Are you Xavier Naturel?

Claim your profile

Publications (9)1.01 Total impact

  • Xavier Naturel, Sid-Ahmed Berrani
    [Show abstract] [Hide abstract]
    ABSTRACT: One of the promises of digital television is the possibility of creating interactive and innovative television services, like catch-up TV. However, these services need external resources, coming from the channels themselves or from manual annotation. In this paper, a system for automatically building a Catch-up TV service from the available EPG and the broadcasted TV stream, is presented. The system combines several content-based techniques for extracting exact program boundaries from the TV stream. Traditional commercial detection and recognition methods are used, as well as novel techniques to detect and classify repetitions. Identification of the TV program is then performed by matching the detected boundaries with the EPG. Extensive experiments on three weeks of TV assess the effectiveness of the proposed system.
    ISM 2009, 11th IEEE International Symposium on Multimedia, San Diego, California, USA, December 14-16, 2009; 01/2009
  • Gaël Manson, Xavier Naturel, Sid-Ahmed Berrani
    [Show abstract] [Hide abstract]
    ABSTRACT: This demo presents a complete system that automatically structures TV streams on the fly. The objective is to precisely and automatically determine the start and the end of each broadcasted TV program. The extracted programs can then be stored in a database to be used in novel services such as TV-on-Demand. The system performs on-the-fly detection of inter-programs using a reference database of inter-programs, as well as an offline detection and classification of repeated sequences. This offline phase allows us to automatically detect inter-programs as repeated sequences. The macro-segmentation is performed using the online and the offline results of inter-program detection as well as metadata, when available, in order to label extracted programs. The demo shows results on large real TV streams.
    Advances in Multimedia Modeling, 15th International Multimedia Modeling Conference, MMM 2009, Sophia-Antipolis, France, January 7-9, 2009. Proceedings; 01/2009
  • Source
    Xavier Naturel, Patrick Gros
    [Show abstract] [Hide abstract]
    ABSTRACT: Television daily produces massive amounts of videos. Digital video is unfortunately an unstructured document in which it is very difficult to find any information. Television streams have however a strong and stable but hidden structure that we want to discover by detecting repeating objects in the video stream. This paper shows that television streams are actually highly redundant and that detecting repeats can be an effective way to detect the underlying structure of the video. A method for detecting these repetitions is presented here with an emphasis on the efficiency of the search in a large video corpus. Very good results are obtained both in terms of effectiveness (98% in recall and precision) as well as efficiency since one day of video is queried against a 3 weeks dataset in only 1 s.
    Multimedia Tools and Applications 01/2008; · 1.01 Impact Factor
  • Source
    Xavier Naturel, Patrick Gros
    [Show abstract] [Hide abstract]
    ABSTRACT: This article presents a method for detecting duplicate se- quences in a continuous television stream. This is of in- terest to many applications, including commercials monito- ring and video indexing. Repetitions can also be used as a way of structuring television streams by detecting inter- program breaks as a set of duplicate sequences. In this con- text, we present a shot-based method for detecting repeated sequences ecien tly. Experiments show that this fast shot matching strategy allows us to retrieve duplicated shots be- tween a 1 hour long query and a 24 hours database in only 10 ms.
    Proceedings of the Second International Workshop on Computer Vision meets Databases, CVDB 2005, June 17, 2005, Baltimore, MD, USA; 01/2005
  • Source
    Xavier Naturel, Guillaume Gravier, Patrick Gros
    [Show abstract] [Hide abstract]
    ABSTRACT: An original task of structuring and labeling large television streams is tackled in this paper. Emphasis is put on simple and efficient methods to detect precise boundaries of programs. These programs are further analysed and labeled with information coming from a standard television program guide using an improved Dynamic Time Warping algorithm (DTW) and a manually labeled reference video dataset. It is shown that the labeling process yields a very high accuracy and opens the way to many applications. We eventually indicate how the dependency to a manually labeled video dataset can be removed by providing an algorithm for a dynamic update of the reference video dataset.
    01/1970: pages 222-231;
  • Source
    Xavier Naturel, Patrick Gros
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper investigates the problem of managing very large digital television archives. This problem is called television structuring (Or TV broadcast macro-segmentation) and is defined as the process of identifying the structure of a television stream as watchers perceive it: a succession of programs. This is the very first step in order to manage a television collection. In this report, a complete solution for television structuring is proposed, which makes use of simple yet efficient methods in order to deal with huge datasets. Methods from commercial detection are generalized to be able to distinguish regular programs from non-programs. It is shown how television program guides can be used to label the identified programs. It is finally shown how an update procedure can improve the segmentation results over time. Results are provided on 3 weeks of French television.
  • Source
    Xavier Naturel, Guillaume Gravier, Patrick Gros
  • Source
    Gaël Manson, Xavier Naturel, Sid-Ahmed Berrani
    [Show abstract] [Hide abstract]
    ABSTRACT: This demo presents a method of detecting television programs in TV streams. The objective is to automatically detect the precise start and end of broadcasted television programs. The method first detects inter-programs (commercials, trailers. . .) as repeated se-quences in the broadcast stream, and then deduces the programs boundaries. The extracted programs can then be stored in a database to be used in novel services such as TV On Demand. The demo shows examples of repeated inter-programs detection and program extraction, performed on several days of French television from two different channels.
  • Source
    Xavier Naturel