Python library for YOLO small object detection and instance segmentation

icip2022 paper: sahi benchmark on visdrone and xview datasets using fcos, vfnet and tood detectors

Sakuli is an end-2-end testing and monitoring tool for web sites and common UIs with multiple monitoring integrations

SAHI(Slicing Aided Hyper Inference)をYOLOX(ONNX)で動かしたサンプル

Yolov5 Modelini Kullanarak Özel Nesne Eğitimi ve SAHI Kullanımı

Bu repo SAHI uygulamasını mantığını öğreniyoruz.

Tool for detecting cracks on construction materials (based on YOLO V8 + SAHI). Simple measurements made easy with a GUI

Tile (Slice) a YOLO Dataset for Small Object Detection and Instance Segmentation

使用SAHI模块完成超分辨率以及小目标检测是应用示例，仅供参考。

Sliced Detection and Clustering Analysis Toolkit

YOLOv9 (Ultralytics) Python interface for training, validating and running detection on waste detection dataset + detection using SAHI.

This project contains service to parse Sahi run result and display it in JSON. This service is a part of SwiftOps project

SAHI Kütüphanesi için Detaylı Türkçe Dokümantasyonu

This Repo contains code for used for age and gender detection in CCTV videos

This is a plugin for SwiftOps Bot Engine.

Tile your highres images to squares in FiftyOne directly to train for small object detection purpose (e.g. with SAHI)

This work is an extension of the project https://github.com/ablanco1950/LicensePlate_Yolov8_Filters_PaddleOCR adding the possibility to detect the speed, tracking and counting cars. Has been updated to feature and leverage improvements brought by Roboflow and Ultralytics.

This repository contains a Python-based program that detects and tracks people in a video, counting the number of individuals entering and exiting a defined area. It uses the YOLOv8 model for object detection and the SORT (Simple Online and Realtime Tracking) algorithm for tracking.

This project integrates SAHI with YOLOv8 for efficient object detection, supporting image, video, and real-time webcam feeds. By using SAHI's slicing technique, it improves detection in complex or high-resolution scenarios. Ideal for versatile object detection needs with state-of-the-art performance.