CleanStream is an OBS plugin that uses AI to clean live audio streams from unwanted words and utterances

Realtime Sign Language Detection: Deep learning model for accurate, real-time recognition of sign language gestures using Python and TensorFlow.

Search for objects on video (possible with IP cameras)

Real-time lane and car detection system using YOLOv8 and OpenCV, with distance estimation for vehicles. Ideal for autonomous driving and traffic analysis.

使用ONNXRuntime部署LSTR基于Transformer的端到端实时车道线检测，包含C++和Python两个版本的程序

Computer vision-based system for real-time detection and localization of road surface defects such as potholes and cracks, is proposed to overcome the limitations and inefficiency of human-based visual onsite road inspection.

A Keras implementation of YOLOv3 (Tensorflow backend)

Logo Detection of a custom small dataset. The dataset contains logos of 6 famous brands: Nike, Jordans, Adidas, Puma, Kappa, Quicksilver. The model was developed using the Object-Detection-API by Tensorflow

computer vision based face tracking and facial expression android app.

A model was built to detect ASL alphabets in real time that uses Deep Learning and Neural Networks to predict ASL alphabets.

Realtime Face Mask Detection Flutter App.

Big data and image processing project

A mobile application to track your macronutrients intake, calories and set up goals on your fitness journey

An application which can run object detection tasks from device cameras and IP cameras in real time.

Detect a player in real-time and extract hand positions. Interact with a projected video-game.

This project focuses on a real-time Helmet Detection System to ensure road safety by identifying riders who are not wearing helmets. Integrated with a Flutter application, it provides an automated system for issuing challans (e-tickets) and capturing images of the rider and their number plate for record-keeping.

ProcessING Inspection Eyes

A simple Python script to check the strength of a password based on length, the inclusion of numbers, special characters, and upper/lower case letters.

Based on Faster R-CNN, we train model on our mask dataset and leverage data augmentation to preprocess our data. Mean average precision is introduced to evaluate the model performance. You'll see data augmentation and mAP evaluation in detailed explainations, and tutorials of faster-rcnn training

Using ML 🧠 , detect in realtime if people are wearing mask or not 😷.