Browsing and exploration of video sequences: a new scheme for key frame extraction and 3D visualization using entropy based Jensen divergence
This paper proposes a unified scheme for video browsing and exploration. Our scheme involves two components, video key frame extraction and 3D visualization. For key frame extraction, we develop a generic approach in which the Jensen-Shannon divergence (JSD), Jensen-Rényi divergence (JRD) and Jensen...
| Autores: | , , , , , , |
|---|---|
| Tipo de documento: | artigo |
| Estado: | Versão publicada |
| Data de publicação: | 2014 |
| País: | España |
| Recursos: | Varias* (Consorci de Biblioteques Universitáries de Catalunya, Centre de Serveis Científics i Acadèmics de Catalunya) |
| Repositório: | Recercat. Dipósit de la Recerca de Catalunya |
| OAI Identifier: | oai:recercat.cat:10256/11658 |
| Acesso em linha: | http://hdl.handle.net/10256/11658 |
| Access Level: | Acesso embargado |
| Palavra-chave: | Video key frame extraction Three-dimensional display systems) Vídeo digital Digital video Registre d'imatges Image registration Imatges -- Processament -- Tècniques digitals Image processing -- Digital techniques |
| Resumo: | This paper proposes a unified scheme for video browsing and exploration. Our scheme involves two components, video key frame extraction and 3D visualization. For key frame extraction, we develop a generic approach in which the Jensen-Shannon divergence (JSD), Jensen-Rényi divergence (JRD) and Jensen-Tsallis divergence (JTD) are investigated for measuring the difference between neighboring video frames, segmenting a video clip into shots then possibly into sub-shots, and choosing key frames in each shot. Our novel approach is computationally inexpensive and yet effective, as shown by experimental results. As for 3D visualization, an innovative prototype, in which key frames, their related information and video contexts are displayed, is created for video exploration. Our visualization tool also contributes to easily locating, highlighting and removing the possible redundant key frames. The key frames selected by our proposed approach and their corresponding visualization interface are combined to lead to a fast grasp of video contents |
|---|