jalywang123's repositories
PersonDetection_SORT
Person detection using deep learning model and SORT (KalmanFilter) / DeepSort (KalmanFilter + Embeddings) algorithm for tracking
pytorch-arm-builds
Unofficial PyTorch and torchvision builds for ARM devices
chineseocr
yolo3+ocr
docker-nvenc
FFmpeg docker images with NVIDIA NVENC (Hardware-Accelerated Video Encoding)
facenet_trt
NVIDIA TensorRT implementation for facenet with pre-trained Inception-ResNet v1 SavedModel/Ckpt and MTCNN Networks
yolov3ha
Yolov3 server for HomeAssistant integration via MQTT
YOLOv3-model-pruning
在 oxford hand 数据集上对 YOLOv3 做模型剪枝(network slimming)
Real-Time-Video-Stabilization
real time video stabilization using Kalman Filter
voxelnet
This is an unofficial inplementation of VoxelNet in TensorFlow.
Vehicle-Counting
Video-based vehicle counting system built with Python/OpenCV.
Python-100-Days
Python - 100天从新手到大师
ScratchDet
The code and models for paper: "ScratchDet: Exploring to Train Single-Shot Object Detectors from Scratch"
JetsonNanoInsulatorDetection
compare tensorflow and TensorRT approach in image detection and classification with TF and TensorRT approach
MTCNN_FaceDetection_TensorRT
MTCNN C++ implementation with NVIDIA TensorRT Inference accelerator SDK
yolo-dd
Using the yolov3 neural network to detect and classify model construction vehicles real time and display the distance of them using the ZED stereo camera.
SRN
Selective Refinement Network for High Performance Face Detection, AAAI, 2019
C-3-Framework
An open-source PyTorch code for crowd counting
Real-Time-Object-Detection-and-Tracking
This project combines object detection and object tracking. The object detecting algorithm is SSD and the object detecting algorithm is SiamRPN. Both algorithms are real-time.
object-detection-tracking
Light-weight real-time multi-object detection and tracking in Nvidia TX2
DeepSORT
MOT using deepsort yolo3 with C++
Pelee-TensorRT
Pelee(NeurIPS'18)-TensorRT Implementation
deep_sort
Simple Online Realtime Tracking with a Deep Association Metric
TensorRT-Yolov3
TensorRT for Yolov3
Online-Realtime-Action-Recognition-based-on-OpenPose
A skeleton-based real-time online action recognition project, classifying and recognizing base on framewise joints, which can be used for safety surveilence.
unofficial-implement-of-openpose
Simple Implement of Openpose use Tensorflow代码贼简单,tensorboard很全
Traffic-Survalance-with-Computer-Vision-and-Deep-Learning
The system takes video footage of a highway as input and provides statistics like the count of vehicles and an average estimated speed of vehicles on the highway. The statistics provided by the system can have many applications. Like, pricing the billboards on a highway for advertisement, higher the count of vehicles, higher the price. Moreover, the government can use this statistic to know how many vehicles are entering a city each day. The system internally uses YOLO object detection algorithm for vehicle detection, followed by, Centroid Tracking algorithm for tracking the detected vehicles.