Real-time multi-object tracker using YOLOv7 and StrongSORT with OSNet
-
Updated
May 28, 2024 - Python
Real-time multi-object tracker using YOLOv7 and StrongSORT with OSNet
Local, API-free multi-stage identity pipeline: face recognition → liveness detection → super-resolution fallback → person Re-ID, built for realistic CCTV-style conditions.
OpenVINO Training Extensions Object Re-identification
python notebook super resolution
Person Re-Identification pipeline (YOLOv8 + TorchReID/OSNet) — the fallback stage for matching people by appearance when face recognition can't resolve an identity.
Vehicle Re-Identification (ReID) dataset contains over 55,000 images for training and validation of the vehicle re-identification model
Multi-camera people tracking with cross-camera ReID association and bird's-eye-view spatial mapping. YOLOv8m + StrongSort + OSNet on EPFL Laboratory dataset — CMC@3: 90.0%
A production-ready, locally-hosted Retrieval-Augmented Generation (RAG) system for chatting with your documents.
Detect, track and count different classes of objects.
C++ pipeline for cross-camera person tracking: YOLO person detection + BYTETrack/IoU in-process or NvDCF via DeepStream, OSNet ReID + Hungarian global ID assignment
Multi-sport forecasting platform: CV player tracking from broadcast video, calibrated win-probability models across 4 sports, leak-guarded walk-forward validation, pre-registered claims ledger, and a documented reject graveyard. Calibration rigor, not edge claims.
Implementing and training an AI model for Person Re-Identification task
Successfully developed a real-time soccer player re-identification and tracking system using YOLOv11 and Deep SORT with TorchreID-based embeddings.
AI-powered retail analytics platform for footfall tracking, heatmaps, zone analytics, demographics, and cross-camera ReID.
Track video objects with YOLOv8 and StrongSORT, using appearance embeddings and MOT metrics for reliable multi-object tracking
Advanced multi-object tracking pipeline — YOLOv8m + StrongSORT with OSNet appearance embeddings on MOT17. HOTA 41.6 | MOTA 38.1 | IDF1 50.8. Benchmarked against ByteTrack baseline with TrackEval.
AI-powered PPE compliance monitoring system using YOLO, ByteTrack, Flask and OpenCV.
A novel implementation of Multi-Object Tracking in fixed frame setting
Scalable Flask API powering the Badminton AI Analyser. Orchestrates a concurrent multi-model pipeline (YOLO, StrongSORT, TrackNet, SlowFast) on Google Cloud Platform to process high-speed match footage.
This is the pipeline 1 for player detection in football match. (gpu recommended)
Add a description, image, and links to the osnet topic page so that developers can more easily learn about it.
To associate your repository with the osnet topic, visit your repo's landing page and select "manage topics."