Abstract: A method, device and computer-readable medium for generating a super-resolution version of a compressed video stream. By leveraging the motion information and residual information in compressed video streams, described examples are able to skip the time-consuming motion-estimation step for most frames and make the most use of the SR results of key frames. A key frame SR module generates SR versions of I-frames and other key frames of a compressed video stream using techniques similar to existing multi-frame approaches to VSR. A non-key frame SR module generates SR version of the non-key inter frames between these key frames by making use of motion information and residual information used to encode the inter frames in the compressed video stream.

Abstract: Methods, devices, and processor-readable media for adjusting the control-display gain of a gesture-controlled device are described. Adjusting the control-display gain may facilitate user interaction with content or UI elements rendered on a display screen of the gesture-controlled device. The control-display gain may be adjusted based on a property of how a mid-air dragging gesture is being performed by a user's hand. The property may be the location of the gesture, the orientation of the hand performing the gesture, or the velocity of the gesture. A hand that becomes stationary for a threshold time period while performing the dragging gesture may adjust the control-display gain to a different level. Control-display gain may be set to a different value based on the current velocity of the hand performing the gesture. The control-display gain levels may be selected from a continuous range of values or a set of discrete values. Devices for performing the methods are described.

Abstract: Methods and devices for generating a slow motion video segment are described. A first set of video frames captures a video view at a first resolution and at a first frame rate. A second set of video frames captures the video view at a second lower resolution, and at a second frame rate that is greater for at least a portion of the second set. At least two high resolution frames are identified in the first set for generating the slow motion video segment. One or more low resolution frames are identified in the second set corresponding to an inter-frame time period between the identified high resolution frames. The slow motion video segment is generated by generating at least one high resolution frame corresponding to the inter-frame time period using interpolation based on the identified high resolution frames and the identified low resolution frames.

Abstract: Methods, systems, and media for image searching are described. Images comprising one query image and a plurality of candidate images are received. For each candidate image, a first model similarity measure from an output of a first model configured for scene classification to perceive scenes in the images is determined. Further, for each candidate image of the plurality of candidate images, a second model similarity measure from the output of a second model configured for attribute classification to perceive attributes in the images is determined. For each candidate image of the plurality of candidate images, a similarity agglomerate index of a weighted aggregate of the first model similarity measure and the second model similarity measure is computed. The plurality of candidate images based on the respective similarity agglomerate index of each candidate image are ranked and a first ranked candidate images corresponding to the searched images are generated.

Abstract: Methods, devices, and processor-readable media for adjusting the control-display gain of a gesture-controlled device are described. Adjusting the control-display gain may facilitate user interaction with content or UI elements rendered on a display screen of the gesture-controlled device. The control-display gain may be adjusted based on a property of how a mid-air dragging gesture is being performed by a user's hand. The property may be the location of the gesture, the orientation of the hand performing the gesture, or the velocity of the gesture. A hand that becomes stationary for a threshold time period while performing the dragging gesture may adjust the control-display gain to a different level. Control-display gain may be set to a different value based on the current velocity of the hand performing the gesture. The control-display gain levels may be selected from a continuous range of values or a set of discrete values. Devices for performing the methods are described.

Abstract: Systems, methods, and computer media of processing a video are disclosed. An example method may include: receiving a plurality of video frames of a video; generating a plurality of first input features based on the plurality of video frames; generating a plurality of second input features based on reversing a temporal order of the plurality of first input features; generating a first set of joint attention features based on the plurality of first input features; generating a second set of joint attention features based on the plurality of second input features; and concatenating the first set of joint attention features and the second set of joint attention features to generate a final set of joint attention features.

Abstract: Methods, systems, and media for training deep neural networks for cross-domain few-shot classification are described. The methods comprise an encoder and a decoder of a deep neural network. The training of the autoencoder comprises two training stages. For each iteration in the first training stage, a batch of data samples from the source dataset are sampled and fed to the encoder to generate a plurality of source feature maps, then determining a first training stage loss, which updates the autoencoder's parameters. For each iteration in the second training stage, the novel dataset is split into a support set and a query set. The support set is fed to the encoder to determine a prototype for each class label. The query set is also fed to the encoder to calculate a query set metric classification loss. The query set metric classification loss updates the autoencoder's parameters.

Abstract: Methods and systems for gesture-based control of a device are described. A virtual gesture-space is determined in a received input frame. The virtual gesture-space is associated with a primary user from a ranked user list of users. The received input frame is processed in only the virtual gesture-space, to detect and track a hand. Using a hand bounding box generated by detecting and tracking the hand, gesture classification is performed to determine a gesture input associated with the hand. A command input associated with the determined gesture input is processed. The device may be a smart television, a smart phone, a tablet, etc.

Abstract: Systems and methods for multi-frame video frame interpolation. Higher-order motion modeling, such as cubic motion modeling, achieves predictions of intermediate optical flow between multiple interpolated frames, assisted by relaxation of the constraints imposed by the loss function used in initial optical flow estimation. A temporal pyramidal optical flow refinement module performs coarse-to-fine refinement of the optical flow maps used to generate the intermediate frames, focusing a proportionally greater amount of refinement attention to the optical flow maps for the high-error middle frames. A temporal pyramidal pixel refinement module performs coarse-to-fine refinement of the generated intermediate frames, focusing a proportionally greater amount of refinement attention to the high-error middle frames.

Abstract: Methods and systems for updating the weights of a set of convolution kernels of a convolutional layer of a neural network are described. A set of convolution kernels having attention-infused weights is generated by using an attention mechanism based on characteristics of the weights. For example, a set of location-based attention multipliers is applied to weights in the set of convolution kernels, a magnitude-based attention function is applied to the weights in the set of convolution kernels, or both. An output activation map is generated using the set of convolution kernels with attention-infused weights. A loss for the neural network is computed, and the gradient is back propagated to update the attention-infused weights of the convolution kernels.

Abstract: Methods and systems for determining a surface color of a target surface under an environment with an environmental light source. A plurality of images of the target surface are captured as the target surface is illuminated with a variable intensity, constant color light source and a constant intensity, constant color environmental light source, wherein the intensity of the light source on the target surface is varied by a known amount between the capturing of the images. A color feature tensor, independent of the environmental light source, is extracted from the image data, and used to infer a surface color of the target surface.

Abstract: Methods, processing units and media for multi-bandwidth separated feature extraction convolution in a neural network are described. A convolution block splits input channels of an activation map into multiple branches, each branch undergoing convolution at a different bandwidth by using down-sampling of the inputs. The outputs are concatenated by up-sampling the outputs of the low-bandwidth branches using pixel shuffling. The concatenation operation may be a shuffled concatenation operation that preserves separated multi-bandwidth feature information for use by subsequent layers of the neural network. Embodiments are described which apply frequency-based and magnitude-based attention to the weights of the convolution kernels based on the frequency band locations of the weights.

Abstract: Methods and apparatus for gesture-based control of a device in a multi-user environment are described. The methods prioritize users or gestures based on a predetermined priority ruleset. A first-user-in-time ruleset prioritizes gestures based on when in time they were begun by a user in the camera FOV. An action-hierarchy ruleset prioritizes gestures based on the actions they correspond to, and the relative positions of those actions within an action hierarchy. A designated-master-user ruleset prioritizes gestures performed by an explicitly designated master user. Methods for designating a new master user and for providing gesture-control-related user feedback in a multi-user environment are also described.

Abstract: Methods and systems for processing an image are described. A saliency map is generated from the image. The saliency map indicates one or more salient portions of the image that have saliency values satisfying a saliency criterion. A scene graph is generated for at least the one or more salient portions. The scene graph represents a plurality of objects detected in the image. The scene graph further represents one or more relationships between each respective object pairs. One or more dataset entries associated with the image are generated. Each of the one or more relationships for each of the one or more object pairs is indicated by a respective dataset entry. The one or more dataset entries are stored in a first dataset.

Abstract: Methods, devices, and processor-readable media for adjusting the control-display gain of a gesture-controlled device are described. Adjusting the control-display gain may facilitate user interaction with content or UI elements rendered on a display screen of the gesture-controlled device. The control-display gain may be adjusted based on a property of how a mid-air dragging gesture is being performed by a user's hand. The property may be the location of the gesture, the orientation of the hand performing the gesture, or the velocity of the gesture. A hand that becomes stationary for a threshold time period while performing the dragging gesture may adjust the control-display gain to a different level. Control-display gain may be set to a different value based on the current velocity of the hand performing the gesture. The control-display gain levels may be selected from a continuous range of values or a set of discrete values. Devices for performing the methods are described.

Abstract: Methods and apparatus for gesture-based control of a device in a multi-user environment are described. The methods prioritize users or gestures based on a predetermined priority ruleset. A first-user-in-time ruleset prioritizes gestures based on when in time they were begun by a user in the camera FOV. An action-hierarchy ruleset prioritizes gestures based on the actions they correspond to, and the relative positions of those actions within an action hierarchy. A designated-master-user ruleset prioritizes gestures performed by an explicitly designated master user. Methods for designating a new master user and for providing gesture-control-related user feedback in a multi-user environment are also described.

Abstract: Methods and apparatus for gesture-based control of a device in a multi-user environment are described. The methods prioritize users or gestures based on a predetermined priority ruleset. A first-user-in-time ruleset prioritizes gestures based on when in time they were begun by a user in the camera FOV. An action-hierarchy ruleset prioritizes gestures based on the actions they correspond to, and the relative positions of those actions within an action hierarchy. A designated-master-user ruleset prioritizes gestures performed by an explicitly designated master user. Methods for designating a new master user and for providing gesture-control-related user feedback in a multi-user environment are also described.

Abstract: Systems and methods for multi-frame video frame interpolation. Higher-order motion modeling, such as cubic motion modeling, achieves predictions of intermediate optical flow between multiple interpolated frames, assisted by relaxation of the constraints imposed by the loss function used in initial optical flow estimation. A temporal pyramidal optical flow refinement module performs coarse-to-fine refinement of the optical flow maps used to generate the intermediate frames, focusing a proportionally greater amount of refinement attention to the optical flow maps for the high-error middle frames. A temporal pyramidal pixel refinement module performs coarse-to-fine refinement of the generated intermediate frames, focusing a proportionally greater amount of refinement attention to the high-error middle frames.

Abstract: Methods, devices and processor-readable media for an integrated teacher-student machine learning system. One or more teacher-student modules are trained as part of the teacher neural network training. Each student sub-network uses a portion of the teacher neural network to generate an intermediate feature map, then provides the intermediate feature map to a student sub-network to generate inferences. The student sub-network may use a feature enhancement block to map the intermediate feature map to a subsequent feature map. A compression block may be used to compress intermediate feature map data for transmission in some embodiments.

Abstract: Methods and apparatus for gesture-based control of a device in a multi-user environment are described. The methods prioritize users or gestures based on a predetermined priority ruleset. A first-user-in-time ruleset prioritizes gestures based on when in time they were begun by a user in the camera FOV. An action-hierarchy ruleset prioritizes gestures based on the actions they correspond to, and the relative positions of those actions within an action hierarchy. A designated-master-user ruleset prioritizes gestures performed by an explicitly designated master user. Methods for designating a new master user and for providing gesture-control-related user feedback in a multi-user environment are also described.