🌿 Detect and classify plant leaves in real-time using YOLOv8, enhancing crop health and productivity through early disease identification.

Autonomous robotic manipulation with machine-learning-based object detection and vision-guided control

Deep learning pet breed recognition app

An implementation of k-means clustering that maintains data association.

Deep Hierarchical Semantic Segmentation — Pixel-Wise Classification with Multi-Level Labels

Go package for computer vision using OpenCV 4 and beyond. Includes support for DNN, CUDA, OpenCV Contrib, and OpenVINO.

C source for ESP32 using ESP-IDF framework to comunicate with sensors, control and adjust actuators in the classification system. (DEMO Video link below)

ROS 2 perception system fusing IR/NIR/ToF/LiDAR sensors with point clouds. Combines DBSCAN, Kalman filtering, and PointNet for object detection, tracking, and shape classification in simulation.

3D object classification using dictionary learning and RGB-D covariance descriptors.

Sorting industrial components with an Intel Realsense D435i, a UR3e, and a Robotiq Hand E gripper

A computer vision-based project that uses YOLOv8 to detect and classify plant leaves in real time, helping identify healthy and diseased plants efficiently to support smart agriculture solutions.

Personal practice archive for the 'AI for Autonomous Driving' course (2024).

Real-time zero-shot classifier app

Celebrity face recognition system built using deep learning. It leverages a Convolutional Neural Network to identify individuals from a dataset of celebrity images. The system first pre-processes the image data, trains the model, and then uses a Haar Cascade classifier to detect faces in new images.

A Python project for classifying military aircraft using two distinct deep learning approaches: object detection with YOLOv8 and image classification with a ResNeSt50 CNN. This repository implements and compares both models on the Kaggle Military Aircraft Detection Dataset.

Computes loss between two sets of entities using the optimal assignment based on the Hungarian algorithm.

This project presents a dual-model system using YOLOv8 and YOLOv8-cls to accurately detect and classify highly similar-looking objects. It is built using FastAPI and trained on 5 visually identical medical device classes as a demonstration.

A TensorRt engine that runs any model

This repository provides a curated list of recommended AI tools for diverse applications, including tabular data analysis, sentiment analysis, speech recognition, and more.

AI-based project aiming to detect tree species using image classification