Abstract: Embodiments of the present disclosure provide a method and system for co-operative and cascaded inference on the edge device using an integrated Deep Learning (DL) model for object detection and localization, which comprises a strong classifier trained on largely available datasets and a weak localizer trained on scarcely available datasets, and work in coordination to first detect object (fire) in every input frame using the classifier, and then trigger a localizer only for the frames that are classified as fire frames. The classifier and the localizer of the integrated DL model are jointly trained using Multitask Learning approach. Works in literature hardly address the technical challenge of embedding such integrated DL model to be deployed on edge devices. The method provides an optimal hardware software partitioning approach for components or segments of the integrated DL model which achieves a tradeoff between latency and accuracy in object classification and localization.

Abstract: Systems and methods for generating a road map are provided. The methods include obtaining a navigation route for a navigation terminal and a current location of the navigation terminal. The methods further include obtaining road network data associated with the current location of the navigation terminal based on the navigation route and the current location of the navigation terminal. The road network data includes a plurality of links and a plurality of nodes. Each node is connected to one or more links including at least one type of links. The methods further include determining a level for each node. The methods further include determining one or more target links for the at least one type of links connected to the node based on the level of the node. The methods further include generating a road map based on the target links.

Abstract: Disclosed is an image processing apparatus and a method of operating the same. The image processing apparatus includes: a memory storing information on at least one random patch; and at least one processor configured to: obtain correlations between a pixel block included in an input image and each of a plurality of random patches obtained from the information on the at least one random patch, obtain weights respectively for the plurality of random patches on a basis of the obtained correlations and apply the weights respectively to the plurality of random patches, and obtain an output image by applying, to the pixel block, the plurality of random patches to which the weights are respectively applied.

Abstract: Cross-service usage is dynamically enforced for different network-based services. A cross-service control system may accept requests to specify a usage quota for a client for an interface of a network-based service. A rule to enforce the usage quota may be distributed to request handling nodes for the service. Usage measurements may be collected from the request handling nodes for requests received from the client via the interface. An update to the usage quota may be determined from an evaluation of the usage measurements. The rule at the request handling nodes may be updated to enforce the updated usage quota.

Abstract: A network device executes an account management function that pre-authorizes network service unit consumption by multiple subscribers that are members of a same, shared network service account. The account management function receives, from multiple charging systems distributed within the network, network unit consumption charges associated with each of the multiple subscribers, where the network device is centrally located within a network relative to multiple locations associated with the plurality of charging systems. The account management function updates a consumption counter for the same, shared network service account based on the network unit consumption charges associated with the multiple subscribers received from the multiple charging systems distributed within the network.

Abstract: An iris scanning device that includes a single camera sensor utilized for generating iris scan data is described herein. The iris scanning device includes a selector, a lens, and the camera sensor. The selector is configured to receive a first optical signal representative of a first eye and a second optical signal representative of a second eye. The selector selectively allows a passed optical signal to be optically propagated to the lens and inhibits a blocked optical signal from being optically propagated to the lens during a time period. The passed optical signal is one of the first optical signal or the second optical signal during the time period. The blocked optical signal is a differing one of the first optical signal or the second optical signal during the time period. The lens causes the passed optical signal to be incident on the camera sensor during the time period.

Abstract: A device and a method and a non-transitory readable storage medium, for face recognition are provided, the method comprise: extracting a face sample image from a predetermined face sample library and performing feature point detection to obtain multiple face feature points; obtaining multiple mask images; selecting first to fourth face feature points from the multiple face feature points; defining a distance between the first and second face feature point as a mask image height, and defining a distance between the third and fourth face feature point as a mask image width; adjusting a size of each mask image according to the mask image height and the mask image width; fusing each adjusted mask image with the face sample image to obtain multiple face mask images to save into the predetermined face sample library; training a face recognition model based on the predetermined face sample library for face recognition.

Abstract: There are provided measures for coordinated session charging in roaming scenarios. Such measures (of a first session management entity of a first mobile network in a roaming charging related scenario between said first mobile network and a second mobile network) exemplarily comprise transmitting a session creation request message to a second session management entity of said second mobile network, said session creation request message including information indicative of a first configuration utilized for mobile network related charging, and receiving a session creation response message from said second session management entity of said second mobile network, said session creation response message including information indicative of a second configuration utilized for mobile network related charging.

Abstract: Provided is a human posture detection method including: acquiring a plurality of frames of image data; acquiring a plurality of human posture reference maps output by a human posture detection model responsive to inputting a current frame of image data to the human posture detection model with reference to human posture confidence maps of a previous frame of image data, wherein different human posture reference maps correspond different human-posture key points; identifying a human-posture key point in each of the human posture reference maps; and generating human posture confidence maps of the current frame of image data based on credibility of the human-posture key points, wherein the human posture confidence maps of the current frame of image data is configured to participate in generation of human posture confidence maps of a next frame of image data.

Abstract: A method may include obtaining a plurality of training samples with a plurality of classifications that include a first classification and a second classification, training an initial tree with an initial set of training samples selected from the plurality of training samples using an initial set of feature values extracted from the set of training samples, and, in response to determining that the initial tree incorrectly classified the initial set of training samples at an output node of the initial tree, training a subsequent tree using a subsequent set of feature values extracted from a subsequent set of training samples.

Abstract: A processor-implemented method of generating feature data includes: receiving an input image; generating, based on a pixel value of the input image, at least one low-bit image having a number of bits per pixel lower than a number of bits per pixel of the input image; and generating, using at least one neural network, feature data corresponding to the input image from the at least one low-bit image.

Abstract: Methods and apparatuses for data transmission. A method implemented at a first network function may comprise obtaining information related to network slice selection; and sending a reroute message to an access network node, wherein the reroute message including a container comprising at least one information element of the information related to network slicing selection.

Abstract: To quickly and accurately quantify four types of TSNAs contained in a tobacco raw material by using fluorescence fingerprint information, a tobacco raw material containing known quantities of four types of TSNAs is prepared, and fluorescence fingerprint information thereof is acquired. The fluorescence fingerprint information is pre-processed, as needed, and an estimation model (calibration curve) for estimating the total quantity of the four types of TSNAs from the fluorescence fingerprint information is created. After verifying the calibration curve, the total quantity of the four types of TSNAs contained in an unknown tobacco raw material is estimated by applying the calibration curve to the unknown tobacco raw material. On the basis of the estimated value of the total quantity of the four types of TSNAs and the known abundance ratios of the individual TSNAs, the contained quantities of the four types of TSNAs are estimated.

Abstract: Star tracker systems and methods are provided. The star tracker incorporates deep learning processes in combination with relatively low cost hardware components to provide moderate (e.g. ˜1 arc second attitude uncertainty) accuracy. The neural network implementing the deep learning processes can include a Hinton's capsule network or a coordinate convolution layer to maintain spatial relationships between features in images encompassing a plurality of features. The hardware components can be configured to collect a blurred or defocused image in which point sources of light appear as blurs, and in which the blurs create points of intersection. Alternatively or in addition, a blurred or defocused image can be created using processes implemented as part of application programming. The processing of collected images by a neural network to provide an attitude determination can include analyzing a plurality of blurs and blur intersections across an entire frame of image data.

Abstract: The present disclosure method and device for wireless communications, comprising: operating a first data block set; and transmitting first charging information; a size of data in the first data block set is used to generate the first charging information, the first charging information comprises a first identity (ID) set, and the first ID set comprises a first ID and a second ID; the first ID is a link layer ID; the operating action is receiving, a destination ID field of a MAC header of a MAC PDU used to carry the first data block set comprises at least partial bits in a first ID, when a destination link layer ID list maintained by a node indicated by the second ID comprises the first ID. The present disclosure determines a receiver or a generator of a first data block set, which improves accuracy of charging and enriches charging types of services.

Abstract: A method and system for determining a geographical location and orientation of a vehicle travelling through a road network is disclosed. The method comprises obtaining, from one or more cameras associated with the vehicle travelling through the road network, a sequence of images reflecting the environment of the road network on which the vehicle is travelling, wherein each of the images has an associated camera location at which the image was recorded. A local map representation representing an area of the road network on which the vehicle is travelling is then generated using at least some of the obtained images and the associated camera locations. The generated local map representation is compared with a section of a reference map, the reference map section covering the area of the road network on which the vehicle is travelling, and the geographical location and orientation of the vehicle within the road network is determined based on the comparison.

Abstract: An object of the present invention is to provide an analysis method capable of analyzing time-series images by a method simpler than ever.

Abstract: Embodiments of this application provide a charging method. The method includes that a session management function determines, a data flow within a protocol data unit, PDU, session and a charging key corresponding to the data flow within the PDU session, wherein a plurality of user plane functions are involved in the PDU session. For each of the plurality of user plane functions, the session management function sends to an online charging system, a quota request for a charging key without available quota for the UPF.

Abstract: This disclosure provides a method of detecting and deterring a target animal from a target area. A target area is positioned within the field of vision of a video camera connected to a computer processing system. An animal identification computer program using convolution neural networks and deep learning computer programs and camera images rapidly detects a target animal. The animal identification computer program is trained to identify target animals accurately using a learning algorithm and related machine learning technology. The time to deploy a deterrent against a target animal from the instant of detection is 2 seconds or less so that little or no time is available to the target animal to damage the target area.

Abstract: A stretchable net cover for a mobile electronic device that provides multiple grasp points and configurations, providing a user with a plurality of ways to hold and use their devices. A user may slide one or more fingers through openings in the netting material to grasp the device, may slide the netting over a forearm, or may hook the device to a purse, belt loop, carabiner, lanyard, retractable or static leash via the netting material. The cover may be provided as an add-on, i.e., designed for use over a mobile electronic device alone or a device enclosed within a case, or may be provided as part of a case for a mobile electronic device.