Abstract: Provided is an efficient approach to establish a foundational video-text model for tasks including open-vocabulary video classification, text-to-video retrieval, video captioning and video question-answering. Some example implementations include a model which can be referred to as VideoCoCa. Example implementations reuse a pretrained image-text contrastive captioner (CoCa) model and adapt it to video-text tasks with little or minimal extra training. While previous works adapt image-text models with various cross-frame fusion modules (for example, cross-frame attention layer or perceiver resampler) and finetune the modified architecture on video-text data, aspects of the present disclosure leverage findings that the generative attentional pooling and contrastive attentional pooling layers in the image-text CoCa design are instantly adaptable to “flattened frame embeddings”, yielding a strong zero-shot transfer baseline for many video-text tasks.

Abstract: Examples described herein relate to processing packets. In some examples, based on receipt of a Hypertext Transfer Protocol (HTTP) packet at a network interface device, the HTTP packet comprising an HTTP body and HTTP header: provide the HTTP header, but not the HTTP body, for processing in user space; modify solely the HTTP header in user space; and in kernel space, combine the modified HTTP header and the HTTP body prior to transmission of the HTTP packet with modified HTTP header to a client.

Abstract: Provided is a stress blank angle eliminating device and method for a dynamic true triaxial electromagnetic Hopkinson bar. Objectives of eliminating influence of a stress blank angle in a sample in a test and preventing permanent deformation caused by mutual collision between the Hopkinson bars are achieved. The adjustable test device for eliminating the stress blank angle eliminates influence of the stress blank angle generated by the sample in a triaxial six-direction dynamic impact test process, and is convenient for obtaining real dynamic failure parameters of the sample.

Abstract: An electric valve/door, and a control apparatus and method thereof are provided. The electric valve/door includes a control apparatus. The control apparatus includes an alternating current (AC) asynchronous motor, a transmission device, and a valve driver. The valve driver is connected to the AC asynchronous motor. The AC asynchronous motor is connected to the valve/door through the transmission device. The valve driver is configured to collect information of a real-time torque of the AC asynchronous motor or the valve/door during operation, and take a set torque of the valve driver as an input and the real-time torque as a feedback to immediately correct an output torque of the valve driver or a limit value of the output torque by using a proportional-integral-derivative (PID) control algorithm, to meet requirements for response speed and control precision of each stage in opening and closing processes of the valve/door.

Abstract: Methods and systems for media trend identification of content sharing platforms are provided herein. A set of audiovisual embeddings that represent audiovisual features of a media item is obtained. A set of textual embeddings that represent textual features of the media item is obtained. The obtained set of audiovisual embeddings and the obtained set of textual embeddings are provided as an input to an artificial intelligence (AI) model trained to predict whether a respective media item is associated with one or more media trends of a platform based on given embeddings for the media item. One or more outputs of the AI model are obtained. A determination is made, based on the one or more outputs of the AI model, whether the media item is associated with the one or more media trends of the platform.

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 sealing system for connecting a drainage device to a vehicle window, includes a fixing section for being connected to the vehicle window glass and an engaging section for being detachably connected to the drainage device, the engaging section including a supporting body, an elastic arm, and a transition portion connecting the supporting body to the elastic arm; wherein the sealing system further includes a positioning element connected to the transition portion and the supporting body, the positioning element being at least partially elastically deformable; further, the positioning element includes an elastic leg connected to the transition portion and an engaging head connected to the supporting body, the engaging head having a cross section presenting a U-shape or V-shape, and the U-shape or the V-shape having an opening oriented towards the supporting body.

Abstract: A method for authorizing a blockchain data access on a blockchain platform includes: obtaining from a third-party institution a data access request containing a user ID of data access, a data scope, and a random number of a signature, which is generated by the third-party institution after receiving a random number sent by a client terminal and signing on the received random number; performing a verification on the random number of the signature and the data scope; and after the verification is passed, sending data ciphertext corresponding to the user ID to the third-party institution, such that the third-party institution decrypts the data ciphertext using a data secret key sent by the client terminal.

Abstract: The present invention discloses a fully bio-based, highly filled lignin-rubber masterbatch, a method for preparing same, and use thereof. The lignin-rubber masterbatch is prepared by a method comprising: (1) reacting a lignin, acetic acid, and oleic acid in the presence of a catalyst to give a modified lignin; and (2) blending the modified lignin and a rubber, and granulating to give the lignin-rubber masterbatch. The highly filled lignin-rubber masterbatch prepared by the present invention can replace the conventional reinforcing agent carbon black and provide a better reinforcing effect and higher mechanical properties for rubber materials. The present invention can also reduce the rubber content of the rubber composite materials while retaining the mechanical properties, thus featuring cost-efficiency.

Abstract: An electric valve/door, and a control apparatus and method thereof are provided. The electric valve/door includes a control apparatus. The control apparatus includes an alternating current (AC) asynchronous motor, a transmission device, and a valve driver. The valve driver is connected to the AC asynchronous motor. The AC asynchronous motor is connected to the valve/door through the transmission device. The valve driver is configured to collect information of a real-time torque of the AC asynchronous motor or the valve/door during operation, and take a set torque of the valve driver as an input and the real-time torque as a feedback to immediately correct an output torque of the valve driver or a limit value of the output torque by using a proportional-integral-derivative (PID) control algorithm, to meet requirements for response speed and control precision of each stage in opening and closing processes of the valve/door.

Abstract: A method, a system and a device for determining interlaminar shear strength of a composite laminated plate are provided. The method includes following steps: constructing a force-displacement curve of a composite laminated plate specimen based on strain data, corresponding to different displacements of the indenter, of the composite laminated plate specimen; determining a dominant strain based on plane strain nephograms; reading any defect in the plane strain nephogram corresponding to one displacement as a current defect, and determining a test force acting on the whole composite laminated plate specimen corresponding to the reference displacement of the indenter as a critical strain value; determining a product of the critical strain value and shear modulus under the dominant strain as the shear strength between adjacent single layers to be tested.

Abstract: Methods and systems for media item characterization based on multimodal embeddings are provided herein. A media item including a sequence of video frames is identified. A set of video embeddings representing visual features of the sequence of video frames is obtained. A set of audio embeddings representing audio features of the sequence of video frames is obtained. A set of audiovisual embeddings is generated based on the set of video embeddings and the set of audio embeddings. Each of the set of audiovisual embeddings represents a visual feature and an audio feature of a respective video frame of the sequence of video frames. One or more media characteristics associated with the media item are determined based on the set of audiovisual embeddings.

Abstract: The present disclosure provides a display control method of a spliced screen including a plurality of sub-display screens, including: sending a display control signal to each sub-display screen in a wireless sending manner according to a time code signal; receiving, by each sub-display screen, the display control signal in a wired receiving manner, and reporting status information of each sub-display screen to the display controller; and controlling, by the display controller, displaying of each sub-display screen according to the status information of each sub-display screen, where the time code signal indicates a current time point; the display control signal is configured for controlling the sub-display screen to perform corresponding display processing on display content stored, so that the sub-display screens display a target display content together. The present disclosure further provides a display control device, a display system, an electronic device, and a computer readable storage medium.

Abstract: The present disclosure relates to a method for predicting an amount of water-sealed gas in a high-sulfur water-bearing gas reservoir. The method solves the problem that no method has yet been proposed for predicting the amount of water-sealed gas in a high-sulfur water-bearing gas reservoir. According to the technical solution, the method includes: considering that the volume of the gas reservoir does not change during the production of the constant-volume gas reservoir, deriving, based on a material balance method, a material balance equation of the high-sulfur water-bearing gas reservoir in consideration of water-sealed gas and water-soluble gas, solving and drawing a chart of water-sealed gas in the high-sulfur water-bearing gas reservoir by an iterative algorithm, obtaining a recovery factor of the high-sulfur water-bearing gas reservoir in consideration of water-sealed gas and water-soluble gas, and further obtaining the amount of water-seal gas in the high-sulfur water-bearing gas reservoir.

Abstract: Provided is a portable display device, which comprises a clamping mechanism and a display body. The clamping mechanism comprises a bracket, a left mounting member, a left connecting member, a right mounting member, and a right connecting member. The left end and the right end of the bracket are respectively provided with a mounting cavity. The left connecting member and the right connecting member are respectively provided with a left L-shaped clamping element and a right L-shaped clamping element, which are symmetrically arranged. The portable display device of the disclosure can accommodate the varied usage habits of different users.

Abstract: A pneumatic soft grabbing sensing device includes a support; a flexible grabbing mechanism including at least two flexible claws, wherein the at least two flexible claws are arranged on the support in a circumferential direction, each flexible claw includes a telescopic air bag, an adapter mechanism and a flexible plate connected with the adapter mechanism, the telescopic air bag is hollow and elongated, the telescopic air bag includes a closed end, an open end and an air bag main body connected between the closed end and the open end, the air bag main body includes at least two air bag bodies and a node portion formed between adjacent air bag bodies, and the adapter mechanism is connected to the node portion, the closed end and open end; and a sensing mechanism including at least two sensing parts, the sensing parts being arranged on inner sides of the corresponding flexible plates.

Abstract: Disclosed are a nebulization cup and application in nebulized inhalation administration thereof, and especially application in nebulized inhalation administration of a preventive and/or therapeutic drug for a respiratory disease (such as SARS-CoV-2 vaccine). After adding an antistatic agent, the nebulization cup can effectively maintain the stability of drug mist within a certain period of time, with stable particle size, less drug residue in the cup, thus ensuring effective inhalable amount, and the administration operation is simple and convenient, thus the nebulization cup can significantly improve the inoculation efficiency and can be used for large-scale inoculation.

Abstract: Aspects presented herein relate to methods and devices for graphics processing including an apparatus, e.g., GPU. The apparatus may obtain an indication of a set of primitives for a draw call operation. The apparatus may also identify a subset of primitives in the set of primitives, each of the subset of primitives including a primitive portion that is outside of a viewing frustum for the draw call operation, and the primitive portion corresponding to less than all of each of the subset of primitives. Further, the apparatus may calculate an area of each of the subset of primitives including the primitive portion that is outside of the viewing frustum. The apparatus may also perform, or refrain from performing, a clipping operation for each of the subset of primitives based on the area of each of the subset of primitives being less than or greater than an area threshold.

Abstract: A power-supply mode switching circuit and an aerosol generating device are provided. The power-supply mode switching circuit includes an external-power-source power-supply circuit, a protection circuit, and a battery power-supply circuit. A node where the external-power-source power-supply circuit is connected to the protection circuit serves as a power-source input end. A node where the external-power-source power-supply circuit is connected to the battery power-supply circuit serves as a power supply end. The protection circuit is configured to control the battery power-supply circuit to be cut off when the power-source input end is connected to the external power source, to make the external power source supply power to the power supply end. The protection circuit is further configured to control the battery power-supply circuit to be conducted when the power-source input end is not connected to the external power source, to make the battery supply power to the power supply end.