A5114636679- Video Coding and Compression Technologies
- Stochastic Gradient Optimization Techniques
- Advanced Vision and Imaging
- Privacy-Preserving Technologies in Data
- Advanced Image Processing Techniques
Simon Fraser University
2024
Abstract Video coding has traditionally been developed to support services such as video streaming, videoconferencing, digital TV, and so on. The main intent was enable human viewing of the encoded content. However, with advances in deep neural networks (DNNs), is increasingly being used for automatic analytics performed by machines. In applications traffic monitoring, vehicle detection, tracking counting, would run continuously, while could be required occasionally review potential...
Federated Learning (FL) enables multiple clients to train a collaborative model without sharing their local data. Split (SL) allows be trained in split manner across different locations. Split-Federated (SplitFed) learning is more recent approach that combines the strengths of FL and SL. SplitFed minimizes computational burden by balancing computation servers, while still preserving data privacy. This makes it an ideal framework various domains, especially healthcare, where privacy utmost...