Abstract: A method may include obtaining a frame of a video stream of multiple video streams of a video conference, obtaining face detection information identifying a face size and a face position of at least one face detected in the frame, and cropping and scaling the frame according to at least one crop and scale parameter using the face detection information to obtain a modified first frame. The at least one crop and scale parameter is based on frames of the multiple video streams. The frames include the frame. The method may further include presenting the modified frame.

Abstract: A camera shooting program control method, the method is applied to terminal equipment, and the method includes obtaining motion parameters collected by at least one motion sensor in response to a situation that the terminal equipment enters a camera shooting program from a screen locking state, and determining a motion state of the terminal equipment according to the motion parameters collected by the at least one motion sensor; obtaining a mistaken touch rate of a display screen of the terminal equipment in response to the motion state of the terminal equipment being a moving state; and exiting the camera shooting program in response to a condition that the mistaken touch rate is greater than a preset first proportion threshold value.

Abstract: A method (1000) for operating cameras (202) in a cascaded network (100), comprising: capturing a first view (1200) with a first lens (326) having a first focal point (328) and a first centroid (352), the first view (1200) depicting a subject (1106); capturing a second view (1202) with a second lens (326) having a second focal point (328) and a second centroid (352); detecting a first location of the subject (1106), relative the first lens (326), wherein detecting the first location of the subject (1106), relative the first lens (326), is based on audio captured by a plurality of microphones (204); estimating a second location of the subject (1106), relative the second lens (326), based on the first location of the subject (1106) relative the first lens (326); selecting a portion (1206) of the second view (1202) as depicting the subject (1106) based on the estimate of the second location of the subject (1106) relative the second lens (326).

Abstract: A method of tracking a presenter during a videoconference, comprising: detecting, in a data stream, data indicating a presence of one or more persons at a videoconferencing endpoint, the data stream comprising a plurality of frames; detecting tracker initiation data (e.g., that a person is actively talking, or that a person has raised their hand, or both) in the data stream; determining a subject of interest responsive to the tracker initiation data, and tracking the subject of interest as they move while making a presentation.

Abstract: A videoconferencing endpoint is described that uses a cascading sequence of convolutional neural networks to perform face detection and upper body detection of participants in a videoconference at the endpoint, where at least one member of the sequence of neural networks performs upper body detection, and where the final member of the sequence of neural networks performs face detection based on the results of the upper body detection. The models of the neural networks are trained on both large datasets of faces well as images that have been distorted by a wide-angle camera of the videoconferencing endpoint.

Abstract: A method may include identifying primly camera speaker coordinates of a location of a speaker relative to a primary camera, converting the primary camera speaker coordinates to secondary camera speaker coordinates, the secondary camera speaker coordinates being of the location and relative to a secondary camera, and adjusting, using the secondary camera speaker coordinates of the secondary camera, a view setting of the secondary camera to track the location of the speaker.

Abstract: A real-time method (600) for enhancing facial images (102). Degraded images (102) of a person—such as might be transmitted during a videoconference—are rectified based on a single high definition reference image (604) of a person who is talking. Facial landmarks (501) are used to map (210) image data from the reference image (604) to an intervening image (622) having a landmark configuration like that in a degraded image (102). The degraded images (102) and their corresponding intervening images (622) are blended using an artificial neural network (800, 900) to produce high-quality images (108) of the person who is speaking during a videoconference.

Abstract: Systems and methods are provided for identifying and displaying whiteboard text and/or an active speaker in a video-based presentation, e.g., a video conference. Video images of an environment including a whiteboard may be captured by a video camera system. The video images may be analyzed to detect at least one text-containing area in the environment. Each text-containing area may be analyzed to determine whether it is an area of a whiteboard. When a text-containing area is identified as a whiteboard area, an area of view including the text-containing whiteboard area may be selected for display, e.g., a subset of the full frame captured by the video system. A video feed from the video camera system may be controlled to display the selected area of view at a client device, to provide a useful view of the whiteboard text and/or a speaking person located near the whiteboard text.

Abstract: A method may include calculating a color gain by applying an automatic white balance (AWB) algorithm to a video frame of a video feed, calculating an illumination color by applying a machine learning model to the video frame, transforming the illumination color into an equivalent color gain, determining that a difference between the color gain and the equivalent color gain exceeds a difference threshold, reversing an effect of the illumination color on the video frame based on the threshold being exceeded to obtain a corrected video frame, and transmitting the corrected video frame to an endpoint.

Abstract: A camera shooting program control method, the method is applied to terminal equipment, and the method includes obtaining motion parameters collected by at least one motion sensor in response to a situation that the terminal equipment enters a camera shooting program from a screen locking state, and determining a motion state of the terminal equipment according to the motion parameters collected by the at least one motion sensor; obtaining a mistaken touch rate of a display screen of the terminal equipment in response to the motion state of the terminal equipment being a moving state; and exiting the camera shooting program in response to a condition that the mistaken touch rate is greater than a preset first proportion threshold value.

Abstract: An edge anti-false-touch method is provided. The method includes: obtaining, by a mobile terminal, a face direction of a user of a mobile terminal, and obtaining gesture data of the mobile terminal; determining, by the mobile terminal according to the face direction and the gesture data, a screen display state of the mobile terminal; and adjusting, by the mobile terminal based on the screen display state, a size of an anti-false-touch region of the mobile terminal to trigger an anti-false-touch response of the anti-false-touch region.

Abstract: A method performs zooming based on gesture detection. A visual stream is presented using a first zoom configuration for a zoom state. An attention gesture is detected from a set of first images from the visual stream. The zoom state is adjusted from the first zoom configuration to a second zoom configuration to zoom in on a person in response to detecting the attention gesture. The visual stream is presented using the second zoom configuration after adjusting the zoom state to the second zoom configuration. Whether the person is speaking is determined, from a set of second images from the visual stream. The zoom state is adjusted to the first zoom configuration to zoom out from the person in response to determining that the person is not speaking. The visual stream is presented using the first zoom configuration after adjusting the zoom state to the first zoom configuration.

Abstract: A method of tracking a presenter during a videoconference, comprising: detecting, in a data stream, data indicating a presence of one or more persons at a videoconferencing endpoint, the data stream comprising a plurality of frames; detecting tracker initiation data (e.g., that a person is actively talking, or that a person has raised their hand, or both) in the data stream; determining a subject of interest responsive to the tracker initiation data, and tracking the subject of interest as they move while making a presentation.

Abstract: One illustrative method includes: (a) obtaining a video frame sequence having alternating fast and slow exposure frames; (b) applying a convolutional neural network twice to each frame in the video frame sequence, first when the frame is paired with a preceding frame, and again when the frame is paired with a subsequent frame, each time converting a pair of fast and slow exposure frames into an enhanced dynamic range video frame; and (c) outputting an enhanced video frame sequence. In another illustrative method, the convolutional neural network converts pairs of adjacent fast and slow exposure frames into corresponding pairs of enhanced dynamic range video frames. In yet another illustrative method, neighboring video frames for each given video frame are interpolated to form a fast and slow exposure frame pair, which the convolutional neural network converts into a corresponding enhanced dynamic range video frame.

Abstract: A teleconferencing system (100) comprises: a first camera (202) including a first lens (326) having a first focal point (328) and a first centroid (352), and configured to capture a first view (900) corresponding to a subject (702); a second lens (326) having a second focal point (328) and a second centroid (352), and configured to capture a second view (902) corresponding to the subject (702); and a processor (206) coupled to the first camera device (202) and the second camera device (202). The processor (206) is configured to: estimate a first orientation (351) of the subject (702) relative the first lens (326) and a second orientation (351) of the subject relative the second lens (326); and determine that the first orientation (351) is more closely aligned with a first line (307) from the first centroid (352) to the first focal point (328) than is the second orientation (351) aligned with a second line (307) from the second centroid (352) to the second focal point (328).

Abstract: A method (1000) for operating cameras (202) in a cascaded network (100), comprising: capturing a first view (1200) with a first lens (326) having a first focal point (328) and a first centroid (352), the first view (1200) depicting a subject (1106); capturing a second view (1202) with a second lens (326) having a second focal point (328) and a second centroid (352); detecting a first location of the subject (1106), relative the first lens (326), wherein detecting the first location of the subject (1106), relative the first lens (326), is based on audio captured by a plurality of microphones (204); estimating a second location of the subject (1106), relative the second lens (326), based on the first location of the subject (1106) relative the first lens (326); selecting a portion (1206) of the second view (1202) as depicting the subject (1106) based on the estimate of the second location of the subject (1106) relative the second lens (326).

Abstract: A method may include obtaining a frame of a video stream of multiple video streams of a video conference, obtaining face detection information identifying a face size and a face position of at least one face detected in the frame, and cropping and scaling the frame according to at least one crop and scale parameter using the face detection information to obtain a modified first frame. The at least one crop and scale parameter is based on frames of the multiple video streams. The frames include the frame. The method may further include presenting the modified frame.

Abstract: A method for tracking movement of a presenter during a videoconference comprises tracking the location of subject of interest; quickly detecting error conditions—such as can occur when a subject of interest moves too quickly or when a subject of interest walks past another person—and quickly reinitiating the tracking sequence to help ensure a positive videoconferencing experience.

Abstract: A method for tracking movement of a presenter during a videoconference comprises tracking the location of subject of interest; quickly detecting error conditions—such as can occur when a subject of interest moves too quickly or when a subject of interest walks past another person—and quickly reinitiating the tracking sequence to help ensure a positive videoconferencing experience.

Abstract: A videoconferencing endpoint is described that uses a cascading sequence of convolutional neural networks to perform face detection and upper body detection of participants in a videoconference at the endpoint, where at least one member of the sequence of neural networks performs upper body detection, and where the final member of the sequence of neural networks performs face detection based on the results of the upper body detection. The models of the neural networks are trained on both large datasets of faces well as images that have been distorted by a wide-angle camera of the videoconferencing endpoint.