Real-time multi-object tracker using YOLOv7 and StrongSORT with OSNet

OpenVINO Training Extensions Object Re-identification

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: YOLOv8 + pluggable tracker (BYTETrack/IoU/NvDCF) + OSNet ReID + Hungarian global ID assignment

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.

Multi-sport AI forecasting platform built solo by directing Claude agent fleets: CV tracking from broadcast video, calibrated win-prob models across 4 sports, leak-free walk-forward validation, pre-registered claims ledger, honest reject graveyard. Calibration rigor, not edge claims.

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.

AI-powered retail analytics platform for footfall tracking, heatmaps, zone analytics, demographics, and cross-camera ReID.

Real-time person re-identification and object matching pipeline using YOLO for detection, OSNet for feature extraction, and FAISS for high-speed similarity search and dynamic ID matching.

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 project builds a person detection and tracking system that distinguishes between children and adults, assigns unique IDs, and handles occlusions, re-entries, and new appearances. It’s designed to monitor interactions between children with Autism Spectrum Disorder (ASD) and therapists, supporting personalized treatment planning.

This is the pipeline 1 for player detection in football match. (gpu recommended)

A complete DeepSORT multi-object tracking pipeline implemented from scratch — training, evaluation, fine-tuning, ONNX export, and real-time inference on webcam or video files.