An automatic and serialized ROI extraction framework for the slow-motion video frames

Abstract

Currently, high-speed cameras has been a very common equipment in many important application fields. How to effectively and automatically extract the ROI (region of interest) for the slow-motion video has been a novel interesting challenge. In recent research work, we designed a ROI extraction framework for the video frames produced by high-speed cameras. The entire framework includes two parts: a novel but simple color similarity measure model is improved to distinguish different pixels; a skeleton feature points based serialized segmentation tactics is proposed to generate seed points. By using the multithreading patterns of parallelizing computations in the extraction process, the ROI in the serialized color slow-motion video frames can be marked automatically and accurately. Comparing with the common methods, this method has advantage in segmentation effect and computational efficiency. It can establish the technical basis for the pertinent subsequent studies.