powerdoudou Goto Github PK

-

agv-visual-detect-and-track

alphapose_yolovx

AlphaPose_yolovx include yolov3, yolov4, yolov5.

basic

cs224w-2021-winter-colab

cs224w（图机器学习）2021冬季课程的colab

dgl

Python package built to ease deep learning on graph, on top of existing DL frameworks.

doudou-image

Save My Images

hero21-vision-code

img

img

iot-the_vision_based_on_traffic_analysis_system-for_multicopter_drone_application

This is an IoT project. It can be divided into two parts: an app for Android and a drone mount that runs on the Raspberry Pi 4. The app can use the mobile network or Wi-Fi to remotely control the camera of the Raspberry Pi and watch the camera's real-time video stream. Once the image is acquired, the application will use OpenCV to image recognition. The purpose of image recognition is to analyze traffic conditions. However, there are still some limitations, such as the inability to accurately calculate the distance. So probably a relative comparison. This project uses quite a few third-party APIs. On the application side, the Haar and histogram analysis of OpenCV is used and the OpenVPN service that solves the TCP connection of the mobile network. On the Raspberry Pi side, D-link's product DWM-222 is used to provide the mobile network. Turn Raspberry Pi Wi-Fi into a hotspot. It is convenient to connect the mobile phone directly. No need to prepare a Wi-Fi base station. Video streaming is supported by V4l2 API. The streaming source of CCTV is the Highway Authority. The results at this stage are over. OpenVPN's servers are down. Only Wi-Fi remains functional. To learn more, watch the demo video. 這是一個物聯網專案，他可以分為兩個部分：Android的應用程式和一個在樹梅派4上運行的無人機掛載物，應用程式可以使用行動網路或是Wi-Fi遙控樹梅派的攝影機轉動並觀看攝影機的即時影像串流，當影像取得後，應用程式將使用OpenCV進行影像辨識，影像辨識的目的為交通狀況分析，但是目前仍有一些限制，像是無法準確的計算距離，因此可能是一個相對的比較。 這個專案用到相當多的第三方API，在應用程式端，使用到了OpenCV的Haar和直方圖分析以及解決行動網路進行TCP連線的OpenVPN服務，在樹莓派端，使用了D-link的產品DWM-222提供行動網路，將Wi-Fi改成熱點，方便手機直接連線，不需準備Wi-Fi基地台，影像的串流藉助了V4l2的API支援，CCTV的串流來源為高速公路管理局，現階段的成果已經結束，OpenVPN的伺服器已經關閉，只剩Wi-Fi可以正常運作，如想了解，可以觀看示範影片。

lighttrack-ncnn

Ncnn version demo of [CVPR21] LightTrack: Finding Lightweight Neural Network for Object Tracking via One-Shot Architecture Search

llm-universe

本项目是一个面向小白开发者的大模型应用开发教程，在线阅读地址：https://datawhalechina.github.io/llm-universe/

p_5c39354b

prune

prunerune

real-time-audio-video

Stream real-time audio and video over TCP in Linux

st-gcn-pytorch

Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition

tensorrt_pro

C++ library based on tensorrt integration

testproject