Abstract: A method for optimizing the performance of an algorithm for detecting predetermined content in a media information stream, and a program and apparatus that operate in accordance with the method. The algorithm is a function of a set of parameters. The method comprises the steps of performing the algorithm at least once to detect the predetermined content in the media information stream, while employing a respective set of parameters in the algorithm for each performance thereof, and automatically evolving at least one respective set of parameters employed in the algorithm to maximize the degree of accuracy at which the algorithm detects the predetermined content in the media information stream.

Abstract: A method and system for providing the creation of a visual summary of a video source during fast forward/rewind of the video can be created by using extracting frames either automatically or manually to create an initial visual summary. A series of weights may be assigned to the extracted frames, which are then filtered according to the relative weights to create a modified visual summary. The keyframe display rate is then adjusted according tot he fast forward/rewind speed, which can be either a standard speed or user selected speed, so as to display the keyframes while the video source is being fast forwarded/rewound. The keyframes may be substituted by selected images and audio, so that an advertiser can substitute an image of the product and a brief audio summary while the user is fast forwarding/rewinding past the commercial message.

Abstract: A method for operating a neural network, and a program and apparatus that operate in accordance with the method. The method comprises the steps of applying data indicative of predetermined content, derived from an electronic signal including a representation of the predetermined content, to an input of at least one neural network, to cause the at least one network to generate at least one output indicative of either a detection or a non-detection of the predetermined content. Each neural network has an architecture specified by at least one corresponding parameter. The method also comprises a step of evolving the at least one parameter to modify the architecture of the at least one neural network, based on the at least one output, to increase an accuracy at which that at least one neural network detects the predetermined content indicated by the data.

Abstract: A method and system for providing the creation of a visual summary of a video source during fast forward/rewind of the video can be created by using extracting frames either automatically or manually to create an initial visual summary. A series of weights may be assigned to the extracted frames, which are then filtered according to the relative weights to create a modified visual summary. The keyframe display rate is then adjusted according tot he fast forward/rewind speed, which can be either a standard speed or user selected speed, so as to display the keyframes while the video source is being fast forwarded/rewound. The keyframes may be substituted by selected images and audio, so that an advertiser can substitute an image of the product and a brief audio summary while the user is fast forwarding/rewinding past the commercial message.

Abstract: A method for optimizing the performance of an algorithm for detecting predetermined content in a media information stream, and a program and apparatus that operate in accordance with the method. The algorithm is a function of a set of parameters. The method comprises the steps of performing the algorithm at least once to detect the predetermined content in the media information stream, while employing a respective set of parameters in the algorithm for each performance thereof, and automatically evolving at least one respective set of parameters employed in the algorithm to maximize the degree of accuracy at which the algorithm detects the predetermined content in the media information stream.

Abstract: A method and device for detecting commercials using the encoding parameters of a compressed video stream is provided. A video encoder receives uncompressed video data and generates a compressed video data. A plurality of separators, each defined by at least two consecutive scene changes in a sequence of the compressed video data, is detected. Then, the beginning and ending of a commercial break is derived by comparing the gap between these separators to a predetermined threshold value.

Abstract: There is disclosed, for use in a system for detecting a movement of a selected object in a sequence of visual animated data frames, a video processing device for generating a descriptor data structure representative of a trajectory of the selected object. The video processing device comprises an image processor for identifying the selected object in the visual animated data frames and determining a trajectory of the selected object in a two dimensional (2D) or three dimensional (3D) coordinate space. The image processor generates a descriptor data structure that contains a position (usually a starting position) of the object in the coordinate space; a speed of the object in the coordinate space; and an acceleration of the object in the coordinate space.

Abstract: A system and method for non-intrusively embedding a multimedia advertisement associated with an object in an audiovisual program provided by one of various sources. The advertisement may be accessed and extracted from the program by pointing and clicking on the object while the object is displayed. A summary of advertisements may thereby be created. The accessed advertisements may be bookmarked or stored to create the summary. A user may create a profile of preferred types(s) of advertisements to filter out the display of advertisements for undesired types of advertisements.

Abstract: The invention relates to a method and device for estimating motion between images forming a sequence of segmented images or partitions composed of I regions R.sub.i identified by labels. The method comprises a step of initializing motion parameters of each region R.sub.i, an intermediate processing step of the images on which the estimation of the motion is performed, and an iterative refining step for the definitive determination of said motion parameters in the form of a vector (D.sub.x, D.sub.y) for all the pixels of each of the regions R.sub.i. This refining step, including a minimizing operation of the prediction errors over each object by means of the minimization of an associated function, allows to select the most representative points and reject the others as outliers. Application: very low bitrate coding, multimedia applications.

Abstract: The invention relates to a method and system for coding segmented pictures, and to corresponding decoding method and system. In the coding system, a first sub-system defines the time evolution of said segmented pictures, or partitions, and a second one encodes both contours and textures of the regions of the successive partitions. The time evolution definition leads to a partition tree (PT) from which regions are extracted during an analysis step (AS) in order to form a decision tree (DT). A decision operation allows to select during a choice step (CS) distinct regions R.sub.S from various levels of said partition tree, in order to construct an optimal final partition (BFP) and, simultaneously, to choose the coding technique (BCT) that is the best one for each region of said optimal partition. Reciprocally the decoding method and system comprise means for defining the coding strategy chosen for each of said regions and decoding correspondingly the coded information.

Abstract: Method and device for estimating motion between segmented images, or partitions, composed of a number I regions R.sub.i. For performing the method, this device comprises a stage (41) for initializing parameters which describe the motion of each region R.sub.i, in which one of different motion hypotheses chosen a priori is selected after prediction, computation of the prediction error and selection of the motion hypothesis to which the smallest prediction error corresponds. This stage (41) is followed by a stage (42) for intermediate processing by filtering, and a stage (43) for refining the motion parameters and is looped to the input of the stage (42) for an iterative operation until it ends at a given criterion so as to finally obtain I motion information M.sub.i (t) corresponding to the regions R.sub.i.