Abstract: A method for making a zoomable user interface for a television (TV) comprises the steps of: assigning a plurality of tags to a video title; collecting tags of a plurality of video titles of a video group; building a tag-relationship map based on a relationship of the tags, wherein the tag-relationship map represents a hierarchical structure of the tags in the video group; building a plurality of tag trees based on the hierarchical structure of the tags; calculating each total distance of each of the plurality of tag trees, wherein the total distance of the tag tree is a sum of distances from a root node to all nodes in the tag tree; and displaying the zoomable user interface having a minimum value of the total distance of the tag tree.

Abstract: A method for displaying digital content in an interactive navigation interface is provided. The method includes: displaying a plurality of top-level category groups on a display screen; receiving a user pointing operation selecting one of the top-level category groups from a remote controller; providing visual feedback for the selected top-level category group; and in response to a zoom-in operation on the selected top-level category group received from the remote controller, displaying a plurality of digital content objects within the selected top-level category group.

Abstract: A privacy reserving camera including a frame buffer, an object detection and tracking module, an object identification module, an instance segmentation module and an object removal and inpainting module. The frame buffer stores frames of a video stream captured by the privacy preserving camera. The object detection and tracking module detects a moving object in the video stream. The object identification module determines whether the moving object is a privacy protecting object. The instance segmentation module parses the moving object in the video stream. The object removal and inpainting module removes at least a portion of the moving object from the video stream, fills an area of the removed moving object in the video stream with a compensated patch from a neighbor frame of the video stream, and outputs a modified video stream having at least a portion of the moving object replaced with the compensated patch.

Abstract: A method for displaying digital content in an interactive navigation interface is provided. The method includes: displaying a plurality of top-level category groups on a display screen; receiving a user pointing operation selecting one of the top-level category groups from a remote controller; providing visual feedback for the selected top-level category group; and in response to a zoom-in operation on the selected top-level category group received from the remote controller, displaying a plurality of digital content objects within the selected top-level category group.

Abstract: A privacy reserving camera including a frame buffer, an object detection and tracking module, an object identification module, an instance segmentation module and an object removal and inpainting module. The frame buffer stores frames of a video stream captured by the privacy preserving camera. The object detection and tracking module detects a moving object in the video stream. The object identification module determines whether the moving object is a privacy protecting object. The instance segmentation module parses the moving object in the video stream. The object removal and inpainting module removes at least a portion of the moving object from the video stream, fills an area of the removed moving object in the video stream with a compensated patch from a neighbor frame of the video stream, and outputs a modified video stream having at least a portion of the moving object replaced with the compensated patch.

Abstract: A method is provided for recommending a desired func sequence schedule to a user. The method includes obtaining at least one user intention inputted by the user; creating a func sequence including a plurality of funcs for fulfilling the user intention; and determining user preference on scheduling of the plurality of funcs in the func sequence. The method also includes determining a dependency relationship among the plurality of funcs in the func sequence; and determining scheduling constraints to the scheduling of the plurality of funcs in the func sequence. The method further includes, based on the user preference, the dependency relationship, and the scheduling constraints, performing an optimization process to determine a desired schedule for the plurality of funcs in the func sequence to fulfill the user intention; and recommending the func sequence with the desired schedule to the user.

Abstract: A method for optimized wake-up strategy via sleeping stage prediction with recurrent neural networks is provided. The method comprises receiving a set of raw sensory data of a user's sleeping status; translating the set of raw sensory data into a plurality of sleeping stages; saving the plurality of sleeping stages as a history of sleeping stages; determining a current sleeping stage; predicting a plurality of future sleeping stages of the user in an upcoming interval; calculating a wakeup score of the current sleeping stage and a wakeup score of each future sleeping stage; determining an optimized sleeping stage for wake up based on the wakeup score through comparing the wakeup score of the current sleeping stage and the wakeup scores of the future sleeping stages; and triggering an alarm impulse to wake up the user at the optimized sleeping stage for wake up.

Abstract: A method is provided for providing desired application usage experience to a user on a user device. The method includes automatically converting a plurality of App pages into FUNC descriptors to be stored, wherein each FUNC descriptor includes at least non-functional properties of a FUNC, capability of the FUNC, and an interface of the FUNC to invoke the FUNC. The method also includes obtaining at least one user intention inputted by the user; determining at least one FUNC descriptor from the FUNC descriptors for fulfilling the user intention; recommending to the user with at least one FUNC corresponding to the at least one FUNC descriptor by rendering the at least one FUNC descriptor on the user device; and invoking the at least one FUNC directly from the user device without going through a launching page of an App page providing the at least one FUNC.

Abstract: A method for making a zoomable user interface for a television (TV) comprises the steps of: assigning a plurality of tags to a video title; collecting tags of a plurality of video titles of a video group; building a tag-relationship map based on a relationship of the tags, wherein the tag-relationship map represents a hierarchical structure of the tags in the video group; building a plurality of tag trees based on the hierarchical structure of the tags; calculating each total distance of each of the plurality of tag trees, wherein the total distance of the tag tree is a sum of distances from a root node to all nodes in the tag tree; and displaying the zoomable user interface having a minimum value of the total distance of the tag tree.

Abstract: The present disclosure provides a method and device for visual appearance based person identity inference. The method may include obtaining a plurality of input images. The input images include a gallery set of images containing, persons-of-interest and a probe set of images containing person detections, and one input image corresponds to one person. The method may further include extracting N feature maps from the input images using a Deep Neural Network, N being a natural number; constructing N structure samples of the N feature maps using conditional random field (CRF) graphical models; learning the N structure samples from an implicit common latent feature space embedded in the N structure samples; and according to the learned structures, identifying one or more images from the probe set containing a same person-of-interest as an image in the gallery set.

Abstract: A system and a method for vision-based self-stabilization by deep gated recurrent Q-networks (DGRQNs) for unmanned arial vehicles (UAVs) are provided. The method comprises receiving a plurality of raw images captured by a camera installed on a UAV; receiving an initial reference image for stabilization and obtaining an initial camera pose from the initial reference image; extracting a fundamental matrix between consecutive images and estimating a current camera pose relative to the initial camera pose, wherein the camera pose includes an orientation and a location of the camera; based on the estimated current camera pose, predicting an action to counteract a lateral disturbance of the UAV based on the DGRQNs; and based on the predicted action to counteract the lateral disturbance of the UAV, driving the UAV back to the initial camera pose.

Abstract: In accordance with some embodiments of the disclosed subject matter, a method and a system for recommending and sharing customized multimedia route planning are provided. The method includes: receiving a query image from a user, the query image containing an object-of-interest of the user, performing an integrative segmentation process to determine one or more contours of the object-of-interest in the query image; determining a route having a maximum area overlap with the one or more contours of object-of-interest on a map image; generating an output image including the object-of-interest and the route; and recommending the output image to the user, and sharing the output image on a social network platform.

Abstract: The present disclosure provides a method for grouping devices in a same place. A plurality of devices may be connected to a server. The devices may be classified two types based on mobility, including a stable device type and a mobile device type. One or more neighborhood devices associated with at least one stable device may be identified based on status information of the plurality of devices. The at least one stable device and the one or more neighborhood devices may respectively record an audio sample at a set time. A similarity score between the audio sample of the one or more neighborhood devices and the audio sample of the at least one stable device may be calculated. Further, one or more devices among the one or more neighborhood devices may be determined as in a same space as the at least one stable device based on the similarity score.

Abstract: A method for a conversational recommender system includes receiving at least one user preference from a user, generating an abstract explanation reflecting the at least one user preference and based on the abstract explanation, and retrieving a list of items to be evaluated by the user from an item database. The items on the list have a relevance with the user preference. The method further includes determining an initial ranking order of the items to be evaluated by the user, and ranking the items to be evaluated by the user based on a random walk algorithm and the initial ranking order. Items not ranked are remaining items. The method further includes receiving a user feedback of the ranked item, and re-ranking the remaining items based on the user feedback of the ranked items and the random walk algorithm.

Abstract: A companion launcher comprises: a context-sensing module configured to sense context information and user behavior; a moment-recognition engine configured to recognize a moment based on the context information and user behavior, the moment indicating a timing that a user of a mobile device has a predicted need; a FUNN-recommending module configured to: mine a plurality of functional tools and/or services of one or more applications based on the moment, the plurality of functional tools and/or services of one or more applications being linked to the predicted need of the user, and recommend a FUNN to the user through a user interface of the mobile device, the FUNN including a plurality of function accessing points to the plurality of functional tools and/or services of one or more applications; and a user-interaction module configured to launch the FUNN on the mobile device in response to receiving a user confirmation to the recommendation.

Abstract: A system and a method for enhancing target tracking via detector and tracker fusion for unmanned aerial vehicles (UAVs) are provided. The method comprises receiving at least one raw input image of objects to be detected; based on the at least one raw input image of objects, generating the objects' candidate information; based on the objects' candidate information, calculating location and velocity estimation of an object at a current timestamp based on a detector and tracker fusion; and based on the location and velocity estimation of the object at the current timestamp, predicting the location and velocity estimation of the object at a future timestamp.

Abstract: A method for personalized intelligent wake-up system based on multimodal deep neural network comprises monitoring a sleeping status of a user; obtaining a current sleeping-stage of the user within a current time frame and a prediction of a next sleeping-stage of the user for a next time frame; correcting the current sleeping-stage of the user through combining the current sleeping-stage and the prediction of the next sleeping-stage; determining a wake up strategy for the current time frame; determining a relationship between each of a plurality of alarm impulses adopted to wake up the user and a corresponding reaction of the user; identifying a change in the current sleeping-stage for the current time frame; determining an alarm impulse to be triggered for waking up the user; and triggering the determined alarm impulse.

Abstract: An information presentation method includes steps of providing a plurality of displays comprising a main screen and at least one picture-in-picture (PiP) window, the main screen having a first displaying region for displaying a first content in the first displaying region, and the at least one PiP window displaying a second content in at least a portion of the first displaying region; providing a surrounding display circumambiently located around the plurality of displays for displaying a third content; determining an immersive impact region and a non-immersive impact region in the plurality of displays; calculating an immersive effect (IE) of the plurality of displays; and displaying the first content or the second content on the immersive impact region and the non-immersive impact region with different intensities based on the IE.

Abstract: A face detection and tracking method of a robotic device. The method includes obtaining a video frame from a camera of the robotic device; performing a face detection process on the video frame to detect one or more faces in the video frame and, after the face detection process, identifying the detected one or more faces in the video frame. The method also includes performing a vision-based tracking process to track the identified one or more faces using a combination of a feature points tracker and a correlation tracker and, after the vision-based tracking process, performing a detection and tracking fusion process and providing desired target prediction of the identified one or more faces.

Abstract: A system and a method for vision-based self-stabilization by deep gated recurrent Q-networks (DGRQNs) for unmanned arial vehicles (UAVs) are provided. The method comprises receiving a plurality of raw images captured by a camera installed on a UAV; receiving an initial reference image for stabilization and obtaining an initial camera pose from the initial reference image; extracting a fundamental matrix between consecutive images and estimating a current camera pose relative to the initial camera pose, wherein the camera pose includes an orientation and a location of the camera; based on the estimated current camera pose, predicting an action to counteract a lateral disturbance of the UAV based on the DGRQNs; and based on the predicted action to counteract the lateral disturbance of the UAV, driving the UAV back to the initial camera pose.