Harvesting information from video surveillance systems
Abstract:
Video surveillance systems have become ubiquitous in the modern world. Surveillance cameras are used for monitoring, tracking, emergency event detection, etc. The amount of data generated by such systems is enormous and will be increased by an order of magnitude in coming years. Computer vision In this project we will concentrate on data from public highway surveillance camera. The goal will be to classify the traffic flow based on images from the cameras. We will start from a simple approach for classification ( feature extraction + linear models ) and will gradually progress to more advanced machine learning techniques like random forest and neural networks.
Project prerequisites:
- Math
- Linear algebra
- Calculus
- Introduction to machine learning:
- Feature Extraction
- Classification problem
- Cross validation techniques
- Overfitting/underfitting
- Model evaluation
- Programming language: Python
- Basics of computer vision (optional)
Associated topics:
supervised machine learning, image classification, computer vision
Planned lectures:
- Basics of Computer Vision
About lecturer:
Mr. Andrey Lyubonko,
Lead Engineer at Samsung R&D Institute Ukraine (SRK)