Spherical torus-based video hashing for near-duplicate video detection

Dear editor, With the rapid development of multimedia tech- nologies, users can easily generate and share multi- ple videos through the Internet. Similarly, numer- ous illegal and useless near-duplicate videos gen- erated through simple reformatting, transforma- tion, and editing appear on the web. These near- duplicate videos inconvenience users when they are surfing the Internet and are considered a copyright infringement. Robust video hashing, which is also called video fingerprinting, does not require access to the video contents at the time of creation and can be used to detect existing contents. In general, a robust video hash is a short digest extracted from a video; however, it is robust to content-preserving attacks such as noise, logo addition, and contrast change. That is, similar to a human fingerprint that identifies a specific person, video hashing can classify video contents by extracting and compar~ ing short digests. As such, video hashing is a fea~ sible means to detect near-duplicate videos.