Abstract: Provided are a method for training a font generation model, a method for establishing a font library, and a device. The method for training a font generation model includes the following steps. A source-domain sample character is input into the font generation model to obtain a first target-domain generated character. The first target-domain generated character is input into a font recognition model to obtain the target adversarial loss of the font generation model. The model parameter of the font generation model is updated according to the target adversarial loss.

Abstract: A meme generation method, an electronic device, and a storage medium are provided. The method includes: determining a plurality of second expression images corresponding to a target face image based on a plurality of first expression images contained in a first meme; generating a second meme corresponding to the target face image based on the plurality of second expression images corresponding to the target face image; wherein, determining an affine transformation parameter between the target face image and an i-th first expression image in the plurality of first expression images according to a corresponding relation between a face key point in the target face image and a face key point in the i-th first expression image; and transforming the target face image based on the affine transformation parameter to obtain an i-th second expression image corresponding to the target face image.

Abstract: There is provided cross-modal feature extraction, retrieval, and model training methods and apparatuses, and a medium, which relates to the field of artificial intelligence (AI) technologies, and specifically to fields of deep learning, image processing, and computer vision technologies. A specific implementation solution involves: acquiring to-be-processed data, the to-be-processed data corresponding to at least two types of first modalities; determining first data of a second modality in the to-be-processed data, the second modality being any of the types of the first modalities; performing semantic entity extraction on the first data to obtain semantic entities; and acquiring semantic coding features of the first data based on the first data and the semantic entities and by using a pre-trained cross-modal feature extraction model.

Abstract: The present application discloses a method and an apparatus for processing map data, which relate to autonomous driving technologies in the field of data processing. The specific implementation is that: a controlling unit inputs initial positioning data collected by a data collecting unit to a data fusing unit to obtain fused positioning data, where the initial positioning data and the fused positioning data are data in a first coordinate system; the controlling unit obtains target positioning data according to the fused positioning data, where the target positioning data is data in a second coordinate system, and the second coordinate system is a coordinate system obtained by offsetting the first coordinate system; and the controlling unit performs a positioning operation on the target positioning data through at least one positioning unit, to determine a position of a vehicle corresponding to an autonomous driving system.

Abstract: A vehicle information detection method, a method for training a detection model, an electronic device and a storage medium are provided, and relates to the technical field of artificial intelligence, in particular to the technical field of computer vision and deep learning. The method includes: performing a first target detection operation based on an image of a target vehicle, to obtain a first detection result for target information of the target vehicle; performing an error detection operation based on the first detection result, to obtain error information; and performing a second target detection operation based on the first detection result and the error information, to obtain a second detection result for the target information.

Abstract: Provided are a streaming media processing method based on inference service, an electronic device, and a storage medium, which relates to the field of artificial intelligence, and in particular, to the field of inference service of artificial intelligence models. The method includes: detecting, in a process of processing a k-th channel of streaming media through an i-th inference service pod, the i-th inference service pod, to obtain a detection result of the i-th inference service pod, i and k being positive integers; determining a replacement object of the i-th inference service pod, in the case where it is determined that the i-th inference service pod is in an abnormal state based on the detection result of the i-th inference service pod; and processing the k-th channel of streaming media through the replacement object of the i-th inference service pod.

Abstract: Provided are a connection establishment method, a server, an accessed node, an access node, and a storage medium. The method includes: receiving a connection establishment request from an accessed node, establishing a connection to the accessed node, and acquiring accessed connection information of the accessed node; receiving an accessed address request from an access node and determining access connection information of the access node; sending the access connection information to the accessed node so that the accessed node opens up a connection channel between the access node and the accessed node according to the access connection information; and sending the accessed connection information of the accessed node to the access node so that the access node establishes a connection to the accessed node according to the accessed connection information in the case where the connection channel is opened up.

Abstract: A method and apparatus for processing an audio and video. The method includes: acquiring a target processing request including a target audio and video data stream; determining a target audio and video pipeline corresponding to the target processing request; the audio and video pipeline being constituted based on a plurality of functional components arranged in a chain structure, and the functional components being uniformly dispatched input data and recovered output data by a preset data stream dispatching module; and calling the target audio and video pipeline to continuously process the target audio and video data stream, and continuously outputting a processed audio and video data stream obtained after processing.

Abstract: A method and apparatus for sending a debugging instruction, an electronic device and a computer readable storage medium are provided. The method may include: after acquiring a debugging instruction sent by an operating terminal, determining a debugged terminal and a first edge communication node corresponding to the debugged terminal according to the debugging instruction, and determining a debugging communication link between the first edge communication node and the debugged terminal, the first edge communication node being determined based on first edge communication node information sent by the debugged terminal, and the first edge communication node information being determined and obtained based on an edge node computing application locally installed on the debugged terminal, and sending an debugging operation included in the debugging instruction to the debugged terminal through the debugging communication link.

Abstract: A text recognition method includes: acquiring an image including text information, the text information including M characters, M being a positive integer greater than 1; performing text recognition on the image to acquire character information about the M characters; recognizing reading direction information about each character in accordance with the character information about the M characters, the reading direction information being used to indicate a next character corresponding to a current character in a semantic reading order; and ranking the M characters in accordance with the reading direction information about the M characters to acquire a text recognition result of the text information.

Abstract: A method for compressing a neural network model includes acquiring a to-be-compressed neural network model. A first bit width, a second bit width and a target thinning rate corresponding to the to-be-compressed neural network model are determined. A target value is obtained according to the first bit width, the second bit width and the target thinning rate. Then the to-be-compressed neural network model is compressed using the target value, the first bit width and the second bit width to obtain a compression result of the to-be-compressed neural network model.

Abstract: The present application discloses a map data updating method, an apparatus, a device and a readable storage medium. The specific implementation solution is: after receiving road information reported by an electronic device, a server obtains multiple sequences according to the road information, and each road information belonging to the same sequence has the same type and location. After that, the server inputs each road information contained in the sequences to a pre-trained neural network model, so that the neural network model outputs a recognition result according to the sequences. The server updates map data according to the recognition result. With such solution, valid road information is recognized by combining context of each road information in the sequences and the neural network technology, and the map data is updated, which achieves the purpose of accurately updating the map data.

Abstract: An image rendering method includes the steps below. A model of an environmental object is rendered to obtain an image of the environmental object in a target perspective. An image of a target object in the target perspective and a model of the target object are determined according to a neural radiance field of the target object. The image of the target object is fused and rendered into the image of the environmental object according to the model of the target object.

Abstract: A human body identification method, an electronic device and a storage medium, related to the technical field of artificial intelligence such as computer vision and deep learning, are provided. The method includes: inputting an image to be identified into a human body detection model, to obtain a plurality of preselected detection boxes; identifying a plurality of key points from each of the preselected detection boxes respectively according to a human body key point detection model, and obtaining a key point score of each of the key points; determining a target detection box from each of the preselected detection boxes, according to a number of the key points whose key point scores meet a key point threshold; and inputting the target detection box into a human body key point classification model, to obtain a human body identification result for the image to be identified.

Abstract: The present disclosure provides a method, an apparatus, a computer device and a computer-readable storage medium for positioning, and relates to the field of autonomous driving. The method obtains point cloud data collected by a LiDAR on a device at a current time; determines, based on the point cloud data and a global map built in a global coordinate system, global positioning information of the device in the global coordinate system at the current time; and determine, based on the point cloud data and a local map built in a local coordinate system, local positioning information of the device in the local coordinate system at the current time. A positioning result of the device at the current time is determined based on at least the global positioning information and the local positioning information. Techniques of the present disclosure can provide an effective and stable positioning service.

Abstract: A method, apparatus, device and storage medium for recognizing a bill image may include: performing text detection on a bill image, and determining an attribute information set and a relationship information set of each text box of at least two text boxes in the bill image; determining a type of the text box and an associated text box that has a structural relationship with the text box based on the attribute information set and the relationship information set of the text box; and extracting structured bill data of the bill image, based on the type of the text box and the associated text box that has the structural relationship with the text box.

Abstract: A method for training generative network, a method for generating near-infrared image and device. The method includes: obtaining a training sample set, in which the set includes near-infrared image samples and visible-light image samples; obtaining an adversarial network to be trained, in which the generative network of the adversarial network is configured to generate each near-infrared image according to an input visible-light image, the discrimination network of the adversarial network is configured to determine whether an input image is real or generated; constructing a first objective function according to a first distance between each generated near-infrared image and the corresponding near-infrared image sample in an image space and a second distance between each generated near-infrared image and the corresponding near-infrared image sample in a feature space; performing an adversarial training on the adversarial network with the set based on optimizing a value of the first objective function.

Abstract: A method, apparatus, and electronic device for processing a video, a medium and a product are presented. An implementation of the method includes: acquiring a target video; selecting, from at least one preset model, a preset model as a target model; determining output data of the target model based on the target video and the target model; reselecting, in response to determining that the output data does not meet a condition corresponding to the target model, another preset model as the target model from the at least one preset model until the output data of the target model meets the condition corresponding to the target model; and determining, based on the output data, a dynamic cover from the target video.

Abstract: The present disclosure provides a method for visual question answering, which relates to fields of computer vision and natural language processing. The method includes: acquiring an input image and an input question; detecting visual information and position information of each of at least one text region in the input image; determining semantic information and attribute information of each of the at least one text region based on the visual information and the position information; determining a global feature of the input image based on the visual information, the position information, the semantic information, and the attribute information; determining a question feature based on the input question; and generating a predicted answer for the input image and the input question based on the global feature and the question feature. The present disclosure further provides a device for visual question answering, a computer device and a medium.