Abstract: According to this disclosure, an alternative to Document Object Model (DOM)-based co-browsing uses a cloud-based browser that is shared by the participants. The cloud browser enables the participants to view a website together but from within their own respective local browser resource. The cloud-based browser is made available by a service provider. Using a cloud browser and streaming of the website image (from the cloud browser) to both peers guarantees that the participants in the co-browse session see the same image, regardless of the local browsers that they use. Further, by transferring the actual user actions (e.g., clicks, taps, gestures, etc.) and executing them in the cloud browser as if the respective user is present in front of the cloud browser, the approach eliminates dependency on the site/page DOM. Rather, in this approach the cloud browser handles all code injections necessary for the streaming of the images to the endpoints.

Abstract: Techniques for performing key frame extraction, recording, and navigation in collaborative video presentations. The techniques include extracting a plurality of frames from media content at a predetermined rate, removing frame areas that do not correspond to a screen area for displaying electronic meeting/webinar content, de-duplicating the plurality of frames, identifying frames that correspond to the “slide type” or similar type of frames, and extracting key frames from the slide type of frames. The key frames can be recorded in a slide deck or other similar collection of key frames, as well as displayed as clickable thumbnails in a UI. By clicking or otherwise selecting a thumbnail representation of a selected key frame in the UI, or clicking-and-dragging a handle of a key frame locator bar to navigate the thumbnails to the selected key frame, users can quickly and more efficiently access desired slide presentation content from an electronic meeting/webinar.

Abstract: A technique for processing video includes receiving a pixel array, such as a block or layer of video content, as well as a mask that distinguishes masked, “don't-care” pixels in the pixel array from unmasked, “care” pixels. The technique encodes the pixel array by taking into consideration the care pixels only, without regard for the don't-care pixels. An encoder operating in this manner can produce a simplified encoding of the pixel array, which represents the care pixels to any desired level of precision, without regard for errors in the don't-care pixels, which are irrelevant to reconstruction. Further embodiments apply a polynomial transform in place of a frequency transform for encoding partially-masked video content, and/or video content meeting other suitable criteria.

Abstract: A technique for synchronizing video receivers with a video stream already in progress includes caching a key frame in a transport protocol component of a video communication system and providing the key frame on demand to any receiver attempting to join the stream and/or to rejoin the stream after an error, such as a dropped packet. Once a receiver obtains the key frame, the receiver requests that the source of the video stream issue a new sync-point frame, where the sync-point frame depends on the key frame but on no other frame for its content. The receiver may then proceed to display rendered video beginning with the sync-point frame.

Abstract: Technology for generating a consistently labeled training dataset. For each one of multiple previously labeled texts, a distance between the previously labeled text and a current text to be labeled is generated by comparing a list of tokens for the previously labeled text to a list of tokens for the current text to determine an overlap value equal to a number of tokens that match between the list of tokens for the previously labeled text and the list of tokens for the current text, and using the overlap value to calculate a distance between the previously labeled text and the current text that is inversely correlated to the overlap value. Previously labeled texts that are most similar to the current text are identified as those previously labeled texts having the shortest distances to the current text, and are displayed with their previously assigned labels in a label selection user interface.

Abstract: A password management service provides automated password management. In one embodiment, a method for automating password changes begins in response to a determination that automated password changes are authorized. In response, a data mining session is initiated. Within the data mining session, a set of third party applications or sites are identified. Then, and responsive to receipt of a password reset flow authorization, a password reset flow to one or more of the third party applications or sites is initiated by the service. Thereafter, and still within the data mining session, and for each of the one or more third party applications or sites, a determination is made whether a password reset confirmation link has been received by the service.

Abstract: A technique for reducing acoustic feedback in audio communications includes measuring variations in round-trip delay over an audio signal pathway. The technique varies a delay interval of an adjustable-delay element in real time based on the measured variations in round-trip delay, effectively canceling the delay variations. Further techniques are disclosed for detecting and eliminating howling frequencies which arise as a result of acoustic feedback in the audio signal pathway.

Abstract: Private real-time communications are provided between meeting attendees during a meeting using one or more augmented reality headsets. An augmented reality headset worn by a sending meeting attendee continuously monitors a current direction of the sending meeting attendee's gaze during the meeting, and detects that the sending meeting attendee has selected a receiving meeting attendee when the current direction of the sending attendee's gaze intersects with a location of the receiving meeting attendee. The augmented reality headset worn by the sending meeting attendee also receives a messaging command issued by the sending meeting attendee, also based on monitoring the sending meeting attendee's gaze during the meeting. In response to the messaging command, the augmented reality headset worn by the sending meeting attendee automatically transmits the private message to a device associated with the receiving meeting attendee, such as an augmented reality headset worn by the receiving meeting attendee.

Abstract: Techniques are provided for a source computer to generate an encoded video stream having layered sub-streams with differing bitrates while allowing a streaming server to intelligently distribute the appropriate sub-streams to recipients based on their available bandwidth. This may be accomplished by having the source computer generate and send metadata along with the encoded stream to allow the streaming server to detect which data packets belong to each sub-stream. The streaming server is then able to selectively send consistent video sub-streams at appropriate bitrates to each recipient.

Abstract: A network-accessible service enables participating end users to collaborate with one another over a network. End users have computing devices (e.g., computers, mobile phone, tablet devices, or the like) that include hardware and software to enable the device to access a network, such as the public Internet, a Wi-Fi network connected to the Internet, a 3G or higher wireless network connected to the Internet, a private network, or the like. The network-accessible service provides a publicly-available site (such as a Web site) or a local software application from which a first participating end user initiates a “meeting,” e.g., by selecting a “share” button. In response, the site or software application provides an HTTP link that includes a “meeting” code, which may be a one-time unique code. The first participating end user then shares the link with whomever he or she desires to collaborate. Upon receiving the link (e.g.

Abstract: Several methods are described herein to allow users to manage and synchronize views in a multi-user application with canvas (infinite or otherwise) and limited or unlimited zoom. Without these synchronization methods, coordinating views is very difficult, and may lead to chaos as one or more of the users gets lost on the canvas. A “View Sync” method provides for many-to-one synchronization, where many users are automatically synchronized with one user. A “Sync View” method provides for one-to-one synchronization, where one user is synchronized to another user's view. Preferably, off-screen “activity indicators” also are used within the viewport to provide visual cues regarding where other users are drawing or creating content on the canvas. Tapping on an activity indicator preferably performs one-time Sync View with the selected other user.

Abstract: A host-client protocol facilitates delivery of screen images from a host to a remote access client when the host content is continuous or large (such as high definition (HD) content). To facilitate this operation, the protocol implements several functions and operations that provide the desired result of an HD stream viewable on the remote access client. These functions and operations include one or more of: frame rate adjustment, rate control, dynamic adjustment to the stream based on client-provided screen resolution or changes, eliminating portions of the host screen that are not in view on the remote access client, implementing client-initiated zoom/pan operations on the host instead of the client, performing zoom/pan operations, dynamic control over screen updates, and (where multiple clients are receiving the stream) delivering streams to different clients at different frame rates.

Abstract: Subscriber (user) data is encrypted and stored in a service provider cloud in a manner such that the service provider is unable to decrypt and, as a consequence, to view, access or copy the data. Only the user knows a user-specific secret (e.g., a password) that is the basis of the encryption. The techniques herein enable the user to share his or her data, privately or publicly, without exposing the user-specific secret with anyone or any entity (such as the service provider).

Abstract: A host-client protocol facilitates delivery of screen images from a host to a remote access client when the host content is continuous or large (such as high definition (HD) content). To facilitate this operation, the protocol implements several functions and operations that provide the desired result of an HD stream viewable on the remote access client. These functions and operations include one or more of: frame rate adjustment, rate control, dynamic adjustment to the stream based on client-provided screen resolution or changes, eliminating portions of the host screen that are not in view on the remote access client, implementing client-initiated zoom/pan operations on the host instead of the client, performing zoom/pan operations, dynamic control over screen updates, and (where multiple clients are receiving the stream) delivering streams to different clients at different frame rates.

Abstract: Subscriber (user) data is encrypted and stored in a service provider cloud in a manner such that the service provider is unable to decrypt and, as a consequence, to view, access or copy the data. Only the user knows a user-specific secret (e.g., a password) that is the basis of the encryption. The techniques herein enable the user to share his or her data, privately or publicly, without exposing the user-specific secret with anyone or any entity (such as the service provider).

Abstract: A network-accessible service enables participating end users to collaborate with one another over a network. End users have computing devices that include hardware and software to enable the device to access a network. The service provides a publicly-available site or a local software application from which a first participating user initiates a “meeting,” e.g., by selecting a “share” button. In response, the site or software application provides an HTTP link that includes a “meeting” code, which may be a one-time unique code. The first participating user then shares the link with whomever he or she desires to collaborate. Upon receiving the link, a second participating user joins the meeting “on-the-fly” by simply selecting the link or navigating to the site and entering the “meeting” code. The service connects the second participating user to the meeting immediately and without requiring any registration, software download, or the like.

Abstract: A mobile device has a display that supports gesture-based control operations. Computers maintain connections to a remote access service. The user desires to copy and/or move resources between first and second of his connected computers. After receipt of data indicating a first gesture requesting an operation with respect to the resource, an overview of one or more target devices is displayed. The overview is a folder level view of all folders and files located on the target devices. In addition, an additional display element is shown. This element is a representation of the resource from which the operation has been requested. The element is displayed as an overlay, and it is adapted to be selectively positioned in the overview using a gesture. Upon receipt of data indicating a second gesture positioning the representation over an available location within a computer connected to the service, the operation is completed.

Abstract: Subscriber (user) data is encrypted and stored in a service provider cloud in a manner such that the service provider is unable to decrypt and, as a consequence, to view, access or copy the data. Only the user knows a user-specific secret (e.g., a password) that is the basis of the encryption. The techniques herein enable the user to share his or her data, privately or publicly, without exposing the user-specific secret with anyone or any entity (such as the service provider).

Abstract: A network-accessible service enables participating end users to collaborate with one another over a network. End users have computing devices that include hardware and software to enable the device to access a network. The service provides a publicly-available site or a local software application from which a first participating user initiates a “meeting,” e.g., by selecting a “share” button. In response, the site or software application provides an HTTP link that includes a “meeting” code, which may be a one-time unique code. The first participating user then shares the link with whomever he or she desires to collaborate. Upon receiving the link, a second participating user joins the meeting “on-the-fly” by simply selecting the link or navigating to the site and entering the “meeting” code. The service connects the second participating user to the meeting immediately and without requiring any registration, software download, or the like.

Abstract: A motion-based authentication method is operative in a mobile computing device having a display interface and that includes an accelerometer. Normally, the device software includes a locking mechanism that automatically locks the display interface after a configurable timeout. The authentication method operates to un-lock the display interface (and thus allow the user access to the device) by movement of the device in a predetermined series of physical movements and without display-based entry of a password or other access code on the display itself. In this manner, the user can un-lock the device without display-based entry of a password (on the display itself) by simply holding the device and performing the necessary movement(s) to generate the unique code.