Abstract: The text recognition method includes: acquiring a first high-frequency feature map and a first low-frequency feature map of a target image; performing an M-level convolution process on the first high-frequency feature map and the first low-frequency feature map by M cascaded convolution modules to obtain M pairs of target high-frequency feature map and target low-frequency feature map of the target image, where M is a positive integer; merging the M pairs of target high-frequency feature map and target low-frequency feature map to obtain a target feature map of the target image; determining a probability map and a threshold map of the target image based on the target feature map, and calculating a binarization map of the target image based on the probability map and the threshold map; and determining a text area in the target image based on the binarization map, and recognizing text information in the text area.

Abstract: A display system, a method for processing a trajectory point sequence, a storage medium, and a device are provided. The display system includes a display, a processor, and an input component. The processor is configured to control the display to display an interaction interface; receive a trajectory point sequence input by a user by means of the input component; generate a first content based on the trajectory point sequence and control the display to display the first content; recognize a first identification sequence included in the trajectory point sequence, to determine a first feature identifier corresponding to the first content; determine a cluster relationship of the first content according to the first feature identifier; generate a display region in a preset format according to the cluster relationship; and control the display to display the display region on the interaction interface.

Abstract: Provided is a control method including obtaining a first image; performing face recognition and gesture recognition on the first image; turning on a gesture control function when a first target face is recognized from the first image and a first target gesture is recognized from the first image; and returning to the act of obtaining the first image when the first target face is not recognized from the first image or the first target gesture is not recognized from the first image.

Abstract: The present disclosure provides a method for generating handwriting, belonging to the field of computer application technology. The handwriting generation method includes: acquiring sampling points on a handwritten track, and generating a handwritten track curve based on the sampling points wherein the handwritten track is a track of a stroke segment of a handwritten stroke; determining an arc length ? between interpolation points; determining positions of interpolation points on the handwritten track curve according to the arc length ? between interpolation points, where an arc length between two adjacent interpolation points on the handwritten track curve is the same as the arc length ? between interpolation points; and overlaying a handwriting effect parameter at the positions of the interpolation points to generate the track of the marker pen corresponding to the handwritten track curve.

Abstract: Provided are a display device and a method for displaying a chart, the display device including: a display screen and a control circuit. The control circuit includes a processor and a memory; the memory is used for storing a program executable by the processor; and the processor is used for reading the program in the memory and performing the following steps: identifying writing track information in a display area of the display screen to obtain a data identification result; in response to a user's chart drawing instruction, determining a chart type corresponding to the chart drawing instruction; and in response to the user's chart drawing instruction, drawing, according to the data identification result, a chart of the chart type corresponding to the chart drawing instruction, and displaying the drawn chart in the display area.

Abstract: An object behavior analysis method includes: acquiring multiple frames of images to be processed from a video of a target scene; detecting at least one first-type object in any frame of image to be processed, and determining key point information of any first-type object; determining key point merging information of any first-type object in a present image according to key point information of first-type objects in the present image and a reference image group, the present image being any frame of image to be processed during present processing, and the reference image group including at least one frame of image to be processed before the present image; and determining a behavior analysis result of any first-type object in the present image according to the key point merging information of the first-type object in the present image.

Abstract: A method of training an object detection network, an object detection method and related apparatuses are provided. The training method includes: inputting a training image into a to-be-trained object detection network to obtain detection information of a target object, where the detection information includes a detection class of the target object, a detection position of a detection box of the target object, and a detection position of a landmark of the target object; calculating a total loss function of the to-be-trained object detection network, where the total loss function is calculated according to a loss function of the detection class of the target object, a loss function of the detection position of the detection box of the target object, and a loss function of the detection position of the landmark of the target object; and adjusting a parameter of the to-be-trained object detection network according to the total loss function.

Abstract: A coated substrate comprising a coating extending over at least a part of the substrate. The coating is obtainable from a coating composition comprising (a) a polymeric binder; and (b) a feathering reducing agent comprising a carboxylic acid reactive functional group. The coated portion of the substrate comprises a pretreatment layer. The pretreatment layer is obtainable from a pretreatment composition that comprises a trivalent chromium compound. The invention extends to a method of coating a pre-treated substrate and a coated package, such as a metal can.

Abstract: A coating composition comprising (a) a polyester binder material; and (b) a feathering reducing agent. The feathering reducing agent is selected from (i) an acrylic feathering reducing agent comprising a functional group selected from hydroxyl, epoxide, phosphatized epoxide and/or acid-functional; (ii) a hydroxy-functional polyester feathering reducing agent; (iii) a feathering reducing agent comprising a functional group selected from amine, amide, imine and/or nitrile; (iv) a phosphatized epoxy feathering reducing agent; (v) a phenolic resin feathering reducing agent; and/or (vi) a feathering reducing agent comprising an oxazolyl functional group. The invention extends to a substrate coated with the coating composition and use of the coating composition to reduce feathering.

Abstract: A computer-implemented method is provided. The computer-implemented method includes inputting a low-resolution image into a generator; and generating a high-resolution image using the generator based on the low-resolution image. Generating the high-resolution image includes processing the low-resolution image through a plurality of super-resolution generating units arranged in series in the generator. A respective output from a respective one of the plurality of super-resolution generating units has a respective increased image resolution as compared to a respective input to the respective one of the plurality of super-resolution generating units.

Abstract: An interaction method, an electronic device and a computer storage medium. The interaction method includes: acquiring a trajectory point set input by a user, wherein the trajectory point set at least includes one trajectory point; determining a candidate object according to the trajectory point set, wherein the candidate object includes a first candidate object and a second candidate object; displaying a first selection icon for representing the first candidate object and a second selection icon for representing the second candidate object; and displaying the first candidate object or executing an operation corresponding to the first candidate object in response to the user's selection operation on the first selection icon, and displaying the second candidate object or executing an operation corresponding to the second candidate object in response to the user's selection operation on the second selection icon.

Abstract: A handwriting forming method, a handwriting forming apparatus, and an electronic device are provided. The method includes: determining a group of plurality of trajectory points and obtaining trajectory information of each of the plurality of trajectory points, the trajectory information including a pressure value; establishing a plurality of trajectory units corresponding to the plurality of trajectory points in one-to-one correspondence based on the trajectory information; and connecting the plurality of trajectory units to form a handwriting for displaying. A size of each trajectory unit is determined based on a pressure sensing amount corresponding to the trajectory unit, and the pressure sensing amount is obtained based on the pressure value of a corresponding trajectory point. In the case where the pressure value is less than a first threshold, the pressure sensing amount is a first constant regardless of whether the pressure value of the corresponding trajectory point changes.

Abstract: The present disclosure provides a method for generating an animation figure, a device and a storage medium. The method includes: acquiring an image including at least one target object; acquiring position information about key points of the target object in the image; determining target angle information about lines connecting the key points in accordance with the position information about the key points; and adjusting a predetermined animation figure in accordance with the target angle information. A pose of a target animation figure acquired through adjustment is identical to a pose of the target object in the image, and the target animation figure corresponds to the target object.

Abstract: A handwriting forming method, a handwriting forming apparatus, and an electronic device are provided. The method includes: determining a group of plurality of trajectory points according to a plurality of sampling points on a writing path on a working surface of a touch device, and obtaining trajectory information of each of the plurality of trajectory points, and the trajectory information including a coordinate, a flag bit and a pressure value; establishing a plurality of trajectory units in a raindrop shape corresponding to the plurality of trajectory points based on the trajectory information; determining a connection mode between trajectory units corresponding to adjacent trajectory points based on the trajectory information; and calculating connection points of the trajectory units according to the connection mode, and filling a polygon formed by the connection points of the trajectory units corresponding to the adjacent trajectory points, so as to form a handwriting for displaying.

Abstract: Embodiments of the present disclosure provide a method and an apparatus for searching for an image and a related storage medium. The method includes obtaining reference keypoint data of a reference person in a reference image, and searching, based on the reference keypoint data, a set of candidate images for at least one target image containing at least one candidate person that has a pose similar to the reference person.

Abstract: A handwriting processing method, a handwriting processing device and a non-transitory storage medium. The handwriting processing method includes: acquiring a handwriting point group corresponding to a stroke on a working surface of a touch device, the handwriting point group including a plurality of handwriting points arranged in sequence, and data of each handwriting point in the plurality of handwriting points including a coordinate and an action type, determining a plurality of model patterns corresponding to the plurality of handwriting plurality of model patterns being in one-to-one correspondence with the plurality of handwriting points; and sequentially connecting the plurality of model patterns, to determine a handwriting track for displaying corresponding to the handwriting point group.

Abstract: The present disclosure provides an image processing method, an image processing device, an electronic device and a computer-readable storage medium. The image processing method includes: receiving an input image; and processing the input image through a first generator to acquire an output image with definition higher than the input image. The first generator is acquired through training a to-be-trained generator using at least two discriminators. According to the embodiments of the present disclosure, the first generator for repairing the image is acquired through training with at least two discriminators, so it is able to provide the repaired image with more details, thereby to improve a repair effect.

Abstract: Provided is a control method including obtaining a first image; performing face recognition and gesture recognition on the first image; turning on a gesture control function when a first target face is recognized from the first image and a first target gesture is recognized from the first image; and returning to the act of obtaining the first image when the first target face is not recognized from the first image or the first target gesture is not recognized from the first image.

Abstract: An image denoising method includes: acquiring a first data set and a second data set, where the first data set includes a plurality of first images without noise, the second data set includes a plurality of second images with real noise, contents of each first image and each second image are different; training, by using the first data set and the second data set, a first network to obtain a noise generation model; inputting the first image into the noise generation model, and outputting a third image with simulated noise; where a plurality of third images forms a third data set; training, by using the first data set and the third data set, an image denoising network to obtain an image denoising model. The image denoising model is configured to convert an original image with noise into an output image without noise.

Abstract: The present disclosure provides a photo album management method. The method includes obtaining voice search information from a user, performing intent recognition on the voice search information to obtain an intent recognition result which indicates an intent of the user for a photo album, obtaining a voiceprint feature from the voice search information to determine identity information of the user, sending the intent recognition result and the identity information of the user, and opening the photo album according to the intent recognition result and the identity information.