Project description
The project aims to develop an intelligent computer vision system capable of real-time object and action recognition using deep learning neural networks. The system processes camera video, detects objects, and classifies actions (e.g., human movement, gestures, and interactions with objects).
The solution can be based on modern architectures such as YOLO, SSD, or Faster R-CNN for object detection, as well as LSTM or 3D-CNN models for action sequence analysis.
The project can be applied in security systems, smart video surveillance, industrial automation, and educational systems.