|The following paper examines the development of a framework that allows the generation of video synopsis. That is a video file obtained by overlaying the main moving objects in a single scene. This allows for file length reduction thus optimization of the analysis and storage of video surveillance footage. The proposed framework is based on modern methods in the field of machine learning for the automatic recognition and localization of objects in the video frames, their segmentation, tracking, and merging on the extracted background. Machine learning models based on convolutional neural networks were used for this purpose.|
*** Title, author list and abstract as seen in the Camera-Ready version of the paper that was provided to Conference Committee. Small changes that may have occurred during processing by Springer may not appear in this window.