updated 4 years ago
- yolov5s nnie☆46Updated 3 years ago
- TensorRT person tracking RFBNet300☆30Updated 5 years ago
- C++版本多目标跟踪☆27Updated 4 years ago
- MTCNN light + SORT tracking☆43Updated 5 years ago
- toybrick rknn multithread c demo frame work☆20Updated 4 years ago
- 移动端快速人脸检测模型是基于RetinaFace的优化去掉stride8以及stride32和stride16的landmark 在CPU位Intel(R) Pentium(R) CPU G2020 @ 2.90GHz(2900 MHz)的设备中人脸检测可达到40ms/帧☆15Updated 5 years ago
- mobileNet SSD 基于caffe的前向检测☆10Updated 6 years ago
- C++ re-implementation of fast Discriminative Scale Space Tracking.☆40Updated 2 years ago
- YOLOV3-Tiny TensorRT6.0 13个类别☆32Updated 5 years ago
- Implementation of YOLO and IOU tracker in C++☆17Updated 3 years ago
- U版yolov5 2.0的tensorrt加速☆37Updated 4 years ago
- Hi3559a Yolov3工程移植☆9Updated 5 years ago
- Pose Estimation using OpenPose and TensorRT (only c++)☆18Updated 6 years ago
- TensorRT inference sample for ResNet50☆34Updated 6 years ago
- face recognize based on tvm☆20Updated 5 years ago
- 海思hi3516dv300 nnie_mapper 工具 for ubuntu18.04☆23Updated 5 years ago
- hi3516cv500/hi3516dv300/hi3519av100☆17Updated 4 years ago
- ☆12Updated 4 years ago
- vgg16-SSD nnie☆36Updated 5 years ago
- nanodet int8 量化,实测推理2ms一帧!☆37Updated 4 years ago
- 车牌识别,基于HyperLPR实现,修改模型调用方法,使用caffe+tensorRT实现GPU加速,修改了车牌检测模型☆18Updated 3 years ago
- yolov5_ncnn in ununtu16.04☆10Updated 4 years ago
- ☆31Updated 4 years ago