YiHungWONG

Construction-Hazard-Detection. Enhances construction site safety using YOLO for object detection, identifying hazards like workers without helmets or safety vests, and proximity to machinery or vehicles. HDBSCAN clusters safety cone coordinates to create monitored zones. Post-processing algorithms improve detection accuracy.

YOLOv8-License-Plate-Insights. This repository demonstrates YOLOv8-based license plate recognition with GCP Vision AI integration, enabling versatile real-world applications like vehicle identification, traffic monitoring, and geospatial analysis while capturing vital media metadata for enhanced insights.

Llama2-Telegram-Bot. Integration of the advanced llama2 AI model with Telegram to provide real-time chatbot responses. Seamless interaction, smart replies, and scalable deployment.

Image-Classification-with-EfficientNetV2. This repository contains code for an image classification model using TensorFlow and Keras. It includes a script for training the model, a script for predicting new images using the trained model, and a sample dataset of images for testing.

CCTV-Inundation-Detection. Code to detect rain/inundation using CCTV images, estimate affected area/depth and store data in MySQL. Image processing & ML for efficient flood monitoring & management.

YOLOv8-PostProcessing-PRCurve. This repository contains code for post-processing techniques and PR curve visualization for object detection using YOLOv8. The project aims to improve object detection results through advanced post-processing methods and provides an analysis of precision and recall using PR curves.

BIMAPI_Revit-Family-Parameters-Generator. This repository contains a project that employs the Revit API for automated generation of diverse beam and column sizes in large-scale BIM projects, aiming to enhance efficiency, minimise errors, and ensure modelling consistency.

VisionSnap. VisionSnap is a uniquely designed iOS app, employing Swift for seamless camera interactions and sophisticated image recognition. Leveraging the YOLOv8 model on a Google Cloud server, VisionSnap allows real-time image capturing and analysis, providing results instantly back to the device. A perfect blend of AI and user experience.

Reuters-News-Classification-and-Analysis. Train a model to categorize news articles, scrape and translate articles, and predict their categories using TensorFlow, Keras, and Google Translate API.

Traffic-Violation-Report-System. A platform for users to upload and share the responses from law enforcement agencies to their traffic violation reports in Taiwan. This system aims to increase transparency and public oversight of traffic law enforcement.

Vitis-AI-YOLOv3-TF2-Quantization-Evaluation. "Vitis-AI-YOLOv3-TF2-Quantization-Evaluation" Repo for quantization of YOLOv3 on Vitis-AI using TF2, aimed to deploy model on edge devices with limited resources. Includes training & quantization scripts and evaluation metrics. Experiment with different configurations.

Diabetes-Prediction-API. A RESTful API using Flask and XGBoost to predict diabetes in Pima Indians based on various diagnostic measurements. Includes training, saving the best model, and testing the API using Python requests.

Continuous-Media-Capture. This repository contains code for capturing continuous media from a given URL, and saving the media frames locally. The media type, URL, and camera name can be specified. This is useful for security cameras or live feeds.

LSTM-Stock-Forecaster. A deep learning-based stock price prediction application using Long Short-Term Memory (LSTM) neural networks and Flask API.

YOLOv8-Dataset-Transformer. A comprehensive toolkit for converting image classification datasets into object detection datasets and training them using YOLOv8. This project streamlines the process of dataset preparation, augmentation, and training, making it easier to leverage YOLOv8 for custom object detection tasks.

Site-Entry-Tracker-YOLOv8. A YOLOv8-based AI model for monitoring construction sites, trained to identify safety gear, vehicles, and machinery. Utilises safety cones to set a virtual line for tracking entries and exits, aiding in the management of personnel and equipment flow.

YOLOv8-qat. Quantization Aware Training

Azure-Line-Chatbot. An integrated chatbot application utilising OpenAI GPT models, hosted on Azure, with seamless connectivity to LINE messaging platform. Designed for modularity, scalability, and efficient logging with Azure Blob Storage.

yihong1120.

Speech-Commands-Classification-LSTM. A TensorFlow project for classifying speech commands using LSTM neural networks. The project includes a script for training the model on the Speech Commands dataset and a script for making predictions on audio files.

PyQt5-File-Uploader-to-REST-API. This is a PyQt5 based GUI application that allows users to select a file and POST it to a REST API. The application receives a request ID from the API and uses it to query the current status of the computation. The GUI provides an intuitive interface for users to interact with the API and retrieve the results.

Neural-Style-Transfer-Web-App. This web app allows users to apply artistic styles to their own images or a set of images retrieved from Google Images using neural style transfer.

Poly-Shield-Detector. Poly-Shield-Detector is an image processing tool that combines YOLOv8 and Pyside6. Detect objects in images, define polygons, and filter objects inside polygons. Perfect for interactive annotation and object recognition tasks. Efficient and user-friendly.

Web_Equipment-Rental-Website. A robust web-based platform designed to streamline the process of renting construction equipment and finding skilled drivers in the construction industry. Features a user-friendly interface, efficient relational database management, and an automated matching system for optimal equipment and driver pairing.

Automation-Scheduler. "Automation-Scheduler" is a repository for a simple bash script that automate the execution of a python script at specific times, schedule task and manage cron job easily. The script can be modified to suit different automation needs.

Genre-Based-Movie-Recommender. An intuitive movie recommendation system leveraging genre similarity with TF-IDF and cosine similarity for a personalized film discovery experience.

Stock-Investment-Tracker. This repository contains a tool that can help track your stock investments and monitor their performance. With this tool, you can keep an eye on your investments, make informed decisions, and achieve your financial goals.

Omega-AI. Omega-AI：基于java打造的深度学习框架，帮助你快速搭建神经网络，实现模型推理与训练，引擎支持自动求导，多线程与GPU运算，GPU支持CUDA，CUDNN。

Iris-Classification-API. "Iris-Classification-API" is a Github Repository that uses Flask to create an API that accepts JSON data, generates a request ID, and utilizes the Scikit-Learn library to classify Iris data. The API allows users to query the classification progress and returns the results of flower classification.

Web-Zone-Guard. Python

MarketDataScience. Python

automl. Google Brain AutoML

transfiner. Mask Transfiner for High-Quality Instance Segmentation, CVPR 2022

Mask_RCNN. Mask R-CNN for object detection and instance segmentation on Keras and TensorFlow