Real-Time Image Processing and Deep Learning

This conference addresses the real-time aspects of image processing and real-time aspects of deep learning solutions in various imaging and vision applications. These aspects include algorithmic computational complexity, hardware implementation, and software optimization for the purpose of making an image processing or recognition system to operate in real-time for an application of interest. The SPIE Conference on Real-Time Image Processing and Deep learning is the continuation of the SPIE Conference on Real-Time Image and Video Processing that has been held for many years but is now expanded to include real-time deep learning for solving image recognition and identification problems. This conference, similar to the previous real-time image processing conferences, is intended to serve as a catalyst bringing together scientists and researchers from industry and academia working in real-time image processing and deep learning to present recent research results pertaining to real-time solutions to image processing and recognition applications. Papers of interest include, but not limited to, the following general topics addressing real-time aspects of image processing and deep learning: real-time image and video processing algorithms; computational efficiency aspects of image or video processing systems; real-time hardware implementation of image or video processing on embedded processors; computational efficiency aspects of training image recognition deep learning networks; real-time operation of deep learning networks for image recognition/identification; real-time hardware implementation of image recognition/identification deep learning networks on embedded processors; applying machine learning techniques to improve image processing speeds; real-time applications including autonomous vehicles, security systems, smart cameras, industrial inspection, video codecs, etc.; AI-based workflows and automations for real-time image generation and processing.