Abstract: Reflective surfaces and display devices adversely affect a videoconferencing camera. For example, reflections from bright lights may cause the camera to change exposure settings, resulting in participants appearing dimly lit. In another example, reflections of people or displayed images may be incorrectly used by automated image framing and tracking. Reflections from a surface are often polarized, as is light from many types of display devices. A polarizer in front of the camera is moved to provide different polarization angles relative to the camera, with images acquired at these different angles. The images are processed to determine a count of features, such as heads or faces, that are detected in the images. The angle that produces the minimum number of features may be subsequently used during acquisition of images by the camera for various purposes, such as videoconferencing.

Abstract: Scene changes should be done pleasingly and without user input or control. Based on the number of speakers and changes in speakers, either to a different individual or movement by the same speaker, based on the locations of the speakers, and based on the overlap of the current and intended scenes, a decision is made whether to perform a smooth transition or do a cut. It has been determined that the decision on cut versus smooth transition is preferably based on the location of the center of the intended new scene versus the boundaries of the current scene, a cut used if the center is outside the boundaries and a smooth transition if inside. If a smooth transition, an easing function, such as ease in ease out, is performed to change the scene. A preferred value for the smooth transition is to perform the transition over 80 frames.

Abstract: A method for determining camera framing in a teleconferencing system comprises a process loop which includes acquiring an audio-visual frame from a captured a video data frame; detecting objects and extracting image features of the objects within the video data frame, ingesting the audio-visual frame into a context-based audio-visual map in an intelligent manner, and selecting targets from within the map for inclusion in an audio-video stream for transmission to a remote endpoint.

Abstract: The size of a rule of thirds frame covering a presenter and the zooming and panning of a videoconference camera are based on the motion of a presenter. The size of the frame, and thus provided walking space in the frame, varies with the speed of the presenter. The slower the presenter, the smaller and tighter the frame. The faster the presenter, the larger and the looser the frame. If the presenter is pacing, a pacing frame is developed that is centered on the limits of the pacing and is large enough to cover both ends of the pacing. The movement frame and the pacing frame provide a pleasant experience for a viewer, one where the presenter does not appear cramped or walking out of the frame.

Abstract: The frame or image of a video stream of a videoconference is divided into a series of segments for analysis. There is a primary grid, which covers the entire frame, and an alternate grid, which is shifted from the primary grid. Each segment is small enough to allow a neural network to efficiently operate on the segment without requiring downsampling. By operating on full resolution images, a participant can be identified at a greater distance from the camera. The entire frame is analyzed at a lower frequency, such as once per five seconds, but each segment containing a participant in the conference is scanned at a higher frequency, such as once per second, to maintain responsiveness to participant movement but also allow the full resolution operation.

Abstract: A videoconferencing system includes different cameras with different fields-of-view (FOVs). Frames of images acquired by these cameras may be processed to determine states, such as number of users present, talking, and so forth. Based on these states, frames from one camera may be selected to send to a far site. If the two states indicate no change in camera source from a first time to a second time, the current frame from the first camera that is sent to the far site is followed by another frame from the same camera. If the two states indicate a change in camera source from the first time to the second time, the current frame from the first camera that is sent to the far site is followed by a frame from the different camera. Depending on the two states and the cameras used, different transition effects may be applied to improve presentation.

Abstract: A method of view selection in a teleconferencing environment includes receiving a frame of image data from an optical sensor such as a camera, detecting one or more conference participants within the frame of image data, and identifying an interest region for each of the conference participants. Identifying the interest region comprises estimating head poses of participants to determine where a majority of the participants are looking and determining if there is an object in that area. If a suitable object is in the area at which the participants are looking, such as a whiteboard or another person, the image data corresponding to the object will be displayed on a display device or sent to a remote teleconference endpoint.

Abstract: A videoconferencing endpoint determines if there is a whiteboard and if a presenter is near the whiteboard. If there is no whiteboard in view or the presenter is not near the whiteboard, any content from a camera focused on the whiteboard is continued and any presenter framing is done normally. If the presenter is in front of the whiteboard, any whiteboard content is ended, and appropriate portions of the whiteboard are included in the main video stream framed with the presenter. If the whiteboard is empty, framing is done without reference to the whiteboard. If the whiteboard is full or has writing away from the presenter, the entire whiteboard and the presenter are framed together. If the whiteboard only has writing near the presenter, only the relevant portion of the whiteboard is framed with the presenter.

Abstract: Scene changes that are done pleasingly and without user input or control. Based on the number of speakers and changes in speakers, either to a different individual or movement by the same speaker, based on the locations of the speakers and based on the overlap of the current and intended scenes, a decision is made whether to perform a smooth transition or do a cut. It has been determined that the decision on cut versus smooth transition is preferably based on the location of the center of the intended new scene versus the boundaries of the current scene, a cut used if the center is outside the boundaries and a smooth transition if inside. If a smooth transition, an easing function, such as ease in ease out, is performed to change the scene. A preferred value for the smooth transition is to perform the transition over 80 frames.

Abstract: The frame or image of a video stream of a videoconference is divided into a series of segments for analysis. There is a primary grid, which covers the entire frame, and an alternate grid, which is shifted from the primary grid. Each segment is small enough to allow a neural network to efficiently operate on the segment without requiring downsampling. By operating on full resolution images, a participant can be identified at a greater distance from the camera. The entire frame is analyzed at a lower frequency, such as once per five seconds, but each segment containing a participant in the conference is scanned at a higher frequency, such as once per second, to maintain responsiveness to participant movement but also allow the full resolution operation.

Abstract: Scene changes that are done pleasingly and without user input or control. Based on the number of speakers and changes in speakers, either to a different individual or movement by the same speaker, based on the locations of the speakers and based on the overlap of the current and intended scenes, a decision is made whether to perform a smooth transition or do a cut. It has been determined that the decision on cut versus smooth transition is preferably based on the location of the center of the intended new scene versus the boundaries of the current scene, a cut used if the center is outside the boundaries and a smooth transition if inside. If a smooth transition, an easing function, such as ease in ease out, is performed to change the scene. A preferred value for the smooth transition is to perform the transition over 80 frames.

Abstract: A method of view selection in a teleconferencing environment includes receiving a frame of image data from an optical sensor such as a camera, detecting one or more conference participants within the frame of image data, and identifying an interest region for each of the conference participants. Identifying the interest region comprises estimating head poses of participants to determine where a majority of the participants are looking and determining if there is an object in that area. If a suitable object is in the area at which the participants are looking, such as a whiteboard or another person, the image data corresponding to the object will be displayed on a display device or sent to a remote teleconference endpoint.

Abstract: A method for determining camera framing in a teleconferencing system comprises a process loop which includes acquiring an audio-visual frame from a captured a video data frame; detecting objects and extracting image features of the objects within the video data frame, ingesting the audio-visual frame into a context-based audio-visual map in an intelligent manner, and selecting targets from within the map for inclusion in an audio-video stream for transmission to a remote endpoint.

Abstract: Utilization of a state machine to determine participant framing. The states include empty room, group framing, any talker, conversation mode and unambiguous talker. In empty room state, the conference room is framed. In group framing state, any participants in the room are framed. In any talker state, the talking participant is framed. In conversation mode state, all talking participants are framed. In unambiguous talker state, the single talking participant is framed. Various framing conditions define transitions between the states. Conditions include, presence of participants, which and number of participants that are talking for how long, system mute and far site talking. The conversation states and conditions and framing decisions provide a fully automated framing mechanism to provide pleasant framing of the individuals in the near site or end for any of the conditions relating to number of talkers, participants and the like.

Abstract: A method of view selection in a teleconferencing environment includes receiving a frame of image data from an optical sensor such as a camera, detecting one or more conference participants within the frame of image data, and identifying an interest region for each of the conference participants. Identifying the interest region comprises estimating head poses of participants to determine where a majority of the participants are looking and determining if there is an object in that area. If a suitable object is in the area at which the participants are looking, such as a whiteboard or another person, the image data corresponding to the object will be displayed on a display device or sent to a remote teleconference endpoint.

Abstract: Scene changes that are done pleasingly and without user input or control. Based on the number of speakers and changes in speakers, either to a different individual or movement by the same speaker, based on the locations of the speakers and based on the overlap of the current and intended scenes, a decision is made whether to perform a smooth transition or do a cut. It has been determined that the decision on cut versus smooth transition is preferably based on the location of the center of the intended new scene versus the boundaries of the current scene, a cut used if the center is outside the boundaries and a smooth transition if inside. If a smooth transition, an easing function, such as ease in ease out, is performed to change the scene. A preferred value for the smooth transition is to perform the transition over 8o frames.

Abstract: A method for determining camera framing in a teleconferencing system comprises a process loop which includes acquiring an audio-visual frame from a captured a video data frame; detecting objects and extracting image features of the objects within the video data frame, ingesting the audio-visual frame into a context-based audio-visual map in an intelligent manner, and selecting targets from within the map for inclusion in an audio-video stream for transmission to a remote endpoint.

Abstract: Scene changes that are done pleasingly and without user input or control. Based on the number of speakers and changes in speakers, either to a different individual or movement by the same speaker, based on the locations of the speakers and based on the overlap of the current and intended scenes, a decision is made whether to perform a smooth transition or do a cut. It has been determined that the decision on cut versus smooth transition is preferably based on the location of the center of the intended new scene versus the boundaries of the current scene, a cut used if the center is outside the boundaries and a smooth transition if inside. If a smooth transition, an easing function, such as ease in ease out, is performed to change the scene. A preferred value for the smooth transition is to perform the transition over 80 frames.

Abstract: Systems and methods to integrate or interface video conference rooms, video conference infrastructure equipment and electronic calendars are disclosed. Example embodiments include providing: security for meeting information; interfaces for personal electronic calendar applications and video conferencing equipment; and implementations for video conferencing infrastructure components to integrate with meeting scheduling software. Also disclosed are systems and methods for making information automatically available via interfaces to the scheduling software (e.g., via a plug-in or extension to the regular interface software and integration modules executing in video conferencing infrastructure equipment) without a change in how a user interacts with existing meeting scheduling software.

Abstract: Systems and methods to integrate or interface video conference rooms, video conference infrastructure equipment and electronic calendars are disclosed. Example embodiments include providing: security for meeting information; interfaces for personal electronic calendar applications and video conferencing equipment; and implementations for video conferencing infrastructure components to integrate with meeting scheduling software. Also disclosed are systems and methods for making information automatically available via interfaces to the scheduling software (e.g., via a plug-in or extension to the regular interface software and integration modules executing in video conferencing infrastructure equipment) without a change in how a user interacts with existing meeting scheduling software.