Trackers gives you clean, modular re-implementations of leading multi-object tracking algorithms released under the permissive Apache 2.0 license. You combine them with any detection model you already use.
BoT-SORT + YOLOX implemented using only onnxruntime, Numpy and scipy, without cython_bbox and PyTorch. Fast human tracker. OSNet is not used.
Хакатон компании Ренью
Football broadcast video analysis with YOLOv26 + YOLO11--player, referee, and ball detection, identity tracking, and possession analytics
SOTA Multi Object Tracker in Cpp
ZoneGuard - Real-time perimeter security system using YOLO11 for detecting persons entering restricted zones. Tracks movement with distance, speed, and duration analytics.
An AI-driven Adaptive Traffic Signal Control System (ATSCS) that replaces static timers with dynamic green-light phases. Utilizes YOLOv8 for real-time vehicle density estimation and multi-class classification, achieving 96.4% mAP. Optimized for low-latency inference (>30 FPS) on edge devices to reduce urban congestion and commuter wait times.
面向 NVIDIA Jetson Orin 的本地视频目标检测与 BoT-SORT 追踪工作台，提供 TensorRT 推理、多档速度/精度方案、结果导出与 AP 评估。
A production-ready computer vision system for real-time retail intelligence. Features YOLOv8 person detection, BoT-SORT tracking, store occupancy, zone analytics, customer journey mapping, and spatial heatmaps with a React dashboard
Real-time multi-object tracking with YOLOv8 detection and BoT-SORT + CLIP ReID — evaluated on MOT17 (43.3% HOTA).
Modular multi-camera person tracking pipeline: YOLO26 + BoT-SORT detection/tracking, clothes-agnostic Re-ID (C2DResNet50 + CAL) and a global movement graph with an interactive Streamlit dashboard. Validated on the MEVID benchmark.
A complete computer vision pipeline for soccer video understanding. It combines YOLO11, BoT-SORT, and HSV pitch filtering for robust player tracking, alongside tactical behavior analysis via ROI presence estimation
Robust Computer Vision and Uncertainty Estimation for Aquatic Ecosystems
ExplainableTrack — an explainable, interpretable toolkit for object tracking with visualizations.
A Machine Vision pipeline that automates feeding of fishes by detecting, tracking and scheduling feeding interval.
Aerial multi-object tracking on VisDrone: RT-DETR detections with a from-scratch SORT tracker and BoT-SORT.
Dockerized ROS 2 Jazzy node for find & track only: BoT-SORT tracker + PD control. Subscribes to /image_raw/compressed and /detections (JSON), publishes /cmd_vel. Includes MJPEG preview with aiming overlay.
Vehicle tracking and speed estimation (km/h) from drone video without telemetry: YOLO11, BoT-SORT, camera-motion compensation and camera geometry inferred from vehicle sizes