Abstract: A computing device receives a video feed. The video feed is divided into a sequence of video segments. For each video segment, visual features of the video segment are extracted. A predicted spectrogram is generated based on the extracted visual features. A synthetic audio waveform is generated from the predicted spectrogram. All synthetic audio waveforms of the video feed are concatenated to generate a synthetic soundtrack that is synchronized with the video feed.

Abstract: The present disclosure provides an array substrate, a method for manufacturing the array substrate, a display panel and a display device. The array substrate includes: a substrate; a planarization layer on a side of the substrate; a pixel defining layer configured to define a pixel opening region and located on a side of the planarization layer away from the substrate; an anode in the pixel opening region and on a side of the planarization layer away from the substrate. The array substrate further includes an intermediate insulation layer between the planarization layer and the pixel defining layer. The intermediate insulation layer has a chemical polarity between a chemical polarity of the planarization layer and a chemical polarity of the pixel defining layer.

Abstract: A vehicle light signal detection and recognition method, system, and computer program product include bounding, using a coarse attention module, one or more regions of an image of an automobile including at least one of a brake light and a signal light generated by automobile signals which include illuminated sections to generate one or more bounded region, removing, using a fine attention module, noise from the one or more bounded regions to generate one or more noise-free bounded regions, and identifying the at least one of the brake light and the signal light from the one or more noise-free bounded regions.

Abstract: A light-emitting substrate includes; a base, an isolation portion disposed on the base and located in an isolation region located outside a light-emitting region, and a second insulating pattern located in the light-emitting region. The isolation portion includes a first conductive pattern, a second conductive pattern and a first insulating pattern that are sequentially stacked on the base; an orthogonal projection of the first conductive pattern on the base is located within an orthogonal projection of the second conductive pattern on the base; and a side face of the first conductive pattern proximate to the light-emitting region and a corresponding side face of the second conductive pattern proximate to the light-emitting region have a first gap therebetween. A side face of the second insulating pattern proximate to the first insulating pattern and a side face of the first insulating pattern proximate to the second insulating pattern have a second gap therebetween.

Abstract: The present application provides techniques for displaying music score segments in target music videos. The techniques comprise determining a digital music score corresponding to a piece of music comprised in a target music video; determining a segment of the digital music score corresponding to a current playing progress of the target music video based at least in part on a playing progress of the target music video; generating an image of a music score segment corresponding to the segment of the digital music score based on a predetermined condition; and presenting the image on a corresponding interface of playing the target music video.

Abstract: The exemplary embodiments relate to converting standard dynamic range (SDR) content to high dynamic range (HDR) content. An SDR image may be decomposed into a base layer of the SDR image that includes low frequency information from the SDR image and a detail layer of the SDR image that includes high frequency information from the SDR image. A base layer of an HDR image may be generated using the base layer of the SDR image and a detail layer of the HDR image may be generated using the detail layer of the SDR image. An HDR image is then generated using the base layer of the HDR image and the detail layer of the HDR image.

Abstract: Disclosed are an information transmission method for a sidelink, and a user equipment and a computer-readable storage medium. The method comprises: detecting a control channel indication sequence at a pre-set position of each time slot; and based on the detected control channel indication sequence, determining, in the time slot where the sequence is located, a time-frequency resource position where a corresponding physical sidelink control channel is located, wherein the control channel indication sequences correspond to the physical sidelink control channels on a one-to-one basis. By applying the solution, an actual time-frequency resource position of a PSCCH can be indicated to a UE in a sidelink of an NR system.

Abstract: The present disclosure relates to the field of optical lenses and provides a camera optical lens. The camera optical lens includes, from an object side to an image side: a first lens made of a plastic material; a second lens made of a plastic material; a third lens made of a glass material; a fourth lens made of a plastic material; a fifth lens made of a plastic material; a sixth lens made of a plastic material; and a seventh lens made of a plastic material. The camera optical lens satisfies following conditions: 1.51?f1/f?2.50; 1.70?n3?2.20; 0.00?f3/f4?2.00; ?10.00?(R13+R14)/(R13-R14)?10.00; and 0.01?d5/TTL?0.20. The camera optical lens can achieve a high imaging performance while obtaining a low TTL.

Abstract: An image privacy perception method based on deep learning, including the following steps: S1, building a privacy classification data set with labeled categories, and training a privacy perception network with a transfer learning method; S2, recognizing a privacy image using a deep convolutional neural network oriented to privacy perception; and S3, extracting an attention profile according to deep convolutional features of the neural network, and locating an attention focusing region to complete the perception of an image privacy region. The method has the following beneficial effects: by completing end-to-end training and testing based on the deep neural network, the privacy image can be accurately distinguished with the privacy region located, facilitating the selective protection of the privacy information in the image.

Abstract: Embodiments of the present invention are directed to a computer-implemented method for action localization. A non-limiting example of the computer-implemented method includes receiving, by a processor, a video and segmenting, by the processor, the video into a set of video segments. The computer-implemented method classifies, by the processor, each video segment into a class and calculates, by the processor, importance scores for each video segment of a class within the set of video segments. The computer-implemented method determines, by the processor, a winning video segment of the class within the set of video segments based on the importance scores for each video segment within the class, stores, by the processor, the winning video segment from the set of video segments, and removes the winning video segment from the set of video segments.

Abstract: The present invention provides a 6-dithio-substituted-2?-deoxyguanosine compound and a preparation method thereof and use thereof in anti-tumor drugs. According to the present invention, a series of brand-new thioredoxin-1 (Trx-1) inhibitors are designed and synthesized by using 2?-deoxyguanosine as a structure mother nucleus. The compound according to the present invention is a structure of formula I, and Trx-1 activity inhibition testing is performed on the type of compounds, showing apparent inhibitory activity. It is indicated from in vitro and vivo researches that these compounds show significant anti-tumor effects on multiple types of tumors, can be applied to drugs for tumors related to human body cell proliferation, and provide a new path for the research of anticancer drugs.

Abstract: A business prediction method includes: obtaining a first business sample set and a second business sample set; performing training based on the first business sample set and the second business sample set to obtain a business prediction model, and predicting received to-be-predicted business information based on the business prediction model to obtain a business prediction result corresponding to the received to-be-predicted business information. A business prediction apparatus is further provided. The business prediction method and the business prediction apparatus take into account data features of some business samples of being rejected in a business validation, while considering business samples of passing the business validation. This restores a business scenario, reduces the waste of costs of the rejected samples, and balances demands for a modeling sample and a rejected sample reasonably when there are insufficient samples of passing the business validation.

Abstract: Dual-modality relation networks for audio-visual event localization can be provided. A video feed for audio-visual event localization can be received. Based on a combination of extracted audio features and video features of the video feed, informative features and regions in the video feed can be determined by running a first neural network. Based on the informative features and regions in the video feed determined by the first neural network, relation-aware video features can be determined by running a second neural network. Based on the informative features and regions in the video feed, relation-aware audio features can be determined by running a third neural network. A dual-modality representation can be obtained based on the relation-aware video features and the relation-aware audio features by running a fourth neural network. The dual-modality representation can be input to a classifier to identity an audio-visual event in the video feed.

Abstract: This application relates to an electrode plate, including a current collector and an electrode active material layer disposed on at least one surface of the current collector. The current collector includes a support layer and a conductive layer disposed on at least one surface of the support layer, where single-side thickness D2 of the conductive layer satisfies: 30 nm?D2?3 ?m, the support layer includes a polymer material or a polymer composite material, and thickness D1 of the support layer satisfies: 1 ?m?D1?20 ?m. The electrode active material layer includes an electrode active material, a binder, and a conductive agent, where total thickness Dtotal of the electrode active material layer satisfies: 200?Dtotal/D1?1, and based on total weight of the electrode active material layer, the binder content in the electrode active material layer is not less than 1 wt %.

Abstract: This application relates to the battery field, and specifically, to an electrode plate, an electrochemical device, and an apparatus. The electrode plate of this application includes a current collector and an electrode active material layer disposed on at least one surface of the current collector, where the current collector includes a support layer and a conductive layer disposed on at least one surface of the support layer, a single-sided thickness D2 of the conductive layer satisfies 30 nm?D2?3 ?m, the support layer is made of a polymer material or a polymer composite material, and a thickness D1 of the support layer satisfies 1 ?m?D1?30 ?m; and the electrode active material layer includes an electrode active material, a binder, and a conductive agent.

Abstract: This application relates to the battery field, and specifically, to a positive electrode plate, an electrochemical apparatus, and an apparatus. The positive electrode plate in this application includes a current collector and an electrode active material layer disposed on at least one surface of the current collector, where the current collector includes a support layer and a conductive layer disposed on at least one surface of the support layer, a single-sided thickness D2 of the conductive layer satisfies 30 nmD23 ?m, a thickness D1 of the support layer satisfies 1 ?mD130 ?m, and the support layer is made of a polymer material or a polymer composite material; and the electrode active material layer includes an electrode active material, a binder, and a conductive agent, and a total thickness Dtotal of the electrode active material layer is not greater than 170 ?m.

Abstract: This application relates to the battery field, and specifically, to a positive electrode plate, an electrochemical apparatus, and an apparatus. The positive electrode plate in this application includes a current collector and an electrode active material layer disposed on at least one surface of the current collector, where the current collector includes a support layer and a conductive layer disposed on at least one surface of the support layer. A single-sided thickness D2 of the conductive layer satisfies 30 nm?D2?3 ?m. A thickness D1 of the support layer satisfies 1 ?m?D1?30 ?m. The support layer is made of a polymer material or a polymer composite material. The electrode active material layer includes electrode active materials, a binder, and a conductive agent.

Abstract: A plurality of pieces of emotional state information corresponding to a plurality of speech frames in a current utterance are obtained based on a first neural network model; statistical operation is performed on the plurality of pieces of emotional state information, to obtain a statistical result, and then the emotional state information corresponding to the current utterance is obtained based on a second neural network device, the statistical result corresponding to the current utterance, and statistical results corresponding to a plurality of utterances before the current utterance.

Abstract: A housing of a device comprises a screw boss that extends outwardly from a first base plate proximate to a corner of the housing. Additionally, the housing of the device includes a force dissipation member. The force dissipation member includes a central portion that is sized to detachably attach around the screw boss. The force dissipation member further includes at least two arms that extend outwardly from the central portion, wherein each arm of the at least two arms is attached to a different side of the two walls of the housing.

Abstract: Systems, methods, and related technologies for device classification are described. In certain aspects, traffic data associated with a device and data from an external system can be accessed. The data can be processed to determine a device classification for the device. An action can be initiated based on the classification.