Video summarisation by deep visual and categorical diversity
The authors propose a video‐summarisation method based on visual and categorical diversities using pre‐trained deep visual and categorical models. Their method extracts visual and categorical features from a pre‐trained deep convolutional network (DCN) and a pre‐trained word‐embedding matrix. Using...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2019-09-01
|
| Series: | IET Computer Vision |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/iet-cvi.2018.5436 |
| _version_ | 1797684719301689344 |
|---|---|
| author | Pedro Atencio Sánchez‐Torres German John William Branch Claudio Delrieux |
| author_facet | Pedro Atencio Sánchez‐Torres German John William Branch Claudio Delrieux |
| author_sort | Pedro Atencio |
| collection | DOAJ |
| description | The authors propose a video‐summarisation method based on visual and categorical diversities using pre‐trained deep visual and categorical models. Their method extracts visual and categorical features from a pre‐trained deep convolutional network (DCN) and a pre‐trained word‐embedding matrix. Using visual and categorical information they obtain a video diversity estimation, which is used as an importance score to select segments from the input video that best describes it. Their method also allows performing queries during the search process, in this way personalising the resulting video summaries according to the particular intended purposes. The performance of the method is evaluated using different pre‐trained DCN models in order to select the architecture with the best throughput. They then compare it with other state‐of‐the‐art proposals in video summarisation using a data‐driven approach with the public dataset SumMe, which contains annotated videos with per‐fragment importance. The results show that their method outperforms other proposals in most of the examples. As an additional advantage, their method requires a simple and direct implementation that does not require a training stage. |
| first_indexed | 2024-03-12T00:33:49Z |
| format | Article |
| id | doaj.art-0e22299bedda42bd8670fee74c1f8a11 |
| institution | Directory Open Access Journal |
| issn | 1751-9632 1751-9640 |
| language | English |
| last_indexed | 2024-03-12T00:33:49Z |
| publishDate | 2019-09-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Computer Vision |
| spelling | doaj.art-0e22299bedda42bd8670fee74c1f8a112023-09-15T10:01:28ZengWileyIET Computer Vision1751-96321751-96402019-09-0113656957710.1049/iet-cvi.2018.5436Video summarisation by deep visual and categorical diversityPedro Atencio0Sánchez‐Torres German1John William Branch2Claudio Delrieux3Faculty of EngineeringInstituto Tecnológico MetropolitanoMedellinColombiaFaculty of EngineeringUniversidad del MagdalenaSanta MartaColombiaFaculty of MinesUniversidad Nacional de ColombiaMedellinColombiaElectric and Computing Engineering DepartmentUniversidad Nacional del SurBahia BlancaArgentinaThe authors propose a video‐summarisation method based on visual and categorical diversities using pre‐trained deep visual and categorical models. Their method extracts visual and categorical features from a pre‐trained deep convolutional network (DCN) and a pre‐trained word‐embedding matrix. Using visual and categorical information they obtain a video diversity estimation, which is used as an importance score to select segments from the input video that best describes it. Their method also allows performing queries during the search process, in this way personalising the resulting video summaries according to the particular intended purposes. The performance of the method is evaluated using different pre‐trained DCN models in order to select the architecture with the best throughput. They then compare it with other state‐of‐the‐art proposals in video summarisation using a data‐driven approach with the public dataset SumMe, which contains annotated videos with per‐fragment importance. The results show that their method outperforms other proposals in most of the examples. As an additional advantage, their method requires a simple and direct implementation that does not require a training stage.https://doi.org/10.1049/iet-cvi.2018.5436video summarisationdeep visual diversitiescategorical diversitiesvideo-summarisation methodpre-trained deep visual modelscategorical models |
| spellingShingle | Pedro Atencio Sánchez‐Torres German John William Branch Claudio Delrieux Video summarisation by deep visual and categorical diversity IET Computer Vision video summarisation deep visual diversities categorical diversities video-summarisation method pre-trained deep visual models categorical models |
| title | Video summarisation by deep visual and categorical diversity |
| title_full | Video summarisation by deep visual and categorical diversity |
| title_fullStr | Video summarisation by deep visual and categorical diversity |
| title_full_unstemmed | Video summarisation by deep visual and categorical diversity |
| title_short | Video summarisation by deep visual and categorical diversity |
| title_sort | video summarisation by deep visual and categorical diversity |
| topic | video summarisation deep visual diversities categorical diversities video-summarisation method pre-trained deep visual models categorical models |
| url | https://doi.org/10.1049/iet-cvi.2018.5436 |
| work_keys_str_mv | AT pedroatencio videosummarisationbydeepvisualandcategoricaldiversity AT sancheztorresgerman videosummarisationbydeepvisualandcategoricaldiversity AT johnwilliambranch videosummarisationbydeepvisualandcategoricaldiversity AT claudiodelrieux videosummarisationbydeepvisualandcategoricaldiversity |