ZENG: Anonymous person tracking using AI
2025-2026 ősz
Szoftver
Téma leírása
Anonymous person tracking is a rapidly growing area in computer vision that focuses on detecting and monitoring individuals’ movements without identifying their personal identity. It balances technical innovation with ethical and privacy concerns, making it highly relevant in today's surveillance-aware world.
Using deep learning models such as YOLO (You Only Look Once) for detection and DeepSORT or ByteTrack for tracking, AI systems can assign and follow anonymous IDs to individuals in video feeds. The emphasis is on understanding patterns of movement, behavior, or spatial usage, rather than identifying who the person is.
This approach is widely applicable in public safety, smart cities, retail analytics, and transportation, where insights are needed about crowd behavior, space optimization, or anomalous activity—without violating privacy laws or ethical standards.
The project can involve:
-
Designing a non-intrusive tracking system using public datasets or real-time video.
-
Implementing object detection and re-identification techniques that avoid biometrics like faces or gait.
-
Creating visual analytics such as heatmaps, path mapping, and movement statistics.
-
Exploring privacy-preserving strategies, such as blurring faces or storing only abstract movement data.
The work contributes to building AI systems that are effective yet ethically sound, aligning with principles like GDPR, and advancing responsible use of surveillance technology.
Maximális létszám:
2 fő