Patent application number | Description | Published |
20140046935 | Identifying Textual Terms in Response to a Visual Query - A method, system, and computer readable storage medium is provided for identifying textual terms in response to a visual query is provided. A server system receives a visual query from a client system. The visual query is responded to as follows. A set of image feature values for the visual query is generated. The set of image feature values is mapped to a plurality of textual terms, including a weight for each of the textual terms in the plurality of textual terms. The textual terms are ranked in accordance with the weights of the textual terms. Then, in accordance with the ranking the textual terms, one or more of the ranked textual terms are sent to the client system. | 02-13-2014 |
20150161170 | IMAGE CLASSIFICATION - An image classification system trains an image classification model to classify images relative to text appearing with the images. Training images are iteratively selected and classified by the image classification model according to feature vectors of the training images. An independent model is trained for unique n-grams of text. The image classification system obtains text appearing with an image and parses the text into candidate labels for the image. The image classification system determines whether an image classification model has been trained for the candidate labels. When an image classification model corresponding to a candidate label has been trained, the image classification subsystem classifies the image relative to the candidate label. The image is labeled based on candidate labels for which the image is classified as a positive image. | 06-11-2015 |
20150161172 | Image Relevance Model - Methods, systems, and apparatus, including computer program products, for identifying images relevant to a query are disclosed. An image search subsystem selects images to reference in image search results that are responsive to a query based on an image relevance model that is trained for the query. An independent image relevance model is trained for each unique query that is identified by the image search subsystem. The image relevance models can be applied to images to order image search results obtained for the query. Each relevance model is trained based on content feature values of images that are identified as being relevant to the query (e.g., frequently selected from the image search results) and images that are identified as being relevant to another unique query. The trained model is applied to the content feature values of all known images to generate an image relevance score that can be used to order search results for the query. | 06-11-2015 |
20150161482 | REFINING IMAGE RELEVANCE MODELS - Methods, systems and apparatus for refining image relevance models. In general, one aspect of the subject matter described in this specification can be implemented in methods that include re-training an image relevance model by generating a first re-trained model based on content feature values of first images of a first portion of training images in a set of training images, receiving, from the first re-trained model, image relevance scores for second images of a second portion of the set of training images, removing, from the set of training images, some of the second images identified as outlier images for which the image relevance score received from the first re-trained model is below a threshold score, and generating a second re-trained model based on content feature values of the first images of the first portion and the second images of the second portion that remain following removal of the outlier images. | 06-11-2015 |
20150178383 | Classifying Data Objects - Methods, systems, and apparatus, including computer programs encoded on computer storage media, for classifying data objects. One of the methods includes obtaining data that associates each term in a vocabulary of terms with a respective high-dimensional representation of the term; obtaining classification data for a data object, wherein the classification data includes a respective score for each of a plurality of categories, and wherein each of the categories is associated with a respective category label; computing an aggregate high-dimensional representation for the data object from high-dimensional representations for the category labels associated with the categories and the respective scores; identifying a first term in the vocabulary of terms having a high-dimensional representation that is closest to the aggregate high-dimensional representation; and selecting the first term as a category label for the data object. | 06-25-2015 |
20150178596 | Label Consistency for Image Analysis - Systems and techniques are disclosed for labeling objects within an image. The objects may be labeled by selecting an option from a plurality of options such that each option is a potential label for the object. An option may have an option score associated with. Additionally, a relation score may be calculated for a first option and a second option corresponding to a second object in an image. The relation score may be based on a frequency, probability, or observance corresponding to the co-occurrence of text associated with the first option and the second option in a text corpus such as the World Wide Web. An option may be selected as a label for an object based on a global score calculated based at least on an option score and relation score associated with the option. | 06-25-2015 |
Patent application number | Description | Published |
20110047163 | Relevance-Based Image Selection - A system, computer readable storage medium, and computer-implemented method presents video search results responsive to a user keyword query. The video hosting system uses a machine learning process to learn a feature-keyword model associating features of media content from a labeled training dataset with keywords descriptive of their content. The system uses the learned model to provide video search results relevant to a keyword query based on features found in the videos. Furthermore, the system determines and presents one or more thumbnail images representative of the video using the learned model. | 02-24-2011 |
20110191374 | Joint Embedding for Item Association - Methods and systems to associate semantically-related items of a plurality of item types using a joint embedding space are disclosed. The disclosed methods and systems are scalable to large, web-scale training data sets. According to an embodiment, a method for associating semantically-related items of a plurality of item types includes embedding training items of a plurality of item types in a joint embedding space configured in a memory coupled to at least one processor, learning one or more mappings into the joint embedding space for each of the item types to create a trained joint embedding space and one or more learned mappings, and associating one or more embedded training items with a first item based upon a distance in the trained joint embedding space from the first item to each said associated embedded training items. Exemplary item types that may be embedded in the joint embedding space include images, annotations, audio and video. | 08-04-2011 |
20120082371 | LABEL EMBEDDING TREES FOR MULTI-CLASS TASKS - Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for label embedding trees for large multi-class tasks. In one aspect, a method includes mapping each image in a plurality of images and each label in a plurality of labels into a multi-dimensional label embedding space. A tree of label predictors is trained with the plurality of mapped images such that an error function is minimized in which the error function counts an error for each mapped image if any of the label predictors at any depth of the tree incorrectly predicts that the mapped image belongs to the label predictor's respective label set. | 04-05-2012 |
20120290621 | GENERATING A PLAYLIST - Generating a playlist may include designating a seed track in an audio library; identifying audio tracks in the audio library having constructs that are within a range of a corresponding construct of the seed track, where the constructs for the audio tracks are derived from frequency representations of the audio tracks, and the corresponding construct for the seed track is derived from a frequency representation of the seed track; and generating the playlist using at least some of the audio tracks that were identified. | 11-15-2012 |
20150161485 | Learning Semantic Image Similarity - Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for identifying similar images. In some implementations, a method is provided that includes receiving a collection of images and data associated with each image in the collection of images; generating a sparse feature representation for each image in the collection of images; and training an image similarity function using image triplets sampled from the collection of images and corresponding sparse feature representations. | 06-11-2015 |
Patent application number | Description | Published |
20140016490 | MISSED RANK REPORT IN DUAL NETWORK RADIO RESOURCE MANAGEMENT - A method for handling a missed rank report during a tune-away period is provided. The method can include a wireless communication device tuning away from a first network to a second network for a tune-away period; returning to the first network from the tune-away period; determining that a scheduled rank report was missed during the tune-away period; generating a Channel State Indicator (CSI) report based on a previously defined Rank Indicator (RI) value known to the first network in response to missing the scheduled rank report; and sending the CSI report to the first network. | 01-16-2014 |
20140071888 | Data Buffering based on Access Stratum Conditions in a Call Having both Circuit-Switched and Packet-Switched Components - Data buffering based on access-stratum conditions by a wireless user equipment (UE) device. A call may be established with a network via a wireless link. The call may include a circuit-switched component and a packet-switched component. It may be determined that the packet-switched component has been released by the network. The UE may have packet data for transmission to the network when the packet-switched component is released. The packet data may be buffered based on determining that the packet-switched component has been released by the network. | 03-13-2014 |
20140112172 | Load Estimation in 3GPP Networks - Loading estimation of 3GPP networks. One or more metrics relating to a cell of a 3GPP network may be measured. Loading of the cell may be estimated based on the one or more metrics. The metrics may include metrics measured, estimated, or derived at multiple layers, possibly including one or more of physical layer, radio link control layer, radio resource control layer, or application layer metrics. | 04-24-2014 |
20140113630 | Wireless Device Based Inter Radio Access Technology Handover Initiation - User Equipment (UE) based forced inter radio access technology (iRAT) handover. A connection to a network may be established via a first cell operating according to a first radio access technology (RAT). It may be determined to initiate a handover of the UE from the first cell to a second cell operating according to a second RAT. An indication may be transmitted to the network to initiate a handover of the UE from the first cell to the second cell. An indication may be received from the network to perform handover of the UE from the first cell to the second cell in response to the indication to initiate the handover. Handover of the UE from the first cell to the second cell may be performed in response to the indication to perform the handover. Handover may include releasing the connection to the network via the first cell and establishing a connection to the network via the second cell. | 04-24-2014 |
20140126544 | NETWORK CELL TRANSITIONS FOR VOLTE DEVICES AT CALL INITIATION - Apparatus and methods of performing a network cell transition for a voice over long term evolution (VoLTE) capable mobile device. Initially, the mobile device can detect a call initiation event. Thereafter, the mobile device determines whether a radio signal power associated with its serving LTE network base station is less than a radio access technology (RAT) specific threshold (i.e., an LTE-specific threshold). In a scenario where the radio signal power is less than the RAT-specific threshold the mobile device can effectuate a transition to an available neighbor network base station employing a legacy RAT, prior to establishing a voice call. When the mobile device is in a radio resource control (RRC) connected mode the network cell transition may be associated with an induced network handover procedure. Alternatively, when the mobile device is in an RRC idle mode the network cell transition may be associated with network reselection procedure. | 05-08-2014 |
20140146763 | QoS Based Buffering while TTI Bundling is Enabled - QoS based uplink data buffering while TTI bundling is enabled by a wireless user equipment (UE) device. The UE may establish a packet-switched connection with a network via a wireless link. The UE may receive, at a media access control (MAC) layer, an indication to enable TTI bundling. The UE may selectively buffer uplink data at an application layer based on the indication to enable TTI bundling. The uplink data may be buffered selectively based on Quality of Service (QoS) considerations. Uplink transmissions may subsequently be performed using TTI bundling. | 05-29-2014 |
20140176659 | OFFLOADING A VIDEO PORTION OF A VIDEO CALL - A method for offloading a video portion of a video call form a cellular network to a WLAN is provided. The method can include a wireless communication device participating in a video call with a remote device over a connection between the wireless communication device and a cellular network. The video call can include a first video stream carried over a first bearer on the cellular network and an audio stream carried over a second bearer on the cellular network. The method can further include the wireless communication device establishing a connection between the wireless communication device and a WLAN; performing a call setup procedure with the remote device to establish a second video stream over the WLAN; terminating the first video stream; and using the second video stream and the audio stream to continue the video call. | 06-26-2014 |
20140176660 | TRANSITIONING A VIDEO CALL BETWEEN NETWORKS - A method for transitioning a video call is provided. The method can include a wireless communication device participating in a video call with a remote communication device via a first video call session established over a connection between the wireless communication device and a first cellular network. The video call can include a packet switched video stream carried over a first bearer and an audio stream carried over a second bearer. The method can further include the wireless communication device determining a degradation in a connection quality for the first cellular network; transitioning to a legacy cellular network having a circuit switched domain in response to the degradation in connection quality for the first cellular network; establishing a second video call session on the legacy cellular network; and using the second video call session to continue the video call on the legacy cellular network. | 06-26-2014 |
20140198699 | Reducing Power Consumption in Voice over LTE Terminals using Semi Persistent Scheduling in Connected Discontinuous Reception Mode - A connection with a network that includes a base station (BS) may be established by a user device (UE) via a wireless connection, for conducting communications using semi persistent scheduling (SPS) in a connected discontinuous reception (C-DRX) mode. The SPS transmit periodicity may be adjusted with respect to the SPS activation command and the SPS interval UL (for uplink). Data may then be transmitted during the C-DRX On-Duration periods according to the determined SPS transmit periodicity. In some embodiments, the SPS transmit periodicity is adjusted such that following a first C-DRX On-Duration period when an SPS activation command is received, SPS data transmission occurs a specified number of subframes earlier during each subsequent C-DRX On-Duration period than in the first C-DRX On-Duration period. The SPS data transmission in each subsequent C-DRX On-Duration period may take place as soon as the UE device wakes up during the On-Duration period. | 07-17-2014 |
20140219151 | REDUCING POWER CONSUMPTION IN CONNECTED MODE DISCONTINUOUS RECEPTION - A method for reducing power consumption in connected mode discontinuous reception is disclosed. The method can include a wireless communication device sending a transmission for a pending HARQ retransmission process and receiving an ACK for the transmission. The method can further include the wireless communication device determining a subset of remaining uplink transmission opportunities in the pending HARQ retransmission process to monitor for an uplink grant in response to receiving the ACK and monitoring the subset of remaining uplink transmission opportunities for an uplink grant. The method can additionally include the wireless communication device entering a sleep state for any uplink transmission opportunities remaining in the pending HARQ retransmission process after monitoring the subset of remaining uplink transmission opportunities in an instance in which an uplink grant for the pending HARQ retransmission process is not received for any of the subset of remaining uplink transmission opportunities. | 08-07-2014 |
20140219246 | SRVCC HANDOVER INDICATION FOR REMOTE PARTY TO VOICE CALL - A method for providing indication of an SRVCC handover is disclosed. The method can include a first wireless communication device participating in a voice call with a second wireless communication device via a connection between the first wireless communication device and a first network. The method can further include the first wireless communication device determining a condition indicative of an impending SRVCC handover of the first wireless communication device from the first network to a legacy network. In response to the condition, the method can additionally include the first wireless communication device formatting a message including an indication that the first wireless communication device is going to perform the SRVCC handover and sending the message to the second wireless communication device prior to performance of the SRVCC handover. | 08-07-2014 |
20140269624 | VOICE CALL RESUMPTION ON A LEGACY NETWORK - A method for resuming a voice call on a legacy network is provided. The method can include a wireless communication device establishing a voice call with a remote device over a first cellular network; determining that a connection quality of the first cellular network has dropped below a threshold connection quality; sending a notification that the wireless communication device is releasing the voice call to transition to a legacy cellular network and will resume the voice call on the legacy cellular network; releasing the voice call on the first cellular network; transitioning to the legacy cellular network; establishing a second voice call with the remote device on the legacy cellular network; and using the second voice call to resume the voice call with the remote device. | 09-18-2014 |
20140293961 | FACILITATING RETURN TO A FIRST WIRELESS NETWORK FROM A SECOND WIRELESS NETWORK AFTER PERFORMANCE OF A CIRCUIT SWITCHED FALLBACK PROCEDURE - Method and apparatus for facilitating return to a first wireless network from a second wireless network by a wireless communication device is provided. A method includes the wireless communication device engaging in data transfer for an application session on the first wireless network; participating in a CSFB procedure to transition from the first wireless network to the second wireless network for servicing a voice connection; buffering data received for the application session before and/or during the CSFB procedure; releasing the voice connection; discontinuing requests for downlink data for the application session in response to termination of the voice connection to provide a gap in data transfer; reselecting to the first wireless network during the gap in data transfer; using the buffered data to continue the application session during reselection; and resuming data transfer for the application session on the first wireless network after completing reselection. | 10-02-2014 |
20140376429 | Power Consumption Reduction for Cellular Broadcast - Reducing power consumption for cellular broadcast communication by a wireless device. A cellular communication channel for cellular broadcast messages may be monitored in a discontinuous manner according to specified cellular broadcast scheduling periods. Between each pair of cellular broadcast scheduling periods, the wireless device may operate in a reduced power state. During each cellular broadcast scheduling period, at least a cellular broadcast scheduling message may be received. If all scheduled cellular broadcast messages are received prior to an end of a cellular broadcast scheduling period, the wireless device may operate in the reduced power state for the remainder of that cellular broadcast scheduling period after all scheduled cellular broadcast messages are received. | 12-25-2014 |
20150092575 | DEVICE-INITIATED CODEC RATE CHANGE DURING A VOICE CALL - A method for initiating a codec rate change during a VoIP call by a wireless communication device is disclosed. The method can include the wireless communication device establishing a first codec rate for use in the VoIP call during a call establishment phase; using the first codec rate to encode voice data for transmission during a first portion of the VoIP call; determining a channel quality while using the first codec rate; determining that the channel quality satisfies a threshold for requesting a codec rate change; requesting a codec rate change from the first codec rate to a second codec rate in response to the channel quality satisfying the threshold; and using the second codec rate to encode voice data for transmission during a second portion of the VoIP call. | 04-02-2015 |
Patent application number | Description | Published |
20140204845 | FAST TRANSITION FROM PCH TO DCH FOR UMTS - Systems and methods for logically transitioning a client device or user equipment (UE) directly to a DCH state for the data path when an application layer application initiates an operation that may result in the transmission of large amounts of data, are presented. Upon detection of a condition that would warrant allocation of a dedicated channel, the UE may set the Traffic Volume Indicator in a subsequent cell update message to the network node. The node may then assign a dedicated channel and the UE may be transitioned to the dedicated channel state without a portion of the application data being first transmitted on a forward access channel or random access channel in a CELL_FACH state. | 07-24-2014 |
20150043481 | METHOD AND APPARATUS FOR FLUSHING UPLINK HARQ BUFFER IN C-DRX MODE - A method for determining whether an acknowledgement received by a user equipment from an external device is a forced acknowledgement. The method including transmitting a set of data stored in an uplink buffer to an external device, receiving an acknowledgement from the external device, determining if the acknowledgement received from the external device was a forced acknowledgement and flushing out an uplink buffer if determined that the acknowledgement was not a forced acknowledgement. The determining the acknowledgement is a forced acknowledgment being based on whether an uplink retransmission collides with one or more scheduled transmission times, a Physical Hybrid-ARQ Indicator Channel (PHICH) falls between gap measurements and an uplink retransmission collides with one of the gap measurements or a TTI bundling retransmission collides with a gap measurement. If the acknowledgement is not a forced acknowledgment, a set of data stored in the uplink buffer is retransmitted to the external device. | 02-12-2015 |
20150092643 | System and Method for Selective Prevention of Transmitting a Scheduling Request - A station that performs a method to coordinate transmissions of scheduling requests (SR) with OnDurations of a connected discontinuous reception (C-DRX) cycle. The station receives a packet at a packet arrival time relative to a schedule indicating the SR opportunities and onDurations, the SR opportunities occurring at a first interval, the onDurations occurring at a second interval, the first interval being less than the second interval. The station determines an SR opportunity subsequent to the packet arrival time that precedes an entirety of a subsequent onDuration, such that a transmission duration to transmit the SR and the packet maximally overlaps with the selected onDuration. A sleep mode of the processor of the station is used until a time associated with the selected SR opportunity. An active mode of the processor is used to transmit the SR and receive control channel information during the onDuration. | 04-02-2015 |
20150092644 | System and Method for Audio Frame Generation Alignment with LTE Transmission Opportunities - A station that generates data packets to be transmitted by the station such that the data packets spend a minimum amount of time in a buffer prior to transmission. The method includes receiving a specification for a connected discontinuous reception (C-DRX) cycle, the specification indicating when a plurality of onDurations of the C-DRX cycle occurs, the onDurations having a predetermined interval therebetween, receiving data at a known time relative to the C-DRX cycle, determining a modification to a conversion process that converts the data to data packets such that the data packets are stored in a buffer at a subframe immediately preceding one of the onDurations subsequent to the known time, performing the conversion process based upon the modification and storing the data packets at the subframe immediately preceding the one of the onDurations. In one embodiment, the data is raw audio data and the data packets are audio packets. | 04-02-2015 |
20150181472 | ESR Extension for LTE TDD to FDD Redirection for VoLTE - This application presents techniques for an LTE user equipment (UE) to use an extended service request (ESR) extension for LTE TDD to FDD redirection for mobile originated and mobile terminated VoLTE calls. These techniques include the UE informing the network that it supports the particular features of the ESR extensions presented. Once the UE attaches to the network, a radio resource control (RRC) message can be sent to indicate that the UE supports the new ESR extension, after which the UE can use the new ESR extension to facilitate an LTE TDD to FDD redirection for the VoLTE call. | 06-25-2015 |
20150181483 | VoLTE Call Establishment in TD and FDD LTE Networks - This application discloses various techniques for call establishment using voice-over LTE (VoLTE) in networks supporting time division (TD) and frequency division duplexed (FDD) LTE communication systems. Such techniques can include systems and methods for mobile-originated calls for a UE in a TD-LTE cell, systems and methods for mobile-terminated calls for a connected UE in a TD-LTE cell and systems and methods for mobile-terminated calls for an idle UE in a TD-LTE cell. These methods and systems can leverage component carriers from a carrier aggregating capable UE to facilitate more efficient and/or effective UE call establishment. | 06-25-2015 |