Detection of Image Significance in Film Editing
Zuye Yang, Haoyuan Yu · 2020
In film editing graphics size, image content diversity, complexity, and specific practical application scene, how to quickly and efficiently using prior knowledge and accurately detect significant area, accurately evaluate significant spectral quality in order to enhance the robustness of the algorithm, the use of significance in solving practical application is still the direction to solve the problem. In this background, this paper is based on significant researches on the detection method and its application, explores the pretreatment stage and objective prior rapid generation method, the spectrum generated significant stage significant detection method based on the background, the post-processing stage no reference quality evaluation based on network of significant spectral selection methods, significant test in the film the application of the video clip.