Abstract: Embodiments of the present disclosure relate to a method, a device, and a computer program product for image processing. The method includes generating from an input image a corresponding output image, and determining watermark embedding of a target watermark by an embedding network. The method further includes generating, by the embedding network, a watermarked image corresponding to the output image based on the output image and the watermark embedding, wherein the target watermark is invisible in the watermarked image and can be extracted from the watermarked image by an extraction network corresponding to the embedding network. Without reducing the practicability of an image generation tool, this solution can provide more effective protection for an image generated thereby and the tool itself so as to prevent theft of the image and functions of the tool.

Abstract: Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for training an image processing model. The method in an illustrative embodiment includes: obtaining a folding weight of a folded convolutional layer of a pre-trained generator by performing a folding operation on a plurality of weights of a plurality of convolutional layers of the pre-trained generator. The method further includes: embedding the pre-trained generator into the image processing model. The method further includes: training the image processing model using a plurality of pairs of sample images, wherein at least one pair of sample images of the plurality of pairs of sample images includes a first sample image having a first resolution and a second sample image having a second resolution, and wherein the first resolution is less than the second resolution.

Abstract: Embodiments of the present disclosure relate to a method, a device, and a computer program product for image processing. The method includes: obtaining an encoding feature of a reference image and an encoding feature of an input image of a first resolution, wherein the reference image has a resolution greater than the first resolution. The method further includes: obtaining high-frequency information and low-frequency information on the input image by interpolating the input image; obtaining a first output feature based on the encoding feature of the reference image and the high-frequency information; and obtaining a second output feature based on the encoding feature of the input image and the low-frequency information. The method further includes: generating an output image of a second resolution based on the first output feature and the second output feature, wherein the second resolution is greater than the first resolution.

Abstract: Embodiments of the present disclosure provide a method, an electronic device, and a computer program product for switching a chat object. The method includes: using a robot as a chat object to converse with a user; outputting a target sentiment score based on current discourse, historical discourse, and predicted discourse associated with the user; and switching the chat object conversing with the user from the robot to human in response to the target sentiment score indicating that the user's sentiment is a negative sentiment. Implementation of this method can improve human-machine interactivity and communication efficiency, and ensure the sustainability of communication between a chat system and a user, thus improving the user's satisfaction with a service system.

Abstract: Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for generating a three-dimensional (3D) model. The method includes generating two-dimensional (2D) features of a 2D image on the basis of performing feature extraction on the 2D image. The method further includes generating, on the basis of the 2D features by using a symmetric conversion module, 2D features subjected to symmetric conversion, wherein the symmetric conversion module is configured to generate 2D features of an object in the 2D image in different viewing angles. The method further includes generating a 3D model of the 2D image on the basis of the 2D features subjected to symmetric conversion. By means of embodiments of the present disclosure, the generation of the 3D model can be supervised by using the symmetric conversion module, thus achieving generation of a 3D model with a better effect.

Abstract: Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for generating a target object. The method includes upsampling a first feature of an input point cloud. The method further includes determining a reference feature of a reference point cloud having a predetermined shape. The method further includes determining a second feature based on the upsampled first feature and the reference feature. The method further includes generating a three-dimensional target object based on the second feature and the input point cloud, wherein the target object has a fitted curved surface, and the target object has a greater number of points than the input point cloud. With embodiments of the present disclosure, a point cloud of the target object can be made denser, and the number of points can be increased, thereby achieving a more accurate three-dimensional reconstruction of the target object.

Abstract: A method in an illustrative embodiment includes determining, by a first deep learning model on a storage device, whether a fault is caused by an environmental factor. The method further comprises: determining, by a second deep learning model on the storage device in response to determining that the fault is caused by the environmental factor, whether the fault can be solved locally in the storage device. The method further comprises: processing the fault according to a knowledge base in response to determining that the fault can be solved locally in the storage device. The method further comprises: sending the fault to a third deep learning model on a cloud device in response to determining that the fault cannot be solved locally in the storage device, the first deep learning model and the second deep learning model being obtained by model distillation of the third deep learning model.

Abstract: A method, an electronic device, and a computer program product for image processing are disclosed herein. The method in an illustrative embodiment comprises collecting a plurality of audio signals from a plurality of sensors, and combining the plurality of audio signals to generate synthetic audio, wherein the plurality of sensors are located at different locations in the same space, and the plurality of audio signals each have a location identifier. The method further comprises analyzing the synthetic audio to find an audio subset in the plurality of audio signals, and extracting first image features of a first image of the space. The method further comprises modifying the first image features based on the location identifiers of the audio signals in the audio subset to generate second image features, and generating a second image of the space based on the second image features.

Abstract: Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for generating segmented images. The method includes inputting a to-be-processed image to a first generator, where the first generator is obtained from training by using a first discriminator for the first generator, a second generator, and a second discriminator for the second generator, and the first generator, the first discriminator, the second generator, and the second discriminator form a cycle generative adversarial network. The method further includes acquiring segmented images of the to-be-processed image that are generated by the first generator. By means of the method, segmented images with improved accuracy can be obtained, and training resources required in a process of training a generator can be greatly reduced, thereby improving the training speed and saving training resources.

Abstract: A method in an illustrative embodiment includes determining a first loss function for a first sub-model of a speech generation model based on a plurality of feature vectors associated with training image information, training audio information, and training text information used to train the speech generation model. The method may further include determining a second loss function for a second sub-model and a third loss function for a third sub-model of the speech generation model based on the plurality of feature vectors that have been processed. In addition, the method may further include determining a fourth loss function for a fourth sub-model of the speech generation model based on the processed plurality of feature vectors. The method may further include updating parameters of the speech generation model based on the first loss function, the second loss function, the third loss function, and the fourth loss function.

Abstract: Methods, electronic devices and computer program products for model processing are disclosed in embodiments herein. A method in an illustrative embodiment includes encoding first data of a point cloud model to obtain a first matrix, and encoding second data of the point cloud model to obtain a second matrix, where the first data and the second data are data of the point cloud model acquired from different angles. The method further includes respectively decomposing, by an equivariant encoder, the first matrix and the second matrix into a first equivariant matrix and a second equivariant matrix, and respectively decomposing, by an invariant encoder, the first matrix and the second matrix into a first invariant matrix and a second invariant matrix. The method further includes training the equivariant encoder and the invariant encoder based on the first equivariant matrix, the second equivariant matrix, the first invariant matrix, and the second invariant matrix.

Abstract: A method in an illustrative embodiment includes generating, based on a group of utterances in a dialog, multiple representations corresponding to the group of utterances, the multiple representations including a first representation associated with the group of utterances, a second representation associated with the group of utterances and a group of utterances following the group of utterances, and a third representation associated with the group of utterances and at least two groups of utterances in a context of the group of utterances. The method also includes obtaining, based on the multiple representations, multiple reference recognition results corresponding to the group of utterances, each of the multiple reference recognition results indicating whether the dialog needs to be transferred to a target object. The method further includes determining, based on the multiple reference recognition results, a target recognition result indicating whether the dialog needs to be transferred to the target object.

Abstract: Embodiments of the present disclosure relate to a method, a device, and a computer program product for rendering. The method comprises modeling a three-dimensional (3D) scene as a four-dimensional (4D) tensor, wherein a first dimension, a second dimension, and a third dimension in the 4D tensor correspond to an X-Y-Z coordinate axis in the 3D scene, a fourth dimension represents a channel dimension corresponding to an encoded feature, and the encoded feature is obtained by encoding the 3D scene with an encoder corresponding to a decoder. The method further comprises performing interpolation on the 4D tensor to obtain streaming scene information associated with the 3D scene. The method further comprises rendering the 3D scene on the basis of the streaming scene information. According to the method, a unified and flexible architecture can be provided for edge-cloud 3D collaboration, and a 3D scene representation can be processed effectively and efficiently.

Abstract: A method, an electronic device, and a computer program product for generating a three-dimensional scene are provided in embodiments of the present disclosure. The method may include obtaining source image features from a plurality of two-dimensional source images associated with the three-dimensional scene to be generated. The method may further include obtaining editing features from an editing instruction input by a user for the three-dimensional scene, each of the editing features respectively forming a feature pair with each of the source image features. Furthermore, the method may include updating the source image features by maximizing a correlation coefficient of each of the feature pairs, and generating the three-dimensional scene based at least on the updated source image features. Embodiments of the present disclosure can realize arbitrary editing of a three-dimensional scene, thus enhancing the experience of human-computer interaction.

Abstract: A method includes generating a feature map of a second resolution for an input image of a first resolution, the first resolution being less than the second resolution. The method further includes generating a feature map of the second resolution and a feature map of the first resolution through a first residual-based dense back projection (RDBP) network. In addition, the method further includes generating a feature map of the second resolution and a feature map of the first resolution through a second RDBP network, the first RDBP network and the second RDBP network respectively including a downsampling back projection layer, an upsampling back projection layer, and an attention layer. The method further includes generating an output image of the second resolution based on the feature maps of the second resolution and the feature maps of the first resolution generated through the first RDBP network and the second RDBP network.

Abstract: Embodiments of the present disclosure provide a method, an electronic device, and a computer program product for detecting model performance. The method may include acquiring a prediction result of an input feature using a target model to determine a confidence of the prediction result. The method may further include reconstructing the input feature using a self-coding model to determine a reconstruction error, the reconstruction error being a difference between the input feature before being reconstructed by the self-coding model and the input feature after being reconstructed by the self-coding model. In addition, the method may include determining a detection result of the target model at least based on a comparison between the confidence and a first threshold and a comparison between the reconstruction error and a second threshold.

Abstract: Embodiments of the present disclosure relate to a method, an electronic device, and a computer program product for allocating a workload. The method includes acquiring a first state graph of a plurality of devices that run a first workload at a first time point. The method further includes updating the first state graph based on a comparison between an active value of at least one node in the first state graph and a predetermined threshold. The method further includes determining a first load state of the plurality of devices at the first time point based on an updated first state graph. The method further includes allocating a second workload to the plurality of devices at a second time point based on the first load state, where active values of nodes in the updated first state graph are greater than the predetermined threshold.

Abstract: Embodiments of the present disclosure relate to a method, a device, and a computer program product for detecting stickers. The method in an illustrative embodiment includes sending, at a far edge, a sticker image of an object device to a near edge, wherein a sticker is attached to a surface of the object device, and the sticker image comprises an image of the sticker. The method further includes querying, at the far edge, label information of the object device. The method further includes receiving, at the far edge from the near edge, a classification result from classifying the sticker image. The method further includes comparing, at the far edge, the classification result with the label information. The method further includes sending, from the far edge, the sticker image to a cloud via the near edge in response to determining that the classification result is not matched with the label information.

Abstract: Embodiments of the present disclosure relate to a method, a device, and a computer program product for processing data. The method includes determining, based on acquired text, character features for a group of characters in the text and text features for the text. The method further includes determining initial visual features for the text based on the text features. The method further includes determining target visual features for the text based on the initial visual features, the character features, and the text features. The method further includes generating a target image corresponding to the text based on the target visual features. Through the method, the accuracy of conversion between text and an image is improved, the data processing efficiency is improved, and the data compression efficiency is further improved.

Abstract: Embodiments of the present disclosure provide a method, an electronic device, and a medium for bug classification. The method includes determining, based on description information of a bug generated during product testing, classification information of the bug through at least one trained computing model; presenting the classification information of the bug; determining, based on user interaction for the presented classification information, whether performance of the at least one computing model satisfies a predetermined condition; and determining that the at least one computing model needs to be retrained in response to determining that the performance of the at least one computing model does not satisfy the predetermined condition. In this way, automatic classification of the bug is realized, and the computing model can be dynamically adjusted by retraining, so as to ensure accuracy of the automatic classification and improve efficiency of bug processing.