Abstract: This application provides an image processing method and apparatus. The method includes acquiring an identity feature of a source image and an initial attribute feature of at least one measure of a target image in response to receiving a face change request; inputting the identity feature and the initial attribute feature of the at least one measure into a face change model; iteratively performing feature fusion on the identity feature and the initial attribute feature of the at least one measure by using the face change model to obtain a fusion feature; and generating a target face change image based on the fusion feature by using the face change model, and outputting the target face change image, a face in the target face change image being fused with an identity feature of the source face and a target attribute feature of the target face.

Abstract: Embodiments of this application provide a peer-to-peer network scheduling method. The method includes: obtaining a fragment file list, where the fragment file list includes information about a plurality of fragment files, each fragment file includes one or more segments, and each segment corresponds to one segment number; downloading a missing segment in a target fragment file from another client device based on a downloading progress of downloading each segment of the target fragment file by each client device; and when a player requests the target fragment file, stopping downloading the missing segment in the target fragment file from the another client device, and downloading the missing segment in the target fragment file from a CDN server. According to the embodiments of this application, a computation amount and resource consumption of a Tracker server can be reduced, and data transmission efficiency of a P2P network can be improved.

Abstract: The present invention discloses a container equipment for enclosed transportation and heterotopic and aerobic stabilization of reserved garbage, which includes a container body and a functional lining provided on an inner wall of the container body, wherein the functional lining is provided with a circulating air low-temperature evaporation system, a water distribution and drainage system and a heating and insulating system; the circulating air low-temperature evaporation system includes an air inlet manifold, an air distribution perforated pipe, a base plate water discharge and air distribution groove and a top plate water distribution and air guide groove, an air extraction perforated pipe, a fan, an air outlet pipe and a quicklime dehydration and deodorization system; the water distribution and drainage system includes a percolate feeding pipe, a fiber capillary water distribution pipe, the top plate water distribution and air guide groove, a side drain, a main drain and a percolate discharge pipe; the hea

Abstract: An image processing method includes performing additional image feature extraction on a training source face image to obtain a source additional image feature, performing identity feature extraction on the training source face image to obtain a source identity feature, inputting a training template face image into an encoder in a to-be-trained face swapping model to obtain a face attribute feature, inputting the source additional image feature, the source identity feature, and the face attribute feature into a decoder in the face swapping model for decoding to obtain a decoded face image, obtaining a target model loss value based on an additional image difference between the decoded face image and a comparative face image, and adjusting the model parameters of the encoder and the decoder based on the target model loss value to obtain the trained face swapping model.

Abstract: A method for training a face swap model includes concatenating an expression feature of a template image and an identity feature of a source face image to obtain a combined feature; performing encoding based on the source face image and the template image to obtain an encoding feature required for face swap; fusing the encoding feature and the combined feature to obtain a fused feature; performing decoding based on the fused feature to obtain a swapped face image; respectively predicting image attribute discrimination results of the swapped face image and a reference image by using a discriminator network of the face swap model; and calculating a difference between an expression feature of the swapped face image and the template image, calculating a difference between an identity feature of the swapped face image and the source face image, and updating the generator network and the discriminator network.

Abstract: Embodiments of this application provides a peer-to-peer network scheduling method. The method includes: obtaining a fragment file list, where the fragment file list includes information about a plurality of fragment files, each fragment file includes one or more segments, and each segment corresponds to one segment number; downloading one or more target segments corresponding to one or more target segment numbers from a CDN server based on the fragment file list and the one or more target segment numbers, where the one or more target segment numbers are located in a target segment list; adjusting the target segment list based on information about data obtained from and/or sent to another client device, to obtain an adjusted target segment list; and downloading one or more target segments corresponding to one or more target segment numbers in the adjusted target segment list from the CDN server.

Abstract: An image synthesis method includes: obtaining a source image including source image identity information for synthesis and a template image including template image background information for synthesis; performing a target synthesis operation on the source image and the template image to obtain an initial synthesized image, the initial synthesized image including the source image identity information, the template image background information, and a partial region to be corrected; performing a target correction operation on the source image, the template image, and the initial synthesized image to obtain a target residual image; and synthesizing the initial synthesized image with the target residual image to generate a target synthesized image.

Abstract: A model training method includes obtaining a training sample set including a triplet training sample that includes a source image, a template image, and a true value image, performing face swapping on the source image and the template image through a first image processing model having a re-parameterization structure to obtain a face-swapped image, obtaining a second image processing model corresponding to the first image processing model and being a pre-trained image processing model, calculating a fusion loss function of the first image processing model according to the second image processing model, the first face-swapped image, and the true value image, training the first image processing model according to the fusion loss function, and determining a model parameter of the first image processing model in response to a training convergence condition of the first image processing model being reached.

Abstract: An image generation method, performed by an electronic device includes obtaining an original face image frame, audio driving information, and emotion driving information, performing spatial feature extraction on the original face image frame to obtain an original face spatial feature corresponding to the original face image frame; performing feature interaction processing on the audio driving information and the emotion driving information to obtain a face local pose feature of the to-be-adjusted object issuing the voice content with the target emotion; and performing, based on the original face spatial feature and the face local pose feature, face reconstruction processing on the to-be-adjusted object to generate a target face image frame.

Abstract: The processing method comprises: acquiring a first image set corresponding to a product category; labelling at least one region of interest for each image in the first image set, wherein each region of interest is used for representing an object; acquiring a scanning window; using the scanning window to scan the regions of interest, acquiring scanning results, and placing the scanning results into a sample training set; establishing an illustration frame acquisition model by taking the first image set as input and taking the sample training set as output; acquiring an image to be processed, and using the illustration frame acquisition model to acquire a target webpage illustration frame corresponding to the image to be processed; and cutting, according to the target webpage illustration frame, the image to be processed so as to obtain a target webpage illustration.

Abstract: An image processing method including obtaining a fake template sample group comprising a first source image, a real labeled image, and a fake template image, inputting the fake template image into an identity swapping model to obtain a first identity swapping image of the fake template image, obtaining a fake labeled sample group comprising a second source image, a real template image, and a fake labeled image, the fake labeled image being based on identity swapping processing of the real template image, inputting the real template image into the identity swapping model to obtain a second identity swapping image of the real template image, and training the identity swapping model based on the fake template sample group, the first identity swapping image, the fake labeled sample group, and the second identity swapping image to generate a trained identity swapping model.

Abstract: This application discloses a method for generating an image processing model performed by a computer device. The method includes: performing training by using a first source image sample, a first template image sample, and a first standard synthesized image, to obtain a first parameter adjustment model, and combining the first parameter adjustment model and a first resolution update layer into a first update model; adjusting the first update model into a second parameter adjustment model using a second source image sample and a second template image sample and a second standard synthesized image; combining the second parameter adjustment model and a second resolution update layer into a second update model; and adjusting the second update model into a target image fusion model using a third source image sample, a third template image sample and a third standard synthesized image.

Abstract: The present invention discloses a container equipment for enclosed transportation and heterotopic and aerobic stabilization of reserved garbage, which includes a container body and a functional lining provided on an inner wall of the container body, wherein the functional lining is provided with a circulating air low-temperature evaporation system, a water distribution and drainage system and a heating and insulating system; the circulating air low-temperature evaporation system includes an air inlet manifold, an air distribution perforated pipe, a base plate water discharge and air distribution groove and a top plate water distribution and air guide groove, an air extraction perforated pipe, a fan, an air outlet pipe and a quicklime dehydration and deodorization system; the water distribution and drainage system includes a percolate feeding pipe, a fiber capillary water distribution pipe, the top plate water distribution and air guide groove, a side drain, a main drain and a percolate discharge pipe; the hea

Abstract: Embodiments of this application provides a peer-to-peer network scheduling method. The method includes: obtaining a fragment file list, where the fragment file list includes information about a plurality of fragment files, each fragment file includes one or more segments, and each segment corresponds to one segment number; downloading one or more target segments corresponding to one or more target segment numbers from a CDN server based on the fragment file list and the one or more target segment numbers, where the one or more target segment numbers are located in a target segment list; adjusting the target segment list based on information about data obtained from and/or sent to another client device, to obtain an adjusted target segment list; and downloading one or more target segments corresponding to one or more target segment numbers in the adjusted target segment list from the CDN server.

Abstract: Embodiments of this application provide a peer-to-peer network scheduling method. The method includes: obtaining a fragment file list, where the fragment file list includes information about a plurality of fragment files, each fragment file includes one or more segments, and each segment corresponds to one segment number; downloading a missing segment in a target fragment file from another client device based on a downloading progress of downloading each segment of the target fragment file by each client device; and when a player requests the target fragment file, stopping downloading the missing segment in the target fragment file from the another client device, and downloading the missing segment in the target fragment file from a CDN server. According to the embodiments of this application, a computation amount and resource consumption of a Tracker server can be reduced, and data transmission efficiency of a P2P network can be improved.

Abstract: Disclosed in embodiments of the present disclosure are a method and apparatus for virtual fitting. A specific implementation of the method comprises: receiving a fitting request comprising a model picture and a user image; performing human body positioning and surface coordinate analysis on the model picture and the user image respectively; performing clothing segmentation on the model picture and the user image respectively; on the basis of the clothing segmentation result and the surface coordinate analysis result, covering the pixels corresponding to a piece of clothing in the model picture to corresponding positions in the user image to obtain a synthesized image and information to be completed; and inputting the synthesized image, the positioning result of the user image, and said information into a pre-trained image completion network to obtain a completed image.

Abstract: The present disclosure describes techniques for generating live streaming video data and playing live streaming video. The techniques comprise obtaining live broadcasting video data; identifying a target object in a video frame; determining a target area based on the target object in the video frame; encoding data indicative of the target area based on a predetermined format for supplementary enhancement information to obtain encoded target area data; and adding the encoded target area data to a data packet corresponding to the video frame, wherein the data packet is encoded based on a predetermined video data format, and wherein the predetermined video data format and the predetermined format for supplementary enhancement information are mutually compatible data encoding formats.

Abstract: This application provides an image processing method and apparatus. The method includes acquiring an identity feature of a source image and an initial attribute feature of at least one measure of a target image in response to receiving a face change request; inputting the identity feature and the initial attribute feature of the at least one measure into a face change model; iteratively performing feature fusion on the identity feature and the initial attribute feature of the at least one measure by using the face change model to obtain a fusion feature; and generating a target face change image based on the fusion feature by using the face change model, and outputting the target face change image, a face in the target face change image being fused with an identity feature of the source face and a target attribute feature of the target face.

Abstract: This application provides an image processing method performed by a computer device. The image processing method includes: receiving a face swapping request; acquiring an attribute parameter of the image, an attribute parameter of the target face, and a face feature of the target face, the attribute parameter of the image indicating a three-dimensional attribute of the face in the image; determining a target attribute parameter based on the attribute parameter of the image and the attribute parameter of the target face; determining a target comprehensive feature based on the target attribute parameter and the face feature of the target face; encoding the image to obtain an image encoding feature of the image; migrating the target comprehensive feature to the image encoding feature of the image by normalization to obtain a fusion encoding feature; and decoding the fusion encoding feature to obtain a target face-swapped image including a fusion face.

Abstract: An de-identification method includes: performing sampling on a target space of a projector in a target network model to obtain N virtual identity vectors, N being a positive integer; extracting, by using an attribute extractor in the target network model, M attribute vectors from a target image, M being a positive integer; and generating, by using a generator in the target network model, a de-identified image of the target image based on the N virtual identity vectors and the M attribute vectors.