We present a fuzzy color histogram-based shot-boundary detection algorithm specialized for content based copy detection applications. The proposed method aims to detect both cuts and gradual transitions (fade, dissolve) effectively in videos where heavy transformations (such as cam-cording, insertions of patterns, strong re-encoding) occur. Along with the color histogram generated with the fuzzy linking method on L*a*b* color space, the system extracts a mask for still regions and the window of picture-in-picture transformation for each detected shot, which will be useful in a content-based copy detection system. Experimental results show that our method effectively detects shot boundaries and reduces false alarms as compared to the state-of-the-art shot-boundary detection algorithms.

We discuss how gaze behavior of an observer can be simulated as a Monte Carlo sampling of a distribution obtained from the saliency map of the observed image. To such end we propose the Levy Hybrid Monte Carlo algorithm, a dynamic Monte Carlo method in which the walk on the distribution landscape is modelled through Levy flights. Some preliminary results are presented comparing with data gathered by eye-tracking human observers involved in an emotion recognition task from facial expression displays.

Context-sensitive queries for image retrieval in digital libraries

Multimedia applications have been extended over the past decade, and efficient methods for video indexing and retrieval are required to confront this situation. Video temporal segmentation is the first and essential step for visual data processing, which aims to find the location of shot boundaries, including abrupt shot boundaries (AB) and gradual shot boundaries (GB). In this paper we propose an algorithm for video shot detection with Frame-Skipping technique. In our algorithm only some typical frames are selected for initial processing, which can be obtained by Frame-Skipping Module (FSM). As a visual description, we may choose 1 or 2 frames from 10 consecutive frames. To diminish the error caused by Frame-Skipping technique, we use Abrupt-Boundary Module (ABM) and Gradual-Boundary Module (GBM) to detect video shots, where the ABM based on global histogram comparison is used to detect abrupt shot boundaries and the GBM based on local histogram comparison concentrates on detecting gradual transitions. This new algorithm can reduce about 40%∼70 % time-consumption compared with some previous algorithms on 4 typical video clips (movie, cartoon, sports and news).

Shot change detection is an essential step in video content analysis. However, automatic shot change detection often suffers high false detection rates when there are camera or object movements. In this paper, we propose an approach to solve this problem based on local keypoint matching of the video frames. Experimental results show that the proposed algorithm is effective for all kinds of shot changes. 1

journal homepage: www.elsevier.com/locate/cviu

Abstract—With recent advances in multimedia technologies, digital TV and information highways, more and more video data is being captured, produced and stored. However, without appropriate techniques that can make the video content more accessible, all these data are hardly usable. Content Based Video Retrieval (CBVR) becomes a proper solution to handling the video databases. The essential first step in CBVR is Video Segmentation. This paper is a critical survey of current trends / methods for video segmentation. This work has been done with an aim to assist the upcoming researchers in the field of video retrieval to know about the technology and methods available for video segmentation. Keywords-video segmentation; content-based video retrieval; video indexing;key frame; histogram 1

In this paper, a review of certain basic information extraction operations that can be performed on video is presented. These operations are those that are particularly useful as a foundation for further semantic processing. Specifically, the review focuses on shot boundary detection and condensed video representation (also called summarization and abstraction). Shot boundary detection is the complete segmentation of a video into continuously imaged temporal video segments. Condensed video representation is the extraction of video frames or short clips that are semantically representative of the corresponding video. Both tasks are very significant for the organization of video data into more manageable forms. An overview of the fundamental issues in each task is provided, and recent work on the subject is described and is critically reviewed.

Abstract—In this work a novel approach to evaluating video temporal decomposition algorithms is presented. The evaluation measures typically used to this end are non-linear combinations of Precision-Recall or Coverage-Overflow, which are not metrics and additionally possess undesirable properties, such as nonsymmetricity. To alleviate these drawbacks we introduce a novel uni-dimensional measure that is proven to be metric and satisfies a number of qualitative prerequisites that previous measures do not. This measure is named Differential Edit Distance (DED), since it can be seen as a variation of the well-known edit distance. After defining DED, we further introduce an algorithm that computes it in less than cubic time. Finally, DED is extensively compared with state of the art measures, namely the harmonic means (F-Score) of Precision-Recall and Coverage-Overflow. The experiments include comparisons of qualitative properties, the time required for optimizing the parameters of scene segmentation algorithms with the help of these measures, and a user study gauging the agreement of these measures with the users ’ assessment of the segmentation results. The results confirm that the proposed measure is a uni-dimensional metric that is effective in evaluating scene segmentation techniques and in helping to optimize their parameters. I.