Abstract: A method is provided for visual inspection. The method includes learning, by a processor, group disentangled visual feature embedding vectors of input images. The input images include defective objects and defect-free objects. The method further includes generating, by the processor using a weight generation network, classification weights from visual features and semantic descriptions. Both the visual features and the semantic descriptions are for predicting defective and defect-free labels. The method also includes calculating, by the processor, a cosine similarity score between the classification weights and the group disentangled visual feature embedding vectors. The method additionally includes episodically training, by the processor, the weight generation network on the input images to update parameters of the weight generation network.

Abstract: The present disclosure relates to a method, device and computer program product for managing a backup system. In a method for managing a backup system, at least one backup attribute of a backup demand on the backup system is determined, the at least one backup attribute describing at least one aspect of a demand on the backup system from a user of the backup system. A plurality of backup records of a plurality of historical backup operations performed on the backup system are obtained, respectively. Data distribution associated with the at least one backup attribute is determined based on the plurality of backup records.

Abstract: A computer-implemented method is provided for generating following up questions for multi-hop bridge-type question answering. The method includes retrieving a premise for an input multi-hop bridge-type question. The method further includes assigning, by a three-way neural network based controller, a classification of the premise against the input multi-hop bridge-type question as being any of irrelevant, including a final answer, or including intermediate information. The method also includes outputting the final answer in relation to a first hop of the multi-hop bridge-type question responsive to the classification being including the final answer. The method additionally includes generating a followup question by a neural network and repeating said retrieving, assigning, outputting and generating steps for the followup question, responsive to the classification being including the intermediate information.

Abstract: According to embodiments of the present disclosure, a method, a device and a computer program product for managing data backup are provided.

Abstract: Embodiments of the present disclosure relate to a method, a device and a computer program product for managing backup data. The method comprises receiving a first request from a user for managing backup data on a first virtual machine management platform. The method further comprises sending a second request comprising a user identifier and a platform identifier to a backup server. The method further comprises receiving, from the backup server, a first virtual machine identifier of the first virtual machine used by the user on the second virtual machine management platform and backup records corresponding to the first virtual machine identifier. The method further comprises determining, based on the first virtual machine identifier, whether a second virtual machine matched with the first virtual machine is present in the first virtual machine management platform.

Abstract: A special-purpose die for shaping an aluminum-magnesium alloy by rotating extrusion is provided, including a male die and a female die, wherein a trapezoidally-sectioned groove is formed at an end portion of a working area of the male die, an inner portion of the male die is hollow, with the hollow inner portion having sections of equal area, a circumferential wall of a die cavity of the female die is provided with at least two symmetrical axial grooves, and a cavity is formed inside a clamping part of the female die. The present disclosure remarkably reduces the axial extrusion force such that the deformation of the formed workpiece is more uniform, which greatly improves the mechanical property of the formed workpiece.

Abstract: A camera device and camera system for video-based workplace safety is provided. The camera device includes at least one imaging sensor configured to capture one or more video sequences in a workplace environment having a plurality of machines therein. The video camera further includes a processor. The processor is configured to generate a plurality of embedding vectors based on a plurality of observations. The observations include (i) a subject, (ii) an action taken by the subject, and (iii) an object on which the subject is taking the action on. The subject and object are constant. The processor is further configured to generate predictions of one or more future events based on one or more comparisons of at least some of the plurality of embedding vectors. The processor is configured to generate a signal for initiating an action to the at least one of the plurality of machines to mitigate harm.

Abstract: Systems and methods for training a recursive neural network (RNN) is provided. The method includes generating, by the processor using the RNN, a plurality of embedding vectors based on a plurality of observations, wherein the observations include (i) a subject, (ii) an action taken by the subject, and (iii) an object on which the subject is taking the action on, wherein the subject and object are constant. The method further includes generating, by the processor, predictions of one or more future events based on one or more comparisons of at least some of the plurality of embedding vectors. The method also includes initiating, by the processor, based on the predictions, an action to a hardware device to mitigate expected harm to at least one item selected from the group consisting of the hardware device, another hardware device related to the hardware device, and a person related to the hardware device.

Abstract: A luminous display substrate includes a base substrate, a light reflective and electrically conductive film layer on the base substrate, and a light-transmittance metal film layer on a side of the light reflective and electrically conductive film layer away from the base substrate. A side of the light reflective and electrically conductive film layer adjacent to the light-transmittance metal film layer is recessed towards the base substrate to form a curved surface.

Abstract: A computer-implemented method and system are provided for training a model for New Class Categorization (NCC) of a test image. The method includes decoupling, by a hardware processor, a feature extraction part from a classifier part of a deep classification model by reparametrizing learnable weight variables of the classifier part as a combination of learnable variables of the feature extraction part and of a classification weight generator of the classifier part. The method further includes training, by the hardware processor, the deep classification model to obtain a trained deep classification model by (i) learning the feature extraction part as a multiclass classification task, and (ii) episodically training the classifier part by learning a classification weight generator which outputs classification weights given a training image.

Abstract: Systems and methods for processing video are provided. The method includes receiving a text-based description of active scenes and representing the text-based description as a word embedding matrix. The method includes using a text encoder implemented by neural network to output frame level textual representation and video level representation of the word embedding matrix. The method also includes generating, by a shared generator, frame by frame video based on the frame level textual representation, the video level representation and noise vectors. A frame level and a video level convolutional filter of a video discriminator are generated to classify frames and video of the frame by frame video as true or false. The method also includes training a conditional video generator that includes the text encoder, the video discriminator, and the shared generator in a generative adversarial network to convergence.

Abstract: A method is provided for visual inspection. The method includes learning, by a processor, group disentangled visual feature embedding vectors of input images. The input images include defective objects and defect-free objects. The method further includes generating, by the processor using a weight generation network, classification weights from visual features and semantic descriptions. Both the visual features and the semantic descriptions are for predicting defective and defect-free labels. The method also includes calculating, by the processor, a cosine similarity score between the classification weights and the group disentangled visual feature embedding vectors. The method additionally includes episodically training, by the processor, the weight generation network on the input images to update parameters of the weight generation network.

Abstract: A light shield device, a light shield control method, an electronic device and a vehicle are provided. The light shield device includes a controller module and a light shield structure, the light shield structure includes an accommodating cavity, an electrowetting material being in the accommodating cavity and having a light-shielding property, and at least one fluid tube which is communicated with the accommodating cavity; and the controller module is configured to apply a voltage to the light shield structure, so that wettability of a surface of the at least one fluid tube is changed to allow the electrowetting material to fill the at least one fluid tube.

Abstract: A display device includes a housing. A plurality of display screens are provided in the housing, and each display screen includes a transparent display panel. The housing is provided with a viewing portion for viewing an image displayed on the transparent display panel, the plurality of display screens are successively arranged in a direction leaving away from the viewing portion, a plurality of the transparent display panels are configured for respectively displaying images at different depths of field in a same scene, and the image displayed on the transparent display panel has a depth of field which is associated with a distance between the transparent display panel and the viewing portion.

Abstract: A pair of 3D glasses and a 3D display system are disclosed in the field of display technology. The pair of 3D glasses includes a light polarizing component and a light filtering component. The light filtering component is arranged at the light emitting side of the light polarizing component. The light polarizing component includes a plurality of photonic crystals, each of which is configured to reflect preset-color light among different-color light incident into the photonic crystal to enable the preset-color light to be deflected so as to ensure an optical path difference among the different-color light incident into the light filtering component from the light polarizing component. The light filtering component includes two light filters, one of which is configured to filter out the preset-color light among the different-color light and the other of which is configured to filter out light other than the preset-color light among the different-color light.

Abstract: A luminous display substrate includes a base substrate, a light reflective and electrically conductive film layer on the base substrate, and a light-transmittance metal film layer on a side of the light reflective and electrically conductive film layer away from the base substrate. A side of the light reflective and electrically conductive film layer adjacent to the light-transmittance metal film layer is recessed towards the base substrate to form a curved surface.

Abstract: The present disclosure provides an OLED display panel including a display substrate having a displaying region and a surrounding region, a package substrate covering the display substrate, and a UV-cured encapsulant composition located at the surrounding region of the display substrate so as to seal a gap between the display substrate and the package substrate. An organic light-emitting element is formed at the displaying region of the display substrate. The UV-cured encapsulant composition contains in weight percentage 1% to 90% of inorganic nano particles.

Abstract: Systems and methods for predicting new relationships in the knowledge graph, including embedding a partial triplet including a head entity description and a relationship or a tail entity description to produce a separate vector for each of the head, relationship, and tail. The vectors for the head entity, relationship, and tail entity can be combined into a first matrix, and adaptive kernels generated from the entity descriptions can be applied to the matrix through convolutions to produce a second matrix having a different dimension from the first matrix. An activation function can be applied to the second matrix to obtain non-negative feature maps, and max-pooling can be used over the feature maps to get subsamples. A fixed length vector, Z, flattens the subsampling feature maps into a feature vector, and a linear mapping method is used to map the feature vectors into a prediction score.

Abstract: A display panel and a method for manufacturing the same, an electronic paper and a driving method thereof are provided. The display panel includes: a first base substrate, a second base substrate arranged oppositely to the first base substrate, and multiple temperature-induced transmittance reversible components which not contacted with each other and arranged in an array between the first base substrate and the second base substrate. Each of the temperature-induced transmittance reversible components includes: a temperature-induced transmittance reversible unit and a temperature control unit. The temperature control unit is used to adjust temperature of the temperature-induced transmittance reversible unit to control transmittance of the temperature-induced transmittance reversible unit with respect to the visible light.

Abstract: The present disclosure provides a display substrate, a display panel and a display device. The display substrate forms a cell with an opposite substrate. A black matrix is arranged on the display substrate or the opposite substrate. The display substrate includes a base substrate and a photoluminescent device arranged on the base substrate. The photoluminescent device is arranged at a position corresponding to the black matrix and located at a side of the black matrix facing a backlight module. A projection of the photoluminescent device onto the base substrate at least partially overlaps a projection of the black matrix onto the base substrate. The photoluminescent device is configured to emit light under an excitation of a light beam.