Abstract: The present disclosure relates to a method for automatically creating a cartoon image based on an input sentence, performed by a device, the method including: recognizing a sentence; identifying word(s) included in the sentence; identifying a type of punctuation mark located at a start point and an end point of the sentence; determining the sentence as one of a general expression sentence, a dialogue expression sentence, and an emotional expression sentence based on the type of the identified punctuation mark; automatically creating a cartoon image based on a word included in the general expression sentence; understanding a subject of the dialogue expression sentence or the emotional expression sentence; and inserting the dialogue expression sentence or the emotional expression sentence in a form of a speech bubble on a character corresponding to the understood subject among at least one character in the cartoon image.

Abstract: Systems and methods are provided for pitch determination. An example method includes obtaining an image depicting a structure, the image being captured via a user device positioned proximate to the structure. The image is segmented to identify, at least, a roof facet of the structure. An eave vector and a rake vector which are associated with the roof facet are determined. A normal vector of the roof facet is calculated based on the eave vector and the rake vector, and compared to a vector indicating a vertical direction such as gravity. The angle made out by the normal and a gravity vector may be utilized to calculate the pitch of the roof facet.

Abstract: A system comprising: a user device, comprising: sensors configured to sense data related to a physical environment of the user device, displays; hardware processors; and a non-transitory machine-readable storage medium encoded with instructions executable by the hardware processors to: place a virtual object in a 3D scene displayed by the second user device, determine a pose of the user device with respect to the physical location in the physical environment of the user device, and generate an image of virtual content based on the pose of the user device with respect to the placed virtual object, wherein the image of the virtual content is projected by the one or more displays of the user device in a predetermined location relative to the physical location in the physical environment of the user device.

Abstract: Various examples of cross-application systems and methods for authoring, transferring, and evaluating rigging control systems for virtual characters are disclosed. A first application, which implements a first rigging control protocol, can provide an input associated with a request for a behavior from the rig for the virtual character. The input can be converted to be compatible with a second rigging control protocol that is different from the first rigging control protocol. One or more control systems can be evaluated based on the input to determine an output to provide the requested behavior from the virtual character rig. The one or more control systems can be defined according to the second rigging control protocol. The output can be converted to be compatible with the first rigging control protocol and provided to the first application to manipulate the virtual character according to the requested behavior.

Abstract: Provided are a video generation method and apparatus, a device and a storage medium, relating to the field of artificial intelligence and, in particular, to the fields of computer vision and deep learning. The method includes changing a character emotion of an original character image according to a character emotion feature of a to-be-generated video to obtain a target character image; and driving the target character image by use of a character driving network and based on a speech segment to obtain the to-be-generated video.

Abstract: Systems and methods herein describe novel motion representations for animating articulated objects consisting of distinct parts. The described systems and method access source image data, identify driving image data to modify image feature data in the source image sequence data, generate, using an image transformation neural network, modified source image data comprising a plurality of modified source images depicting modified versions of the image feature data, the image transformation neural network being trained to identify, for each image in the source image data, a driving image from the driving image data, the identified driving image being implemented by the image transformation neural network to modify a corresponding source image in the source image data using motion estimation differences between the identified driving image and the corresponding source image, and stores the modified source image data.

Abstract: A manifold voxel mesh or surface mesh is manufacturable by carving a single block of material and a non-manifold mesh is not manufacturable. Conventional techniques for constructing or extracting a surface mesh from an input point cloud often produce a non-manifold voxel mesh. Similarly, extracting a surface mesh from a voxel mesh that includes non-manifold geometry produces a surface mesh that includes non-manifold geometry. To ensure that the surface mesh includes only manifold geometry, locations of the non-manifold geometry in the voxel mesh are detected and converted into manifold geometry. The result is a manifold voxel mesh from which a manifold surface mesh of the object may be extracted.

Abstract: Example methods for generating an animated character in dance poses to music may include generating, by at least one processor, a music input signal based on an acoustic signal associated with the music, and receiving, by the at least one processor, a model output signal from an encoding neural network. A current generated pose data is generated using a decoding neural network, the current generated pose data being based on previous generated pose data of a previous generated pose, the music input signal, and the model output signal. An animated character is generated based on a current generated pose data; and the animated character caused to be displayed by a display device.

Abstract: Disclosed are an imaging method and device. The method includes: acquiring, at a specified imaging time, a pulse sequence in a time period before the specified imaging time, with regard to each pixel of a plurality of pixels; calculating a pixel value of the pixel according to the pulse sequence; and obtaining an image at the specified imaging time according to a space arrangement of the pixels, in accordance with pixel values of the plurality of pixels at the specified imaging time.

Abstract: Systems and methods herein describe novel motion representations for animating articulated objects consisting of distinct parts. The described systems and method access source image data, identify driving image data to modify image feature data in the source image sequence data, generate, using an image transformation neural network, modified source image data comprising a plurality of modified source images depicting modified versions of the image feature data, the image transformation neural network being trained to identify, for each image in the source image data, a driving image from the driving image data, the identified driving image being implemented by the image transformation neural network to modify a corresponding source image in the source image data using motion estimation differences between the identified driving image and the corresponding source image, and stores the modified source image data.

Abstract: An animation processing method is performed by an electronic device. The method including: obtaining a terrain feature in a graphical user interface at a current moment, the graphical user interface including a virtual character, and obtaining state information and task information that correspond to the virtual character in an animation segment at the current moment; performing feature extraction on the terrain feature, the state information, and the task information using an animation processing model, to obtain joint action information corresponding to the virtual character at a next moment; determining a joint torque according to the joint action information; and obtaining gesture adjustment information corresponding to the virtual character from the current moment to the next moment based on the joint torque, and updating the animation segment according to the gesture adjustment information to render the updated animated segment in the graphical user interface.

Abstract: A method and apparatus for processing a pattern piece receives, as an input from a user, a selection of at least a portion of an outer line of a pattern piece to be processed among one or more pattern pieces included in a two-dimensional (2D) pattern of clothes, generates a template including a plurality of lines that divide an area set by the input, processes the template based on fullness settings set by the user for the pattern piece, processes the pattern piece based on the processed template, and outputs the processed pattern piece.

Abstract: An input apparatus for inputting a diagnosis result of a diagnosis target detectable for a structure includes circuitry configured to display a spherical image captured for the structure on a screen, receive an input of a position of the diagnosis target in the spherical image, store position information indicating the received position of the diagnosis target in the spherical image in a memory, display, on the screen, the spherical image and a diagnosis information input field used for inputting diagnosis information of the diagnosis target, receive an input of the diagnosis information of the diagnosis target input via the diagnosis information input field, and store the diagnosis information and the position information indicating the received position of the diagnosis target in the spherical image, in the memory in association with each other.

Abstract: In implementations of systems for generating digital objects to animate sketches, a computing device implements a sketch system to receive input data describing a user sketched digital object having a pose and a non-photorealistic style. The sketch system generates a latent vector representation of the user defined non-photorealistic style using an encoder of a generative adversarial network. A digital object is generated having the pose and a non-photorealistic style using a generator of the generative adversarial network based on the latent vector representation of the user defined non-photorealistic style. The sketch system modifies the latent vector representation of the user defined non-photorealistic style based on a comparison between the user defined non-photorealistic style and the non-photorealistic style.

Abstract: Techniques are described for efficiently reducing the amount of total computation in convolutional neural networks (CNNs) without affecting the output result or classification accuracy. Computation redundancy in CNNs is reduced by exploiting the computing nature of the convolution and subsequent pooling (e.g., sub-sampling) operations. In some implementations, the input features may be divided into a group of precision values and the operation(s) may be cascaded. A maximum may be identified (e.g., by 90% probability) using a small number of bits in the input features, and the full-precision convolution may then be performed on the maximum input. Accordingly, the total number of bits used to perform the convolution is reduced without affecting the output features or the final classification accuracy.

Abstract: Systems and methods are provided for measuring and identifying real-world tremors in order to calculate and display an anti-tremor in a virtual reality world. A system may identify a tremor in a body part, such as hands, and measure the tremor movement from, e.g., body sensors, to calculate an opposite movement and animate virtual hands performing the reverse tremor motion. Depicting inverse motions, such as an anti-tremor, may be used in a few ways to therapeutically train and exercise patients to better control their tremors. Providing a visual cue may help a patient interrupt the movement, as well as provide possibilities for therapy or training muscle memory and stimulating nerve and neuron recovery. By displaying an anti-tremor with virtual hands, a VR system may be used to help train or exercise body parts experiencing tremors.

Abstract: A method for encoding complex-valued signals of a computer-generated hologram into a phase-modulating optical element for the reconstruction of a three-dimensional object, and to a computer program product for encoding complex-valued signals of a computer-generated hologram, and to a holographic display for the reconstruction of a three-dimensional object. The object is to reduce the effort on encoding a complex-valued spatial distribution by an iteration method on the basis of phase encoding, so that the computer-generated hologram resulting therefrom can be represented more rapidly and with the same or an improved reconstruction quality. In particular, the convergence during the iterative optimization is intended to be accelerated. This is achieved by a method in which degrees of freedom of the hologram plane as well as the reconstruction plane are used for optimizing the iteration method for rapid convergence and maximization of the diffraction efficiency in the signal range.

Abstract: An example process executed by a processor of a computing device to communicate a first light field data to a user device, wherein the first light field data is limited to a first volume of space that represents a first plurality of views for display on the user device; receive a sensor data associated with the user device; predict a behavior based at least in part from the received sensor data; generate a second light field data based at least in part on the predicted behavior, wherein the second light field data is limited to a second volume of space that is different from the first volume of space and represents a second plurality of views for display on the user device; and communicate the second light field data to the user device.

Abstract: A method of image reconstruction for a cross-modal communication system is disclosed. The method reconstructs a damaged, lost, or delayed image signal during transmission by using complete haptic signals received by a receiving end in the cross-modal communication system, and further constructs a cross-modal interaction network with reference to an attention mechanism, thus solving the limitation of the conventional generation model that it can only be trained on paired samples. An image reconstruction device for a cross-modal communication system is also disclosed. By fully utilizing semantic correlation between different-modality data and realizing cross-modal generation from haptic signals to image signals for unpaired data, the present invention overcomes the difficulty in acquiring haptic-image signal pairs in the practical cross-modal communication system, and significantly improves the quality and class accuracy of generated image signals.

Abstract: Particular techniques for generating synthetic images and/or for training machine learning model(s) based on the generated synthetic images. For example, training a machine learning model based on training instances that each include a generated synthetic image, and ground truth label(s) for the generated synthetic image. After training of the machine learning model is complete, the trained machine learning model can be deployed on one or more robots and/or one or more computing devices.