🏞 A content-based image retrieval (CBIR) system

Artificial Intelligence Learning Notes.

HoG, PCA, PSO, Hard Negative Mining, Sliding Window, Edge Boxes, NMS

Training SVM classifier to recognize people expressions (emotions) on Fer2013 dataset

Detect eye blinks based on eye aspect ratio (EAR) introduced by Soukupová and Čech in their 2016 paper, Real-Time Eye Blink Detection Using Facial Landmarks.

Histogram Of Oriented Gradients

HOG-based linear SVM for detecting vehicles (or any other object) in videos

物体检测　object-detection 用于检测红灯笼demo

Vehicle detection based on YOLO and SVM

HOG+SVM图像分类-行人检测

This project uses Histogram of Oriented Gradients for pedestrian detection and Kalman Filter for tracking and prediction

A real-time human counter that uses HOG and SVM. Developed as project for the Computer Vision course at Sapienza University of Rome (2021-22)

Image Search Engine

This is "ready from box" face recognition app, based on Mediapipe, dlib and face_recognition modules.

Content-based medical image retrieval - image similarity search for radiology images

Numpy Implementation of classic computer vision feature extraction algorithm.

In this project 4 distinct tasks (gender detection (A1), smile detection (A2), face-shape recognition (B1), eye-color recognition (B2)) are adressed following 4 different approaches and exploiting the potentialities of CNNs and HOG descriptors along with SVMs.

Face Recognition library

A live drowsiness detection system made to run on a single board computer like raspi. It has been tested with different extreme parameters of distance and spectacles. Can be a huge product to run on cars and prevent accidents as the system runs with 0% internet connectivity.

Detect if the driver is feeling drowsy and sound an alarm on doing so to wake him/her up.