Abstract: Embodiments are disclosed that relate to color identification. In one example, an image processing method comprises receiving an infrared (IR) image including a plurality of IR pixels, each IR pixel specifying one or more IR parameters of that IR pixel, identifying, in the IR image, IR-skin pixels that image human skin, identifying a skin tone of identified human skin based at least in part on the IR-skin pixels, the skin tone having one or more expected visible light (VL) parameters, receiving a VL image including a plurality of VL pixels, each VL pixel specifying one or more VL parameters of that VL pixel, identifying, in the VL image, VL-skin pixels that image identified human skin, and adjusting the VL image to increase a correspondence between the one or more VL parameters of the VL-skin pixels and the one or more expected VL parameters of the identified skin tone.

Abstract: The systems and methods generate geometric proxies for participants in an online communication session, where each geometric proxy is a geometric representation of a participant and each geometric proxy is generated from acquired depth information and is associated with a particular virtual box. The systems and methods also include generating a scene geometry that visually simulates an in-person meeting of the participants where the scene geometry includes the geometric proxies, and the virtual boxes of the geometric proxies are aligned within the scene geometry based on a number of the participants and a reference object to which the virtual boxes are aligned. In addition, the systems and methods cause a display of the scene geometry with the geometric proxies, where the display of a particular geometric proxy includes a video of the participant corresponding to the particular geometric painted onto the particular geometric proxy.

Abstract: The electronic devices described herein are configured to enhance user experience associated with a network connection when transitioning the network connection between access points. Determinations to scan for available access points and transfer the network connection to an alternative access point are based on connection attributes and/or access point attributes that are compared to scan criteria and transfer criteria. Further, the scan criteria and transfer criteria are updated, or adjusted, according to machine learning techniques such that the determinations to scan for access points and transfer between access points are tuned on a per-device and/or per-user level to fit patterns of use of a particular device and/or user. Over time, the updates to the scan criteria and transfer criteria based on machine learning provide an increasingly consistent, high quality user experience while roaming efficiently between access points.

Abstract: The systems and methods generate geometric proxies for participants in an online communication session, where each geometric proxy is a geometric representation of a participant and each geometric proxy is generated from acquired depth information and is associated with a particular virtual box. The systems and methods also include generating a scene geometry that visually simulates an in-person meeting of the participants where the scene geometry includes the geometric proxies, and the virtual boxes of the geometric proxies are aligned within the scene geometry based on a number of the participants and a reference object to which the virtual boxes are aligned. In addition, the systems and methods cause a display of the scene geometry with the geometric proxies, where the display of a particular geometric proxy includes a video of the participant corresponding to the particular geometric painted onto the particular geometric proxy.

Abstract: Embodiments are disclosed that relate to color identification. In one example, an image processing method comprises receiving an infrared (IR) image including a plurality of IR pixels, each IR pixel specifying one or more IR parameters of that IR pixel, identifying, in the IR image, IR-skin pixels that image human skin, identifying a skin tone of identified human skin based at least in part on the IR-skin pixels, the skin tone having one or more expected visible light (VL) parameters, receiving a VL image including a plurality of VL pixels, each VL pixel specifying one or more VL parameters of that VL pixel, identifying, in the VL image, VL-skin pixels that image identified human skin, and adjusting the VL image to increase a correspondence between the one or more VL parameters of the VL-skin pixels and the one or more expected VL parameters of the identified skin tone.

Abstract: Implementations provide an in-person communication experience by generating a changable virtual viewpoint for a participant in an online communication. For instance, techniques described herein capture visual data about participants in an online communication, and create a realistic geometric proxy from the visual data. A virtual scene geometry is generated that mimics an arrangement of an in-person meeting for the online communication. According to various implementations, a virtual viewpoint displays a changing virtual viewpoint, such as based on a change in position of a participants face.

Abstract: Embodiments are disclosed that relate to color identification. In one example, an image processing method comprises receiving an infrared (IR) image including a plurality of IR pixels, each IR pixel specifying one or more IR parameters of that IR pixel, identifying, in the IR image, IR-skin pixels that image human skin, identifying a skin tone of identified human skin based at least in part on the IR-skin pixels, the skin tone having one or more expected visible light (VL) parameters, receiving a VL image including a plurality of VL pixels, each VL pixel specifying one or more VL parameters of that VL pixel, identifying, in the VL image, VL-skin pixels that image identified human skin, and adjusting the VL image to increase a correspondence between the one or more VL parameters of the VL-skin pixels and the one or more expected VL parameters of the identified skin tone.

Abstract: Examples of the disclosure dynamically scale receive window auto-tuning. Tuning data is obtained, including the number of bytes in a receive buffer and the distribution of receive packets over time. Aspects of the disclosure use this tuning data to determine rates at which one or more applications on the receiving computer are consuming data and adjust or maintain the receive buffer accordingly in a dynamic manner to scale a receive window to current conditions.

Abstract: The electronic devices described herein are configured to enhance user experience associated with a network connection when transitioning the network connection between access points. Determinations to scan for available access points and transfer the network connection to an alternative access point are based on connection attributes and/or access point attributes that are compared to scan criteria and transfer criteria. Further, the scan criteria and transfer criteria are updated, or adjusted, according to machine learning techniques such that the determinations to scan for access points and transfer between access points are tuned on a per-device and/or per-user level to fit patterns of use of a particular device and/or user. Over time, the updates to the scan criteria and transfer criteria based on machine learning provide an increasingly consistent, high quality user experience while roaming efficiently between access points.

Abstract: A three-dimensional shape parameter computation system and method for computing three-dimensional human head shape parameters from two-dimensional facial feature points. A series of images containing a user's face is captured. Embodiments of the system and method deduce the 3D parameters of the user's head by examining a series of captured images of the user over time and in a variety of head poses and facial expressions, and then computing an average. An energy function is constructed over a batch of frames containing 2D face feature points obtained from the captured images, and the energy function is minimized to solve for the head shape parameters valid for the batch of frames. Head pose parameters and facial expression and animation parameters can vary over each captured image in the batch of frames. In some embodiments this minimization is performed using a modified Gauss-Newton minimization technique using a single iteration.

Abstract: Implementations provide an in-person communication experience by generating a changable virtual viewpoint for a participant in an online communication. For instance, techniques described herein capture visual data about participants in an online communication, and create a realistic geometric proxy from the visual data. A virtual scene geometry is generated that mimics an arrangement of an in-person meeting for the online communication. According to various implementations, a virtual viewpoint displays a changing virtual viewpoint, such as based on a change in position of a participants face.

Abstract: Examples of the disclosure dynamically scale receive window auto-tuning. Tuning data is obtained, including the number of bytes in a receive buffer and the distribution of receive packets over time. Aspects of the disclosure use this tuning data to determine rates at which one or more applications on the receiving computer are consuming data and adjust or maintain the receive buffer accordingly in a dynamic manner to scale a receive window to current conditions.

Abstract: A perspective-correct communication window system and method for communicating between participants in an online meeting, where the participants are not in the same physical locations. Embodiments of the system and method provide an in-person communications experience by changing virtual viewpoint for the participants when they are viewing the online meeting. The participant sees a different perspective displayed on a monitor based on the location of the participant's eyes. Embodiments of the system and method include a capture and creation component that is used to capture visual data about each participant and create a realistic geometric proxy from the data. A scene geometry component is used to create a virtual scene geometry that mimics the arrangement of an in-person meeting. A virtual viewpoint component displays the changing virtual viewpoint to the viewer and can add perceived depth using motion parallax.

Abstract: A method for operating an image processing device coupled to a color camera and a depth camera is provided. The method includes receiving a color image of a 3-dimensional scene from a color camera, receiving a depth map of the 3-dimensional scene from a depth camera, generating an aligned 3-dimensional face mesh from a plurality of color images received from the color camera indicating movement of a subject's head within the 3-dimensional scene and form the depth map, determining a head region based the depth map, segmenting the head region into a plurality of facial sections based on both the color image, depth map, and the aligned 3-dimensional face mesh, and overlaying the plurality of facial sections on the color image.

Abstract: Technology is disclosed for automatically generating a facial avatar resembling a user in a defined art style. One or more processors generate a user 3D head model for the user based on captured 3D image data from a communicatively coupled 3D image capture device. A set of user transferable head features from the user 3D head model are automatically represented by the one or more processors in the facial avatar in accordance with rules governing transferable user 3D head features. In some embodiments, a base or reference head model of the avatar is remapped to include the set of user head features. In other embodiments, an avatar head shape model is selected based on the user 3D head model, and the transferable user 3D head features are represented in the avatar head shape model.

Abstract: Technology is disclosed for automatically generating a facial avatar resembling a user in a defined art style. One or more processors generate a user 3D head model for the user based on captured 3D image data from a communicatively coupled 3D image capture device. A set of user transferable head features from the user 3D head model are automatically represented by the one or more processors in the facial avatar in accordance with rules governing transferable user 3D head features. In some embodiments, a base or reference head model of the avatar is remapped to include the set of user head features. In other embodiments, an avatar head shape model is selected based on the user 3D head model, and the transferable user 3D head features are represented in the avatar head shape model.

Abstract: Technology is disclosed for automatically generating a facial avatar resembling a user in a defined art style. One or more processors generate a user 3D head model for the user based on captured 3D image data from a communicatively coupled 3D image capture device. A set of user transferable head features from the user 3D head model are automatically represented by the one or more processors in the facial avatar in accordance with rules governing transferable user 3D head features. In some embodiments, a base or reference head model of the avatar is remapped to include the set of user head features. In other embodiments, an avatar head shape model is selected based on the user 3D head model, and the transferable user 3D head features are represented in the avatar head shape model.

Abstract: A perspective-correct communication window system and method for communicating between participants in an online meeting, where the participants are not in the same physical locations. Embodiments of the system and method provide an in-person communications experience by changing virtual viewpoint for the participants when they are viewing the online meeting. The participant sees a different perspective displayed on a monitor based on the location of the participant's eyes. Embodiments of the system and method include a capture and creation component that is used to capture visual data about each participant and create a realistic geometric proxy from the data. A scene geometry component is used to create a virtual scene geometry that mimics the arrangement of an in-person meeting. A virtual viewpoint component displays the changing virtual viewpoint to the viewer and can add perceived depth using motion parallax.

Abstract: A perspective-correct communication window system and method for communicating between participants in an online meeting, where the participants are not in the same physical locations. Embodiments of the system and method provide an in-person communications experience by changing virtual viewpoint for the participants when they are viewing the online meeting. The participant sees a different perspective displayed on a monitor based on the location of the participant's eyes. Embodiments of the system and method include a capture and creation component that is used to capture visual data about each participant and create a realistic geometric proxy from the data. A scene geometry component is used to create a virtual scene geometry that mimics the arrangement of an in-person meeting. A virtual viewpoint component displays the changing virtual viewpoint to the viewer and can add perceived depth using motion parallax.

Abstract: A controlled three-dimensional (3D) communication endpoint system and method for simulating an in-person communication between participants in an online meeting or conference and providing easy scaling of a virtual environment when additional participants join. This gives the participants the illusion that the other participants are in the same room and sitting around the same table with the viewer. The controlled communication endpoint includes a plurality of camera pods that capture video of a participant from 360 degrees around the participant. The controlled communication endpoint also includes a display device configuration containing display devices placed at least 180 degrees around the participant and display the virtual environment containing geometric proxies of the other participants. Placing the participants at a round virtual table and increasing the diameter of the virtual table as additional participants are added easily achieves scalability.