We write your reusable computer vision tools. 💜

Ultralytics YOLOv8, YOLOv9, YOLOv10, YOLOv11, YOLOv12 for ROS 2

Converting COCO annotation (CVAT) to annotation for YOLO-seg (instance segmentation) and YOLO-obb (oriented bounding box detection)

OBB implementation in python (using numpy)

AI for remote sensing, remote sense, object detection, oriented object detection, computer vision, cv

Python and C++ examples that show shows how to process 3-D Lidar data by segmenting the ground plane and finding obstacles.

SPARC is an intelligent ROS2 Humble-based pick-and-place robotic system integrating MoveIt, YOLOv8-OBB, and Isaac Sim. It enables real-time object detection, motion planning for Franka Panda and UR5 arms, and natural language control via an Ollama-powered chatbot—making robotic manipulation smart, interactive, and human-friendly.

Cube Detection using YOLOv5 with Oriented Bounding Boxes (OBB)

Some object detection codes for DOTA dataset

Wrapping C++ native code in C# with example of CGAL in Grasshopper.

AIDet is a Code Library for Object Detection in Aerial Images Based on MMDetection.

Repo for the OBBStacking: An Ensemble Method for Remote Sensing Object Detection

📡 Production-grade maritime surveillance system for detecting ships in Synthetic Aperture Radar (SAR) imagery using YOLOv8-OBB (Oriented Bounding Boxes) and Lee speckle filtering. Handles the unique challenges of radar imagery: multiplicative noise and arbitrarily-oriented vessels.

A guide/template for training the YOLOv8 oriented bounding boxes object detection model on custom datasets.

📦 HBB2OBB is a Python library for converting horizontal (axis-aligned) bounding boxes to oriented (rotated) bounding boxes using SAM-based segmentation. Includes tools for visualization, evaluation, format conversion, and hyperparameter tuning.

Little wheels to help better understand graphics principles.

Aircraft Recognition using YOLO-OBB: A deep learning model for detecting and classifying aircraft using oriented bounding boxes. The project includes dataset preprocessing, annotation conversion, model training, validation, and inference using YOLO-OBB.

A Greenfoot template that enhances collision detection using convex hulls and the Separating Axis Theorem, optimized with an efficient hybrid approach.

Quaternion Approximate Networks for Enhanced Image Classification and Object Detection

Python-based 3D geometry pipeline for measuring raw .obj mesh objects using oriented bounding boxes (OBB). It extracts vertex data, applies PCA to find the object’s natural orientation, computes tight bounding box dimensions and volume, and validates results via Open3D-based 3D visualization, similar to real-world 3D scanning, measurement systems.