Abstract: Methods, systems, and apparatus for efficient image analysis. In some aspects, a system includes a camera configured to capture images, one or more environment sensors configured to detect movement of the camera, a data processing apparatus, and a memory storage apparatus in data communication with the data processing apparatus. The data processing apparatus can access, for each of a multitude of images captured by a mobile device camera, data indicative of movement of the camera at a time at which the camera captured the image. The data processing apparatus can also select, from the images, a particular image for analysis based on the data indicative of the movement of the camera for each image, analyze the particular image to recognize one or more objects depicted in the particular image, and present content related to the one or more recognized objects.

Abstract: Systems and methods of the present disclosure are directed to a method for generating a machine-learned multitask model configured to perform tasks. The method can include obtaining a machine-learned multitask search model comprising candidate nodes. The method can include obtaining tasks and machine-learned task controller models associated with the tasks. As an example, for a task, the method can include using the task controller model to route a subset of the candidate nodes in a machine-learned task submodel for the corresponding task. The method can include inputting task input data to the task submodel to obtain a task output. The method can include generating, using the task output, a feedback value based on an objective function. The method can include adjusting parameters of the task controller model based on the feedback value.

Abstract: Methods, systems, and apparatus for efficient image analysis. In some aspects, a system includes a camera configured to capture images, one or more environment sensors configured to detect movement of the camera, a data processing apparatus, and a memory storage apparatus in data communication with the data processing apparatus. The data processing apparatus can access, for each of a multitude of images captured by a mobile device camera, data indicative of movement of the camera at a time at which the camera captured the image. The data processing apparatus can also select, from the images, a particular image for analysis based on the data indicative of the movement of the camera for each image, analyze the particular image to recognize one or more objects depicted in the particular image, and present content related to the one or more recognized objects.

Abstract: The present disclosure is directed to object and/or character recognition for use in applications such as computer vision. Advantages of the present disclosure include lightweight functionality that can be used on devices such as smart phones. Aspects of the present disclosure include a sequential architecture where a lightweight machine-learned model can receive an image, detect whether an object is present in one or more regions of the image, and generate an output based on the detection. This output can be applied as a filter to remove image data that can be neglected for more memory intensive machine-learned models applied downstream.

Abstract: Methods, systems, and apparatus for efficient image analysis. In some aspects, a system includes a camera configured to capture images, one or more environment sensors configured to detect movement of the camera, a data processing apparatus, and a memory storage apparatus in data communication with the data processing apparatus. The data processing apparatus can access, for each of a multitude of images captured by a mobile device camera, data indicative of movement of the camera at a time at which the camera captured the image. The data processing apparatus can also select, from the images, a particular image for analysis based on the data indicative of the movement of the camera for each image, analyze the particular image to recognize one or more objects depicted in the particular image, and present content related to the one or more recognized objects.

Abstract: Methods, systems, and apparatus for efficient image analysis. In some aspects, a system includes a camera configured to capture images, one or more environment sensors configured to detect movement of the camera, a data processing apparatus, and a memory storage apparatus in data communication with the data processing apparatus. The data processing apparatus can access, for each of a multitude of images captured by a mobile device camera, data indicative of movement of the camera at a time at which the camera captured the image. The data processing apparatus can also select, from the images, a particular image for analysis based on the data indicative of the movement of the camera for each image, analyze the particular image to recognize one or more objects depicted in the particular image, and present content related to the one or more recognized objects.

Abstract: Aspects of the disclosure provide for an antenna system. The antenna system includes a main reflector, a phased array feed, and a mechanical steering system. The phased array feed is configured to receive and/or transmit signals reflected off the main reflector and to electronically steer a transmitted signal on a first axis. The main reflector and the phased array feed are both attached to the mechanical steering system, and the mechanical steering system is configured to move the transmitted signal along a second axis.

Abstract: Methods, systems, and apparatus for efficient image analysis. In some aspects, a system includes a camera configured to capture images, one or more environment sensors configured to detect movement of the camera, a data processing apparatus, and a memory storage apparatus in data communication with the data processing apparatus. The data processing apparatus can access, for each of a multitude of images captured by a mobile device camera, data indicative of movement of the camera at a time at which the camera captured the image. The data processing apparatus can also select, from the images, a particular image for analysis based on the data indicative of the movement of the camera for each image, analyze the particular image to recognize one or more objects depicted in the particular image, and present content related to the one or more recognized objects.

Abstract: In a semi-open loop rate adaptation scheme for a multiple-input multiple-output (MIMO) system, a transmitter can advantageously use one or more quality metrics of an uplink as well as knowledge of device characteristics of both ends to perform fast and accurate rate adaptation.

Abstract: Embodiments of the invention are generally directed to multiple protocol tunneling using time division operations. An embodiment of an apparatus includes an interface for communication with a second apparatus, the interface including a shared communication link; and a multiplexer to multiplex data of each of multiple protocols into time slots for transmission, the protocols including a first protocol. The time slots are distributed among the protocols, where the distribution of the time slots among the protocols includes assigning one or more time slots to the first protocol to enable the data of the first protocol to meet one or more performance requirements for the first protocol.

Abstract: Embodiments of apparatuses and methods to decrease a size of a memory in a low-latency video communication system are described. A control unit is configured to monitor a condition associated with at the communication link. The control unit is configured to receive the video content over a link based on monitoring. A memory comprising a full-frame buffer is coupled to the control unit. The full-frame buffer is configured as a history buffer to store a full frame of the video in a coding format that matches the coding format of the video content received over the link. A display unit is coupled to the history buffer. A portion of the full-frame buffer is configured as a network streaming buffer.

Abstract: A system and method are disclosed for transmitting data over a wireless channel. In some embodiments, transmitting data includes receiving convolutionally encoded data and enhancing the transmission of the data by further repetition encoding the data.

Abstract: In a semi-open loop rate adaptation scheme for a multiple-input multiple-output (MIMO) system, a transmitter can advantageously use one or more quality metrics of an uplink as well as knowledge of device characteristics of both ends to perform fast and accurate rate adaptation.

Abstract: A media access controller (MAC) generates a composite packet having an optimized format for carrying audio, video, and data traffic. A header portion of the composite packet is encoded separately from a data portion of the composite packet. A physical device interface (PHY) is coupled to the MAC. The PHY encodes and decodes between a digital signal and a modulated analog signal. The PHY comprises a high rate physical layer circuit (HRP) and a low rate physical layer circuit (LRP). A radio frequency (RF) transmitter is coupled to the PHY to transmit data.

Abstract: A system and method are disclosed for transmitting data over a wireless channel. In some embodiments, transmitting data includes receiving convolutionally encoded data and enhancing the transmission of the data by further repetition encoding the data.

Abstract: A multiple-input multiple-output (MIMO) system can transmit on multiple antennas simultaneously and receive on multiple antennas simultaneously. Unfortunately, because a legacy 802.11a/g device is not able to decode multiple data streams, such a legacy device may “stomp” on a MIMO packet by transmitting before the transmission of the MIMO packet is complete. Therefore, MIMO systems and methods are provided herein to allow legacy devices to decode the length of a MIMO packet and to restrain from transmitting during that period. These MIMO systems and methods are optimized for efficient transmission of MIMO packets.

Abstract: A system and method are disclosed for transmitting data over a wireless channel. In some embodiments, transmitting data includes receiving convolutionally encoded data and enhancing the transmission of the data by further repetition encoding the data.

Abstract: A method and system of optimizing transmit beam forming in a multiple radio wireless system is provided. A stimulus signal can be provided to an analog receive input of a device under test (DUT), wherein the DUT includes multiple radios. A receive phase and amplitude can be measured at baseband using the stimulus signal for each radio. At this point, a receive weight and its conjugate can be determined using the receive phases and amplitudes. A calibration vector and its conjugate can also be determined, wherein a product of the receive weight conjugate and the calibration vector conjugate generate a transmit weight. This transmit weight can be applied to transmit signals during the transmit beam forming using the multiple radios.

Abstract: Embodiments of apparatuses and methods to decrease a size of a memory in a low-latency video communication system are described. A control unit is configured to monitor a condition associated with at the communication link. The control unit is configured to receive the video content over a link based on monitoring. A memory comprising a full-frame buffer is coupled to the control unit. The full-frame buffer is configured as a history buffer to store a full frame of the video in a coding format that matches the coding format of the video content received over the link. A display unit is coupled to the history buffer. A portion of the full-frame buffer is configured as a network streaming buffer.

Abstract: A system and method are described for binding together a plurality of wireless data communications channels, whereby an aggregate throughput improvement is realized. A master channel amongst the channels to be bound is compatible with existing standards-based wireless data communications equipment. The master channel serves to perform MAC association and flow control. Aggregate throughput is improved by sending and receiving either multiple sets of separately encoded packets, commonly encoded packets or redundantly encoded packets.