FIELD: physics.
SUBSTANCE: invention relates to the field of video processing. Method includes obtaining video frame images, calculating the semantic attributes vector of detected faces and frames on the obtained images and storing the obtained attribute vectors in the metadata. Distance metric between vectors of features of adjacent frames is calculated, boundaries of survey plans are obtained and boundaries are stored in video metadata. Method includes obtaining samples of vectors of semantic signs of frames and detected faces of each survey plan, calculating clusters of survey plans and preserving the criteria for assigning survey plans to a certain class in metadata. Selecting representative frames from the filming plans making up the cluster and storing the frame data in the metadata are performed. Based on the metadata, the first visual search facility is provided in the form of a set of lines, each of which contains the same number of images of representative frames. Second means of visual search is formed in the form of images of headings of collections of filming plans grouped as belonging to a cluster of semantically close survey plans. Third means of visual search is formed in the form of images of captions collection headings grouped by the presence of the face of the video participant.
EFFECT: technical result consists in improvement of accuracy and speed of visual search of required fragment of video materials.
6 cl, 8 dwg
| Title | Year | Author | Number |
|---|---|---|---|
| DEVICE FOR SEMANTIC CLASSIFICATION AND SEARCH IN ARCHIVES OF DIGITIZED FILM MATERIALS | 2016 |
|
RU2628192C2 |
| METHOD OF AUTOMATIC PHOTOGRAPH FRAMING | 2006 |
|
RU2329535C2 |
| DETECTING OBJECTS FROM VISUAL SEARCH REQUESTS | 2017 |
|
RU2729956C2 |
| SYSTEM AND METHOD OF FORMING TRAINING SET FOR MACHINE LEARNING ALGORITHM | 2017 |
|
RU2711125C2 |
| METHOD AND SYSTEM FOR VIDEO SCENES SEGMENTATION | 2022 |
|
RU2783632C1 |
| VIDEO RECEPTION FRAMEWORK FOR VISUAL SEARCH PLATFORM | 2017 |
|
RU2720536C1 |
| METHOD AND SYSTEM FOR AUTOMATICALLY CHECKING PRESENCE OF A LIVING HUMAN FACE IN BIOMETRIC SAFETY SYSTEMS | 2005 |
|
RU2316051C2 |
| SIMULTANEOUS RECOGNITION OF PERSON ATTRIBUTES AND IDENTIFICATION OF PERSON IN ORGANIZING PHOTO ALBUMS | 2018 |
|
RU2710942C1 |
| ANNOTATION BY MEANS OF SEARCHING | 2007 |
|
RU2439686C2 |
| METHOD AND SYSTEM FOR AUTOMATED GENERATION OF VIDEO STREAM WITH DIGITAL AVATAR BASED ON TEXT | 2020 |
|
RU2748779C1 |
