scispace - formally typeset
Search or ask a question
Topic

Video browsing

About: Video browsing is a research topic. Over the lifetime, 869 publications have been published within this topic receiving 22985 citations.


Papers
More filters
Journal ArticleDOI
TL;DR: A novel solution to the challenges of representing and browsing large video objects and to this end a novel solution is introduced that is an effective way to browse large news videos that are frequently connected with the authors' multimedia news articles, called video-tiles.
Abstract: In this paper we discuss a multimedia news system that we have developed in the Multimedia Information Research Laboratory at the University of Ottawa and beyond. We focus on the feature set—that is, the tools and facilities associated with the system. We explain the functionality of each and give some real examples of the system in action. We then outline the architecture—the system consists of a production server for document authoring, a conferencing system for collaborative news article creation, a content database for authors, a hypernews database for hypermedia news documents, a news database server with aging and archiving, and user sites. The goal was to have all components of the system communicating on OCRInet—an R&D ATM network in the Ottawa region. We then introduce the challenges of representing and browsing large video objects and to this end we introduce a novel solution that we call video-tiles. This video tool is an effective way to browse large news videos that are frequently connected with our multimedia news articles.
DOI
01 Jan 2013
TL;DR: Correlating Visual Speaker Gestures with Measures of Audience Engagement to Aid Video Browsing finds correlations between speaker Gestures and audience engagement help to aid video browsing.
Abstract: Correlating Visual Speaker Gestures with Measures of Audience Engagement to Aid Video Browsing
Patent
28 Oct 2015
TL;DR: In this paper, a browsing history processing method and device are used for relieving the storage pressure of a server and avoiding the waste of storage space, where only one browsing history is kept and other browsing histories are deleted when the same browsing histories exist, so that repeated storage of the same or similar video browsing histories is avoided.
Abstract: The invention discloses a browsing history processing method and device, which are used for relieving the storage pressure of a server and avoiding the waste of storage space. The method comprises the following steps: caching the browsing histories of videos of a user; judging whether or not the same browsing histories exist in the cached browsing histories of the videos; when the same browsing histories exist, selecting one of the same browsing histories for serving as a target browsing history; and keeping the target browsing history, and deleting other browsing histories which are the same as the target browsing history. Through adoption of the method provided by the invention, only one browsing history is kept and other browsing histories are deleted when the same browsing histories exist, so that repeated storage of the same or similar video browsing histories is avoided; the storage pressure of the server is relieved; and the waste of the storage space is avoided.
Book ChapterDOI
19 Sep 2005
TL;DR: The experimental results for sports video of world cup football games indicate that multimodal analysis is effective for video browsing and retrieval by quickly browsing event-based video clips and inputting keywords according to a predefined sports vocabulary database.
Abstract: An advanced content-based sports video browsing and retrieval system, SportsVBR, is proposed in this work. Its main features include event-based sports video browsing and keyword-based sports video retrieval. The paper first defines the basic structure of our SportsVBR system, and then introduces a novel approach that integrates multimodal analysis, such as visual streams analysis, speech recognition, speech signal processing and text extraction to realize event-based video clips selection. The experimental results for sports video of world cup football games indicate that multimodal analysis is effective for video browsing and retrieval by quickly browsing event-based video clips and inputting keywords according to a predefined sports vocabulary database. The system is proved to be helpful and effective for the overall understanding of the sports video content.
Patent
17 Sep 2014
TL;DR: In this article, a video browsing method and device consisting of an image comparing module, an image determining module and an operation executing module was proposed. But the video browsing was not performed with sequences.
Abstract: The invention relates to a video browsing method and device and belongs to the technical field of computers. The video browsing method comprises the steps that a current image frame in a video is compared with a previous image frame in the video, and target block information is obtained; if the target block information meets a preset condition, the current image frame is determined as a key frame; the next image frame of the current image frame is determined as a current image frame, the step that the current image frame in the video is compared with the previous image frame in the video to obtain target block information is conducted continually and not stopped until the last image frame in the video is compared; at least one determined image frame is played. The device comprises an image comparing module, an image determining module, an operation executing module and an image playing module. The video browsing method and device solve the problem that the video cannot be browsed due to the fact that the key frames cannot be extracted from the video with sequences, and the effect of improving the playing efficiency of the video is achieved.

Network Information
Related Topics (5)
Object detection
46.1K papers, 1.3M citations
79% related
Facial recognition system
38.7K papers, 883.4K citations
78% related
Image segmentation
79.6K papers, 1.8M citations
77% related
Feature extraction
111.8K papers, 2.1M citations
77% related
Feature (computer vision)
128.2K papers, 1.7M citations
76% related
Performance
Metrics
No. of papers in the topic in previous years
YearPapers
20233
20222
202110
202021
201919
201828