Abstract: A method for obtaining a cover image includes: obtaining a plurality of first cropped images of an original image corresponding to a candidate resource; obtaining an aesthetic score of each of the plurality of first cropped images; and determining a target cover image of the candidate resource from the plurality of first cropped images based on the aesthetic score of each first cropped image.

Abstract: A large model-based recommendation method includes: determining description information of interested content corresponding to a target user; inputting a content to be recommended, the description information of interested content and current popular search sentences into a large model to generate at least one recommendation card corresponding to the content to be recommended, in which the recommendation card contains a recommendation word associated with the content to be recommended; obtaining a current behavior characteristic of the target user; and in response to the current behavior characteristic satisfying a display condition of the recommendation card, displaying the recommendation card corresponding to at least one content to be recommended.

Abstract: Provided is a method for training an image cropping model, a method for processing an image, an electronic device and a storage medium, relating to the field of deep learning and image processing technology. The training method includes: obtaining sample data, wherein the sample data at least includes: a sample image, a first cropped image obtained by cropping the sample image in a first manner, and a second cropped image obtained by cropping the sample image in a second manner; determining a target loss function; and using at least the sample data and the target loss function to perform model training on a preset image cropping model to obtain a target image cropping model.

Abstract: An embodiment includes a method of multi-user gesture recognition includes determining positions of a plurality of users based on a first radar information from at least one radar, determining a plurality of regions of interest for the plurality of users based on positions of the plurality of users, in a first duty cycle of a plurality of duty cycles, obtaining a second radar information using a first radar configuration that includes a first set of parameters for gesture recognition at a range of distances based on the plurality of regions of interest, in a second duty cycle, obtaining a third radar information using a second radar configuration, and detecting a first user gesture from a first user of the plurality of users and a second user gesture from a second user of the plurality of users based on the second and third radar information.

Abstract: A method includes obtaining a channel impulse response (CIR) based on multiple ultrawide band (UWB) measurements between a mobile device and a target object. The method also includes estimating a first distance between the mobile device and the target object based on an energy of the CIR. The method also includes estimating a second distance between the mobile device and the target object based on a phase of the CIR. The method also includes obtaining an estimated distance between the mobile device and the target object based on at least one of the first distance and the second distance.

Abstract: A method includes determining, for each time interval in a series of time intervals, (i) whether at least one wireless signal measurement was received in the time interval, each wireless signal measurement associated with a pair of anchors, and (ii) whether at least one step was detected in the time interval. The method also includes selecting a state of a tracking filter state machine based on whether the at least one wireless signal measurement was received in the time interval and whether the at least one step was detected in the time interval. The method further includes determining a location of an object using the tracking filter state machine based on the selected state.

Abstract: A method includes receiving at least one wireless signal measurement and motion sensor measurements. The method also includes generating a location estimate based on the at least one wireless signal measurement. The method also includes determining whether a step is present based on the motion sensor measurements. The method also includes, in response to determining that a step is present, determining a step heading offset based on the location estimate and the motion sensor measurements, and determining a step length and heading based on the motion sensor measurements and the step heading offset. The method also includes determining a location of an object based on at least one of (i) the at least one wireless signal measurement or (ii) the step length and heading.

Abstract: Embodiments of the present disclosure provide a method and system for controlling an image-guided interventional puncture device. The method may include: obtaining an initial movement state of an imaging device and/or a surgical robot; controlling a target movement state of the surgical robot and/or the imaging device based on the initial movement state.

Abstract: A method for presence detection using ultra-wide band (UWB) radar includes obtaining a set of centroids based on UWB radar measurements. The method includes, for each respective centroid among the set of centroids: classifying the respective centroid as one among human movement and non-human movement, based on a set of features for the respective centroid satisfying a human-movement condition; and when the respective centroid is classified as the human movement, determining a two-dimensional (2D) location of human movement based on the UWB radar measurements. The method includes updating a current state value that indicates whether a human presence is detected inside boundaries of a 3D space, based at least in part on the classification of each of the respective centroids among the set of centroids.

Abstract: A switching power supply circuit includes a power transistor circuit, including a plurality of power transistors coupled in series between a power supply end and a ground, where a first electrode of a first power transistor in the plurality of power transistors is coupled to the power supply end by using a second power transistor in the plurality of power transistors, and a second electrode of the first power transistor is coupled to the ground by using a third power transistor in the plurality of power transistors; a drive circuit, configured to drive the second power transistor and the third power transistor; and a first transistor, where both a gate and a first electrode of the first transistor are coupled to a gate of the third power transistor, and a second electrode of the first transistor is coupled to the ground by using a first switch.

Abstract: In a voltage conversion circuit, a sampling signal and a triangular wave signal are synthesized into a first signal, and the first signal and a second signal are compared to output a pulse-width modulation (PWM) signal. A frequency of the PWM signal may be determined by a frequency of the triangular wave signal. Therefore, the frequency of the PWM signal is fixed using the frequency-fixed triangular wave signal. Turning on or switching off of a power transistor in a voltage conversion subcircuit is controlled according to the PWM signal to convert a received direct current input voltage into a direct current output voltage, thereby implementing a fixed operating frequency for the voltage conversion circuit.

Abstract: Embodiment of the present disclosure provides a method for obstacle avoidance of a surgical robot. The method may include: collecting first data through a first acquisition device, wherein the first data is image data of a space in which a target subject is located; constructing a safety zone for the target subject based on the first data; collecting second data outside the safety zone through a second acquisition device, wherein the second data is image data of the target subject; and constructing a three-dimensional model of the target subject based on the second data, wherein the three-dimensional model of the target subject may be used for obstacle avoidance detection during an operation of the surgical robot.

Abstract: The present disclosure provides a collision prediction system and method. The method may include obtaining a target model of a target device. The method may also include determining first posture information of the target model and second posture information of an operating apparatus, wherein the target device is set on an end of the operating apparatus. The method may further include generating a collision prediction between the target model and the operating apparatus based on the first posture information and the second posture information, the collision prediction indicating whether there is a space accommodating the target device during an operation of the operating apparatus.

Abstract: A hand-eye calibration method, apparatus, system, and storage medium for a robot are provided. The method includes: obtaining images of a target body of a calibration target acquired by an imaging device when the robot (110) is in different poses; for any pose, determining, based on at least one image collected in the pose, a first transformation relationship between a first coordinate system of the calibration target and a second coordinate system of the imaging device in the pose; obtaining a second transformation relationship between a third coordinate system and a fourth coordinate system of the robot (110); and determining, based on the first transformation relationships and the second transformation relationships corresponding to the different poses, a third transformation relationship between the second coordinate system and one of the third coordinate system and the fourth coordinate system. Further provided is a calibration target.

Abstract: The present disclosure relates to a surgical robot and a method of controlling the surgical robot. The method may include obtaining pose information of an end tool and providing first feedback information to an operator based on the pose information.

Abstract: This application provides a surgical robot control method, a computer device, and a surgical robot system. The surgical robot control method includes: receiving a user demand, and generating an interactive control command; generating a motion control command according to the interactive control command; and controlling a terminal of a mechanical arm to perform the motion control command. The motion control command includes controlling the terminal of the mechanical arm to act in accordance with a plurality of motion modes. The surgical robot control method may control the terminal of the mechanical arm to act in accordance with the plurality of motion modes through the motion control command, realize different motion schemes in various clinical application scenarios, and realize controlling the terminal of the mechanical arm to be flexibly switched between the plurality of motion modes in any application scenario through the interactive control command.

Abstract: The present disclosure discloses example DC-DC converters. One example DC-DC converter includes a first capacitor, a second capacitor, a switched capacitor SC circuit, and an inductor circuit. An input end of the SC circuit is coupled to a voltage input end, and an output end of the SC circuit is coupled to a voltage output end. The first capacitor and the second capacitor are sequentially connected in series between the voltage input end and the voltage output end. One end of the inductor circuit is selectively coupled between the first capacitor and the second capacitor or to a ground end, and the other end of the inductor circuit is coupled to the voltage output end.

Abstract: Efficient writing to lower-performance storage tiers of a multi-tier storage system evaluates user intention when determining a write process that ultimately writes target data to a priority storage tier according to the user intention. Temporary high-performance storage tiers serve as coordinator tiers to achieve the efficient writing process for writing substantial or massive datasets to local storage.

Abstract: Efficient writing to lower-performance storage tiers of a multi-tier storage system evaluates user intention when determining a write process that ultimately writes target data to a priority storage tier according to the user intention. Temporary high-performance storage tiers serve as coordinator tiers to achieve the efficient writing process for writing substantial or massive datasets to local storage.

Abstract: An actuation assembly of an auto-injector is used for combining a container assembly with a vial. The actuation assembly of the auto-injector includes: an actuation sleeve, an execution sleeve, a first resilient element, a rear cap, a push rod, and a second resilient element. The actuation assembly as a whole corresponds to one axis. The actuation sleeve is designed with two symmetrical cantilevers of the execution sleeve and corresponding parts, which is used not only to control the movement of the push rod before and after the actuation sleeve is actuated, but also to provide an audible sound to alert the assembly in place and the end of automatic injection. Accordingly, when the actuation sleeve is actuated, the push rod can be moved toward the vial to accomplish the effect of automatic injection.