Abstract: Example aspects include techniques for implementing a high-frequency attention network for single image super-resolution. These techniques may include extracting a plurality of features from an original image input into a CNN to generate a feature map, and restoring one or more high-frequency details of the original image via an efficient residual block (ERB) and a high-frequency attention block (HFAB) configured to assign a scaling factor to one or more high-frequency areas. In addition, the techniques may include generating reconstruction input information by performing an element-wise operation on the one or more high-frequency details and cross-connection information from the feature map and performing, by the CNN, an enhancement operation on the reconstruction input information to generate an enhanced image.

Abstract: Described are examples for detecting objects in an image on a device including setting, based on a condition, a number of sparse proposals to use in performing object detection in the image, performing object detection in the image based on providing the sparse proposals as input to an object detection process to infer object location and classification of one or more objects in the image, and indicating, to an application and based on an output of the object detection process, the object location and classification of the one or more objects.

Abstract: Embodiments of the disclosure provide a method, device, storage medium and program product for music screening. The method includes: obtaining at least one image and at least one piece of to-be-selected music; determining an analysis result of the at least one image corresponding to an image classification tag based on N predetermined image classification tags, N being an integer greater than or equal to 1; determining attribute information for each piece of to-be-selected music based on the at least one image and the at least one piece of to-be-selected music; determining target music that matches the at least one image among the at least one piece of to-be-selected music based on the analysis result and the attribute information of each piece of to-be-selected music.

Abstract: Embodiments of the disclosure provide a method and a device for processing a portrait image, the method includes: acquiring a to-be-processed portrait image; inputting the to-be-processed portrait image into an image processing model, and acquiring a head smear image output by the image processing model, where the image processing model is configured to smear a hair area of a portrait located above a preset boundary in the portrait image, and the image processing model is generated by training a sample data set of a sample portrait image and a sample head smear image corresponding to the sample portrait image; rendering the head smear image with a head effect material to obtain a portrait image added with an effect; and displaying the portrait image added with the effect.

Abstract: Example aspects include techniques for implementing a high-frequency attention network for single image super-resolution. These techniques may include extracting a plurality of features from an original image input into a CNN to generate a feature map, and restoring one or more high-frequency details of the original image via an efficient residual block (ERB) and a high-frequency attention block (HFAB) configured to assign a scaling factor to one or more high-frequency areas. In addition, the techniques may include generating reconstruction input information by performing an element-wise operation on the one or more high-frequency details and cross-connection information from the feature map and performing, by the CNN, an enhancement operation on the reconstruction input information to generate an enhanced image.

Abstract: Described are examples for detecting objects in an image on a device including setting, based on a condition, a number of sparse proposals to use in performing object detection in the image, performing object detection in the image based on providing the sparse proposals as input to an object detection process to infer object location and classification of one or more objects in the image, and indicating, to an application and based on an output of the object detection process, the object location and classification of the one or more objects.

Abstract: Systems and methods are directed to inpainting video. More specifically, initial video data including a sequence of image frames containing missing or corrupted pixel information may be received. Optical flow displacement values and optical flow validity masks may be generated for neighboring image frames of initial video data. Image features from image feature maps of one or more neighboring image frames may be warp-shifted to image feature maps of a current image frame using the optical flow displacement values and warp-shifted image features from the feature maps of the one or more neighboring image frames may be selected based on one or more of the optical flow validity masks. A sequence of complete image frames may be generated based on the selected warp-shifted image features from the feature maps of the one or more neighboring image frames and image features from the image feature maps of the current image frame.

Abstract: Systems and method directed to performing video object segmentation are provided. In examples, video data representing a sequence of image frames and video data representing an object mask may be received at a video object segmentation server. Image features may be generated based on a first image frame of the sequence of image frames, image features may be generated based on a second image frame of the sequence of image frames; and object features may be generated based on the object mask. A transform matrix may be computed based on the image features of the first image frame and image features of the second image frame; the transform matrix may be applied to the object features resulting in transformed object features. A predicted object mask associated with the second image frame may be obtained by decoding the transformed object features.

Abstract: Systems and methods are directed to inpainting video. More specifically, initial video data including a sequence of image frames containing missing or corrupted pixel information may be received. Optical flow displacement values and optical flow validity masks may be generated for neighboring image frames of initial video data. Image features from image feature maps of one or more neighboring image frames may be warp-shifted to image feature maps of a current image frame using the optical flow displacement values and warp-shifted image features from the feature maps of the one or more neighboring image frames may be selected based on one or more of the optical flow validity masks. A sequence of complete image frames may be generated based on the selected warp-shifted image features from the feature maps of the one or more neighboring image frames and image features from the image feature maps of the current image frame.

Abstract: Systems and method directed to performing video object segmentation are provided. In examples, video data representing a sequence of image frames and video data representing an object mask may be received at a video object segmentation server. Image features may be generated based on a first image frame of the sequence of image frames, image features may be generated based on a second image frame of the sequence of image frames; and object features may be generated based on the object mask. A transform matrix may be computed based on the image features of the first image frame and image features of the second image frame; the transform matrix may be applied to the object features resulting in transformed object features. A predicted object mask associated with the second image frame may be obtained by decoding the transformed object features.

Abstract: A functional regenerated cellulose fiber includes a graphene structure and non-carbon non-oxygen elements. The non-carbon non-oxygen elements includes elements of Fe, Si, and Al. The elements of Fe, Si, and Al account for 0.018 wt % to 0.8 wt % of the regenerated cellulose fiber.

Abstract: A method, computer program product, and computer system for predicting responses to at least one conversational phrase. At least one conversational phrase may be received. A first probability for a subset of candidate responses of a plurality of candidate responses may be determined based upon, at least in part, context associated with the at least one conversational phrase, the at least one conversational phrase, and each context associated with the plurality of candidate responses. A second probability for the subset of candidate responses may be determined based upon, at least in part, the subset of candidate responses, the at least one conversational phrase, and the context associated with the at least one conversational phrase. At least one candidate response for the at least one conversational phrase may be determined based upon, at least in part, the first probability and the second probability.

Abstract: A method, computer program product, and computer system for identifying, by a computing device, a model for predicting conversational phrases for a communication between at least a first user and a second user. The model may be trained based upon, at least in part, an attribute associated with the second user. At least one conversational phrase may be predicted for the communication between the first user and the second user. The at least one conversational phrase may be provided to the second user as an optional phrase to be sent to the first user.

Abstract: Disclosed is a method and an apparatus for recognizing speech, and the method comprises: separating an input audio signal into at least two separated signals; generating a denoised signal at a current frame; performing a preliminary recognition on each interesting signal at the current frame; and performing a recognition decision according to a recognition score of each interesting signal at the current frame. The method and apparatus of the present disclosure deeply integrate an array signal processing and a speech recognition and use multiway recognitions such that a good recognition rate may be obtained even in a case of a low signal-to-noise ratio.

Abstract: A graphene adsorbing material, a preparation method therefor and application thereof, and a cigarette filter tip and a cigarette are provided. The graphene adsorbing material comprises a fiber carrier, graphene and doped elements wherein the graphene and the doped elements are loaded on the fiber carrier and the doped elements comprise at least one of Al, Si and Fe. The graphene adsorbing material is obtained by contacting the fibers with the graphene material to load the graphene material on the fibers. By introducing the graphene and the doped elements in the fiber carrier, harmful substances with a type of fused aromatic hydrocarbons such as benzopyrene in smoke can be specifically adsorbed, filtered and removed; while nicotine and the other harmless substances are retained, and thus the smoking taste of a cigarette is not affected.

Abstract: Graphene-containing latex. The latex contains a graphene structure and non-carbon, non-oxygen and non-hydrogen elements. The provided latex is enabled to have multiple performances such as far-infrared performance, antibacterial performance and bacteriostasis by selecting a specific carbon nanostructure compound and by means of collocation and combination of a graphene structure and elements such as Fe, Si and Al; and furthermore, higher far-infrared effect and antibacterial effect can be achieved by controlling a specific addition proportion. The experimental results prove that the far-infrared performance of the latex can reach a maximum value of 0.93, and the bacteriostasis can reach a maximum value of 99%.

Abstract: The present invention relates to a composite having a carbon nanostructure, comprising graphene, amorphous carbon and a non-carbon non-oxygen element, wherein the non-carbon non-oxygen element is in an amount of 0.5 wt %-6 wt % of the composite. The present invention discloses controlling the content of the non-carbon non-oxygen element in the composite to obtain excellent far-infrared effect and antibacterial and bacteriostatic effects, wherein the normal emissivity in the far-infrared performance reaches 0.85 or more, and the antibacterial rate reaches 95% or more. The composite having a carbon nanostructure of the present invention is applied to macromolecular materials to modify macromolecular materials under the circumstance that the addition amount is relatively low. The composite having a carbon nanostructure can achieve notable far-infrared performance and antibacterial and bactericidal performances without any pre-modification and activation treatment.

Abstract: A polyester material including a composite having a carbon nanostructure, which comprises carbon element, from 0.5 to 4 wt % of a first non-carbon non-oxygen element substance, and from 0 to 4 wt %, of a second non-carbon non-oxygen element. The first non-carbon non-oxygen element is selected from the group consisting of P, Si, Ca, Al and Na; the second non-carbon non-oxygen element is any one selected from the group consisting of Fe, Ni, Mn, K, Mg, Cr, S or Co, or a combination of at least two selected therefrom. The G peak and D peak of the carbon element in the Raman spectrum has a peak height ratio of 1-20 in the composite having a carbon nanostructure.

Abstract: A method, computer program product, and computer system for identifying, by a computing device, a model for predicting conversational phrases for a communication between at least a first user and a second user. The model may be trained based upon, at least in part, an attribute associated with the second user. At least one conversational phrase may be predicted for the communication between the first user and the second user. The at least one conversational phrase may be provided to the second user as an optional phrase to be sent to the first user.

Abstract: Disclosed is a method and an apparatus for recognizing speech, and the method comprises: separating an input audio signal into at least two separated signals; generating a denoised signal at a current frame; performing a preliminary recognition on each interesting signal at the current frame; and performing a recognition decision according to a recognition score of each interesting signal at the current frame. The method and apparatus of the present disclosure deeply integrate an array signal processing and a speech recognition and use multiway recognitions such that a good recognition rate may be obtained even in a case of a low signal-to-noise ratio.