Abstract: A visual content search and retrieval platform comprising an object detection model, a coarse-grained image classification model, and a selection of fine-grained image classification models. During inference, the object detection model locates and extracts a query object in a query image; the coarse-grained image classification model classifies the query object's meta-category; the fine-grained image classification model trained and optimized for the meta-category of the query object is selected; and the selected fine-grained image classification model searches and retrieves from an image database images of objects most similar to the query object.

Abstract: A method of aligning and overlaying an AR virtual object in an electronic 3D viewing environment, comprising: receiving a motion comprising: a surrounding real-world scene image containing the real-world object captured by a camera of the electronic 3D viewing environment, and a camera real-world pose; recognizing a reference feature for aligning the virtual object to the real-world object; recording the motion and its corresponding reference feature; estimating a real-world object pose using the camera real-world pose and the reference feature; refining the estimated real-world object pose by pose estimation refinement by features and pose estimation refinement by images; and rendering the virtual object according to the estimated and refined real-world object pose in the 3D viewing environment.

Abstract: An image-guided diffusion network has two Convolution Neural Networks (CNNs). A RGB image and an IR image are concatenated with a Gaussian noise image and input to a denoising neural network that merges information from the RGB and IR images as noise is removed over many iterations. Then an enhancement neural network up-samples for Super Resolution (SR) and convolutes to generate a condition vector that controls Global Feature Modulation (GFM) at three convolution layers to generate a SRGFM enhanced fusion image. Timesteps are embedded using adaptive group normalization blocks within Adaptive Bottleneck Residual (ABR) blocks in the denoising network, which is a UNet having many levels of ABRs, and in the enhancement network before feature modulation. Global image features are detected by triple convoluting the image input to the enhancement network to generate the condition vector that controls feature modulation blocks at three layers of convolution.

Abstract: A speed bridge which provides high verification productivity in a hardware emulation system by allowing for time-efficient communication between a host/server and a device under test is disclosed. The speed bridge emulates the host to the device and the device to the host simultaneously while being in communication with both. When the message/response between the host and the device is fixed, the virtual host communicates with the virtual device. When the message/response requires physical interaction, the speed bridge passes the message/response between the host/server and the device.

Abstract: Systems and methods which provide single nanostructure-integrated metalens configurations implementing projector and light shaper functionality are described. A single nanostructure-integrated metalens may comprise an optical substrate having a preconfigured mapping of integrated nanostructures providing metasurfaces for modulating the behaviors of electromagnetic waves to implement a thin, flat lens. A period distance for the nanostructures, a height of the nanostructures, and a quantization for lateral sizes of the nanostructures may be selected for a particular single nanostructure-integrated metalens configuration. A first phase map may be computed with respect to structured light generation and a second phase map may be computed with respect to light collimation, wherein fusion of these phase maps may be used to define a preconfigured mapping for the nanostructures of a single nanostructure-integrated metalens implementation.

Abstract: A method for analyzing environmental sounds to identify repeating noises due to power equipment for purposes of controlling noise pollution is disclosed. Ambient sounds are sampled with microphones and the resulting data analyzed by an edge computing device. The samples are compared to stored data with a convolutional neural network by the edge computing device which also performs pitch shifting and cycle shifting to reduce false negatives. Positive results are reported via LoRa and/or 4G to a cloud platform and/or backend server.

Abstract: An AI-based system for the generation, evaluation, and prediction of performance of banner ads for presentation to specific customers or types of customers. A data set of existing banner ads, ad materials, and design and marketing parameters is used to generate a graph-based model. The graph-based model is used to generate original banner ads, which are then evaluated by a machine learning model, which assigns them scores. The highest-scoring banner ads are then presented to customers. The invention can also use a genetic algorithm in combination with iterative evaluation and generation to diversify design and choose the highest ranked banner ads.

Abstract: A method for processing multiple image data sets, each image data set generated by a different ultrasound frequency emitter, to obtain a single merged image data set for use to display a single merged image. The method can also include steps to account for non-ideal alignment between the different ultrasound frequency emitters.

Abstract: A swallowable capsule has a primary magnet along a long axis that aligns between movable electromagnets in a magnetic endoscopy machine. A flip magnet is orthogonal to the long axis. The capsule is rolled around the long axis to flip orientation of a side camera on a side of the capsule. An external base electromagnet below the patient's feet is energized to roll the capsule. The base magnet acts on the flip magnet but not on the orthogonal primary magnet when the long axis is orthogonal to a magnetic axis of the base electromagnet. Most components such as a controller, the side camera, a side laser, batteries, and an antenna, are mounted to a long Printed Circuit Board (PCB) along the long axis, while a front camera and laser are mounted to a front PCB that is mounted orthogonally to a front edge of the long PCB.

Abstract: A multidimensional knowledge transfer model for predicting CTR of a computational ad, the multidimensional knowledge transfer model comprises: a pre-processor for building an ad group node graph based on similarities among ad group nodes, an ad campaign node graph from merging the ad group node graph, and an ad account node graph from merging the ad campaign node graph. The multidimensional knowledge transfer model further comprises a multi-knowledge CTR prediction model for each of the ad account, ad campaign, and ad group layers. The multi-knowledge CTR prediction model predicts the respective node's CTR from the ad account node graph, ad campaign node graph, or ad group node graph, features of the audience group, and features of the node having its CTR predicted appended with the hidden vector of its parent node extracted from the upper layer multi-knowledge CTR prediction model.

Abstract: A 3D building model generator has a server that constructs a 3D texture model for each floor in a building by extruding a 2D floor plan to 3D using user-inputted heights, and adding photos captured by a smartphone or mobile device as textures to image planes for surfaces in the 3D model. The server displays on the smartphone a floor-plan portion, or a virtual 3D display based on the camera orientation with a facing wall highlighted and confirmed by a smartphone user. The user moves mapping points overlaid upon the photo to wall corners to map the photo to the facing surface in the model. The user flags rooms and exits such as doors and elevators as walkable. The server stacks the 3D texture model for the floors and connects them into a walkable network of the rooms and exits that are flagged as walkable. Hashing removes duplicate floorplans.

Abstract: A line-scanning three-dimensional sensing system measures a surface profile of an object. In the system, a dispersion optical module (DOM) performs a forward optical process of chromatically dispersing a polychromatic linear light beam into constituent narrowband linear light beams (CNLLBs) and focusing the CNLLBs on different focal planes to form a rainbow light pattern for illuminating a scanned surface of the object. The illuminated object displays an information-bearing color image (IBCI) containing height information of the scanned surface. The DOM captures the IBCI, and performs a backward optical process of optically condensing the captured IBCI to form an elongated light pattern. The backward optical process is an inverse of the forward one. A slit spatially filters the elongated light pattern to form an output light line. A height profile of the scanned surface is obtained by analyzing a spectral content at each point of the output light line.

Abstract: A method for detecting and tracking target object in a captured video using convolutional neural network (CNN) is provided. The method includes: inputting image data into a detecting model to generate detection results, wherein the detecting model is constructed by the CNN; inputting the image data into tracking models to generate tracking results; performing detection score enhancement operation according to the detection results and the tracking results to obtain enhanced detection results; matching the enhanced detection results and the tracking results by a matching operation; processing matched results and unmatched target detection results and unmatched target tracking results; and selectively updating the tracking models using tracking reliability estimation according to the matched results.

Abstract: A computer-implemented method for compliance checking related to goods in import supply chain logistics is provided, including: adding received legality data elements into a blockchain ledger corresponding to an importation of the goods; establishing a smart contract of the blockchain ledger; adding received sensing data elements corresponding to the smart contract into the blockchain ledger; adding electronic lock (e-Lock) status data corresponding to the smart contract into the blockchain ledger; and adding one or more inspection results corresponding to the smart contract into the blockchain ledger. Furthermore, determining, by the smart contract, whether the goods are within the compliance according to the received legality data elements, the received sensing data elements, the e-Lock status data and the one or more inspection results, wherein in response to determining that the goods are within the compliance, unsealing the e-Locks by sending a control signal to the e-Locks.

Abstract: A Hybrid TEE device allows a Trusted Execution Environment (TEE) by incorporating hardware comprising a Cache Purging Controller, a Memory Isolation Gateway, and a Memory Clean Up into a System on a Chip device, a general purpose computing device, or a special purpose or proprietary computing or electronic device. The addition of the hardware enables a method of protecting the Trusted Execution Environment and thus reducing vulnerability to malicious software or other program code.

Abstract: Described is a method and device for detecting a DTX state at a UCI receiver in a wireless communication system. The method comprises receiving a linear block encoded signal on an uplink at said UCI receiver and processing the received signal after resource element (RE) demapping to generate a soft bit sequence {right arrow over (s)}. The method includes selecting a plurality of bits in said generated soft bit sequence {right arrow over (s)} as comparison bits and comparing said selected comparison bits to corresponding bits in a reconstructed soft bit sequence {right arrow over (r)}. The reconstructed soft bit sequence {right arrow over (r)} is generated from a plurality of bits selected as reconstruction bits in said generated soft bit sequence {right arrow over (s)}.

Abstract: Described is a method of setting up a direct connection between edge sites on a 5G communications network. The method comprises: configuring 5G network functions (NFs) in the 5G communications network; configuring 5G NFs at the edge sites, if not already configured; and based on said configured NFs, setting up a direct tunnel between the edge sites according to a pre-defined tunnel policy.

Abstract: Systems and methods which provide for auditable data are described. Data management systems of embodiments facilitate storage and management of various data (e.g., information regarding transactions, customers, infrastructure, geographic locations, medical records, etc.) for supporting data creation, manipulation, storing, and/or sharing by and among different parties. Users may create new and/or update data, digitally sign the data, and store the digital signature and data on immutable storage using data structures of a data management system. A data structure in which data and associated metadata is stored on immutable storage in association with a digital signature corresponding to the data and user may be used in providing data traceability and auditability. Snapshots of merged data may be implemented with respect to data having many updates to avoid delays in merging update data for queries for that data. A collaborative rating mechanism may provide trust levels with respect to the data.

Abstract: Described is a method of calibrating a camera. The method comprises defining an area observed in a field of view (FOV) of the camera as a region of interest (ROI). The method includes selecting a plurality of physical reference points in said ROI and obtaining for each reference point physical position coordinates. The method also includes obtaining from said camera FOV or camera image data pixel locations for each point of reference and pairing each pixel location with the physical position coordinates of its respective reference point. Then, a relationship is derived from the paired pixel locations and physical position coordinates to enable a selected or identified pixel location in the camera FOV or camera image data to be transformed to physical position coordinates for a corresponding physical location in said ROI.

Abstract: The present invention provides a method and a system for remote imaging of explosive gases in an area. The method comprises: illuminating the area with a light source having a uniform light intensity distribution over an infrared wavelength range; acquiring images of the illuminated area with an image sensor through gas detection filters having bandpass central wavelength corresponding to absorption curves of target gases respectively; determining existence of the target gases based on the acquired images; predicting distribution of gas concertation for existing target gases respectively by using a non-linear prediction model; and constructing gas distribution images of the area based on the predicted distribution of gas concertation. The present invention can recognize different gases with overlapping absorption curves and provide more accurate prediction of gas concentrations.