No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Use of Probabilistic Clusters Supports for Broadcast News Segmentation
Abstract
Broadcast news segmentation is a recognised important task in audiovisual archives retrieval systems. Though research is very active in the field, the problem has not been satisfactorily solved yet. This paper presents a promising novel methodology at segmenting news programmes based on a probabilistic approach, and innovative evaluation framework to assess the performance of the proposed techniques. Experimental results are very encouraging w.r.t. state of the art. Additionally the method has revealed very robust against the various editorial formats of news programmes.
... It is important that these stories can form a learning cloud similar to the concept introduced in the Newscloud. Newscloud (Iulio and Messina, 2008) is the system that segments news content automatically into stories and provides search and navigation in news content. As shown inFigure 3, the news video is on the left of the webpage, while its stories are on the right. ...
Multimedia analysis techniques can enable e-learning systems and applications to understand multimedia content automatically. Therefore, such techniques can provide various novel services to both e-learning video providers and learners. This paper aims at providing lecturers and learners with a state-of-the-art overview of multimedia analysis techniques used mainly in internet video for e-learning purposes. Finally, the paper presents some of the open issues for further research in the area of video analysis for e-learning.
... under-segmentation). As under-segmentation effects can be considered as being more penalising than over-segmentation effects, we used an evaluation measurement taking into account starting boundaries and ending boundaries with different weights, as well as considering missing material as having more impact than extra material on the measurement [5]. The system was tested against a test set of 84 programmes obtaining precision and recall of 0.76 and 0.73, respectively. ...
In the modern age, cross-media production is an innovative technique used by the media industry to ensure a positive return
on investments while optimising productivity and market coverage. So that, technologies for the seamless fusion of heterogeneous
data streams are increasingly considered important, and thus research efforts have started to explore this area. After having
aggregated heterogeneous sources, what has to be addressed as the next problem is efficiency in retrieving and using the produced
content. Tools for personalised and context-oriented multimedia retrieval are indispensable to access desired content from
the aggregated data in a quicker and more useful way. This chapter describes the problems connected with this scenario and
proposes an innovative technological framework to solve them, in the area of informative content (news) distribution and retrieval.
Extensive experiments prove the effectiveness of this approach in a real-world business context.
... In order to measure the quality of news segmentation we used an alignment measurement that takes into account starting and ending boundaries with different weights. In addition, it considers under-segmentation effects (i.e., when the detected story starts/ends after/before the actual story) as being more penalising than over-segmentation effects (i.e., when the detected story starts/ends before/after the actual story) on the measurement [8]. The system was tested against a test set of 84 programmes (i.e., ≈ 40 hours of material) achieving a precision of 0.76 and a recall of 0.73. ...
Current Web technology has enabled the distribution of in- formative content through dynamic media platforms. In ad- dition, the availability of the same content in the form of di- gital multimedia data has dramatically increased. Content- based, cross-media retrieval applications are needed to e-- ciently access desired information from this variety of data sources. This paper presents a novel approach for cross- media information aggregation, and describes a prototype system implementing this approach. The prototype adopts online newspaper articles and TV newscasts as information sources, to deliver a service made up of items including both contributions. Extensive experiments prove the efiectiveness of the proposed approach in a real-world business context.
In the modern digital age methodologies for professional Big Data analytics represent a strategic necessity to make information resources available to professional journalists and media producers in a more effective and efficient way and enabling new forms of production like data-driven journalism. The challenge lies in the ability of collecting, connecting and presenting heterogeneous content streams accessible through different sources, such as digital TV, the Internet, social networks and media archives, and published through different media modalities, such as audio, speech, text and video, in an organic and semantics-driven way. Rai Active News is a portal for professional information services that addresses these challenges with a uniform and holistic approach. At the core of the system there is a set of artificial intelligence techniques and advanced statistical tools to automate tasks such as information extraction and multimedia content analysis targeted at discovering semantic links between resources, providing users with text, graphics and video news organized according to their individual interests.
This paper describes Hyper Media News (HMNews), a system for the automated aggregation and consumption of information streams
from digital television and the Internet. TV newscasts are automatically segmented, annotated and indexed. Such information
is then integrated with those available from Internet blogs, newspapers and press agencies. The end result is a set of innovative
information services that supplies retrieval, recommendation and browsing of multi-modal news items across different production
paradigms, ranging from traditional professional media, e.g. television and press, to new user-centric media platforms such
as social networking sites, internet forums and blogs.
Automatic news program segmentation and classification becomes a hot topic, which reorganizes the news program according to
the news’ topics, and provides the on-demand services to mobile consumers or Internet/home TV consumers. This paper presents
a personalized news consuming system, including the system architecture, consumption steps and key techniques. Then, focused
on the core technique, i.e., video temporal segmentation, the automatic video temporal segmentation method is proposed, evaluated
and compared with existing ones. Experimental results show that the proposed scheme is computational efficient and gets higher
correct detection rate. These properties make it a suitable choice for the personalized news consuming system.
KeywordsVideo temporal segmentation-Shot boundary detection (SBD)-News segmentation-Personalized news consuming
Internet video is becoming more and more popular in consuming field. The increasing of video data volumes requires some innovative
means to enrich the video consuming experiences. Multimedia analysis is the key means, which can provide conveniences to both
Internet video providers and consumers. This paper reviews the progress of media analysis techniques, investigates typical
innovative Internet video services based on media analysis, introduces some innovative user experiences based on media analysis,
and presents some open issues and potential research topics in media analysis and the related applications in Internet video
services. We hope the paper provides valuable information to the researchers, engineers or decision-makers working in the
fields of Internet video and media analysis.
The University of Iowa participated in the shot boundary detection, story segmentation and feature extraction tracks of TRECVID-2004.
Segmenting news video into stories is among key issues for achieving efficient treatment of news-based digital libraries.
In this paper we present a novel unsupervised algorithm that combines audio and video information for automatic partitioning
news videos into stories. The proposed algorithm is based on the detection of anchor shots within the video. In particular,
a set of audio/video templates of anchorperson shots is first extracted in an unsupervised way, then shots are classified
by comparing them to the templates using both video and audio similarity. Finally, a story is obtained by linking each anchor
shot with all successive shots until another anchor shot, or the end of the news video, occurs. Audio similarity is evaluated
by means of a new index and helps to achieve better performance in anchor shot detection than pure video approach. The method
has been tested on a wide database and compared with other state-of-the-art algorithms, demonstrating its effectiveness with
respect to them.
This paper presents the system used by the LIUM to partici- pate in ESTER, the french broadcast news evaluation campaign. This system is based on the CMU Sphinx 3.3 (fast) decoder. Some tools are presented which have been added on different steps of the Sphinx recognition process: segmentation, acoustic model adaptation, word-lattice rescoring. Several experiments have been conducted on studying the ef- fects of the signal segmentation on the recognition process, on injecting automatically transcribed data into training corpora, or on testing different approaches for acoustic model adapta- tion. The results are presented in this paper. With very few modifications and a simple MAP acoustic model estimation, Sphinx3.3 decoder reached a word error rate of 28.2%. The entire system developed by LIUM obtained 23.6% as official word error rate for the ESTER evaluation, and 23.4% as result of an unsubmited system.
The TREC Video Retrieval Evaluation (TRECVid) is an
international benchmarking activity to encourage research
in video information retrieval by providing a large test collection, uniform scoring procedures, and a forum for organizations interested in comparing their results. TRECVid completed its fifth annual cycle at the end of 2005 and in 2006 TRECVid will involve almost 70 research organizations, universities and other consortia. Throughout its existence, TRECVid has benchmarked both interactive and automatic/manual searching for shots from within a video
corpus, automatic detection of a variety of semantic and
low-level video features, shot boundary detection and the
detection of story boundaries in broadcast TV news. This
paper will give an introduction to information retrieval (IR) evaluation from both a user and a system perspective, highlighting that system evaluation is by far the most prevalent type of evaluation carried out. We also include a summary of TRECVid as an example of a system evaluation benchmarking campaign and this allows us to discuss whether
such campaigns are a good thing or a bad thing. There are
arguments for and against these campaigns and we present
some of them in the paper concluding that on balance they
have had a very positive impact on research progress.
The segmentation of news video into story units is an important step towards effective processing and management of large news video archives. In the story segmentation task in TRECVID 2003, a wide variety of techniques were employed by many research groups to segment over 120-hour of news video. The techniques employed range from simple anchor person detector to soisticated machine learning models based on HMM and Maximum Entropy (ME) approaches. The general results indicate that the judicious use of multi-modality features coupled with rigorous machine learning models could produce effective solutions. This paper presents the algorithms and experience learned in TRECVID evaluations. It also points the way towards the development of scalable technology to process large news video corpuses.
We describe the Automatic News Summarisation and Extraction System (ANSES), which captures television news each day with the accompanying subtitles and identifies and extracts news stories from the video. Lexical chain analysis is used to provide a summary of each story and important entities are highlighted in the text.
In this paper we describe a novel approach to lexical chain based segmentation of broadcast news stories. Our segmentation system SeLeCT is evaluated with respect to two other lexical cohesion based segmenters TextTiling and C99. Using the Pk and WindowDiff evaluation metrics we show that SeLeCT outperforms both systems on spoken news transcripts (CNN) while the C99 algorithm performs best on the written newswire collection (Reuters). We also examine the differences between spoken and written news styles and how these differences can affect segmentation accuracy.