Abstract: A videoconferencing system allows a distributed group of separately addressable multipoint controller signaling layer and multipoint controller application layer units to be logically combined for the purpose of managing a videoconference with a plurality of endpoints. In some embodiments, a primary multipoint controller signaling layer unit may identify other multipoint controller signaling layer units and multipoint controller application layer units to handle requests for joining an ongoing videoconferencing session, routing such requests to the other multipoint controller signaling layer units and multipoint controller application layer units as needed.

Abstract: A conference apparatus reduces or eliminates noise in audio for endpoints in a conference. Endpoints in the conference are designated as a primary talker and as secondary talkers. Audio for the endpoints is processed with speech detectors to characterize the audio as speech or not and to determine energy levels of the audio. As the audio is written to buffers and then read from the buffers, decisions for the gain settings of faders for read audio of the endpoints being combined in the speech selective mix. In addition, the conference apparatus can mitigate the effects of a possible speech collision that may occur during the conference between endpoints.

Abstract: A method of generating a category model for classifying medical images. The method comprises providing a plurality of medical images each categorized as one of a plurality of categorized groups, generating an index of a plurality of visual words according to a distribution of a plurality of local descriptors in each the image, modeling a category model mapping a relation between each visual word and at least one of the categorized groups according to the index, and outputting the category model for facilitating the categorization of an image based on local descriptors thereof.

Abstract: A striped composed video image is disclosed, in which a main segment image is displayed with a strip of small strip segments. The strip of small strip segments is composed from a plurality of video images from a plurality of video streams. A user may pull video from the strip into the main segment. The strip of small strip segments may scrolled without affecting the main segment.

Abstract: A method of generating a category model for classifying medical images. The method comprises providing a plurality of medical images each categorized as one of a plurality of categorized groups, generating an index of a plurality of visual words according to a distribution of a plurality of local descriptors in each the image, modeling a category model mapping a relation between each visual word and at least one of the categorized groups according to the index, and outputting the category model for facilitating the categorization of an image based on local descriptors thereof.

Abstract: A conference apparatus reduces or eliminates noise in audio for endpoints in a conference. Endpoints in the conference are designated as a primary talker and as secondary talkers. Audio for the endpoints is processed with speech detectors to characterize the audio as speech or not and to determine energy levels of the audio. As the audio is written to buffers and then read from the buffers, decisions for the gain settings of faders for read audio of the endpoints being combined in the speech selective mix. In addition, the conference apparatus can mitigate the effects of a possible speech collision that may occur during the conference between endpoints.

Abstract: A system and method is disclosed for adapting a continuous presence videoconferencing layout according to interactions between conferees. Using regions of interest found in video images, the arrangement of images of conferees may be dynamically arranged as displayed by endpoints. Arrangements may be responsive to various metrics, including the position of conferees in a room and dominant conferees in the videoconference. Video images may be manipulated as part of the arrangement, including cropping and mirroring the video image. As interactions between conferees change, the layout may be automatically rearranged responsive to the changed interactions.

Abstract: A technique allows delivering content in a videoconference among a plurality of conferees in multiple quality levels. Content distributed from a multipoint control unit that is connected to an endpoint that delivers the content to receiving endpoints may have two or more quality levels. The quality level can be adapted to the capabilities of the receiving endpoints and the current quality of the connection that is used to deliver the content toward the receiving endpoints. The quality level associated with an endpoint may change during a session dynamically, adapting the delivery mode to changes in the condition of the connection between the receiving endpoints and the multipoint control unit.

Abstract: A method and system are provided for maximizing reception rate of datagrams transmitted in a packet-switched network, by minimizing their transmission rate. The method comprises: forwarding datagram groups from a first location towards a second location, wherein each of the datagram groups includes a plurality of datagrams each comprising a first indication for identifying the group to which it belongs, wherein each group is forwarded at a different time than others, and wherein all datagrams that belong to a certain group are forwarded at substantially the same rate; at the second location, determining an arrival rate of datagrams which belong to a first group; based on the determined arrival rate, determining whether a change should be affected at the forwarding rate of a second group of datagrams; and if so, forwarding the second group of datagrams at the new rate.

Abstract: A method and systems are provided for transmitting encoded video data over a packet switched network, by: (i) providing a plurality of encoded copies of video data, each encoded at a bitrate different from that of the others; (ii) transmitting one of the copies encoded at a first bitrate from a first end of a communication channel towards a second end of the communication channel; (iii) receiving the encoded copy at the second end and detecting the channel error rate associated therewith; (iv) at the first end, providing an indication associated with the detected error rate to enable transmitting from the first end another encoded copy having a different bitrate; and (v) transmitting a second copy encoded at a second bitrate, wherein the second bitrate is substantially equal to a maximal rate at which that encoded copy is recoverable at the second end with no noticeable apparent errors.

Abstract: A method of generating a category model for classifying medical images. The method comprises providing a plurality of medical images each categorized as one of a plurality of categorized groups, generating an index of a plurality of visual words according to a distribution of a plurality of local descriptors in each the image, modeling a category model mapping a relation between each visual word and at least one of the categorized groups according to the index, and outputting the category model for facilitating the categorization of an image based on local descriptors thereof.