Abstract: A method learns unsupervised world models for autonomous driving via discrete diffusion. The method includes encoding an observation of an actor for a geographic region using an encoder to generate a prior frame of prior tokens. The method further includes processing the prior frame with a spatio-temporal transformer to generate a predicted frame of predicted tokens. The spatio-temporal transformer includes a spatial transformer and a temporal transformer. The method further includes processing the predicted frame to generate a predicted action for the actor. The method further includes decoding the predicted frame to generate a predicted observation of the geographic region.

Abstract: A method of training a knowledge distillation based continual semantic segmentation model may include training the continual semantic segmentation model based on training data, predicting a probability for each class of an output of a current continual semantic segmentation model that has been generated by being trained and a probability for each class of an output of an old continual semantic segmentation model, expanding a class related to the old continual semantic segmentation model so that a spatial dimension of a class related to the old continual semantic segmentation model is the same as a spatial dimension of a class related to the current continual semantic segmentation model, calculating knowledge distillation loss based on the predicted probability for each class, and updating the current continual semantic segmentation model based on the knowledge distillation loss.

Abstract: Systems and methods for generating simulation data based on real-world dynamic objects are provided. A method includes obtaining two- and three-dimensional data descriptive of a dynamic object in the real world. The two- and three-dimensional information can be provided as an input to a machine-learned model to receive object model parameters descriptive of a pose and shape modification with respect to a three-dimensional template object model. The parameters can represent a three-dimensional dynamic object model indicative of an object pose and an object shape for the dynamic object. The method can be repeated on sequential two- and three-dimensional information to generate a sequence of object model parameters over time. Portions of a sequence of parameters can be stored as simulation data descriptive of a simulated trajectory of a unique dynamic object. The parameters can be evaluated by an objective function to refine the parameters and train the machine-learned model.

Abstract: Deferred neural lighting in augmented image generation includes performing operations. The operations include generating a source light representation of a real-world scene from a panoramic image of the real-world scene, augmenting the real-world scene in an object representation of the real-world scene to generate an augmented scene, and processing the augmented scene to generate augmented image buffers. The operations further include selecting a target lighting representation identifying a target light source, processing, by a neural deferred rendering model, the augmented image buffers, the source lighting representation, and a target lighting representation to generate an augmented image having a lighting appearance according to the target light source and outputting the augmented image.

Abstract: Systems and methods for generating simulation data based on real-world dynamic objects are provided. A method includes obtaining two- and three-dimensional data descriptive of a dynamic object in the real world. The two- and three-dimensional information can be provided as an input to a machine-learned model to receive object model parameters descriptive of a pose and shape modification with respect to a three-dimensional template object model. The parameters can represent a three-dimensional dynamic object model indicative of an object pose and an object shape for the dynamic object. The method can be repeated on sequential two- and three-dimensional information to generate a sequence of object model parameters over time. Portions of a sequence of parameters can be stored as simulation data descriptive of a simulated trajectory of a unique dynamic object. The parameters can be evaluated by an objective function to refine the parameters and train the machine-learned model.

Abstract: A camera module includes a base frame, a photosensitive chip, a first base plate, a second base plate, a first guide post, a second guide post, a first driving mechanism, and a second driving mechanism; the first guide post is disposed on the base frame, the first base plate is movably connected to the first guide post, and the first base plate moves in an extension direction of the first guide post; the second guide post is disposed on the first base plate, the second base plate is movably connected to the second guide post, the second base plate moves in an extension direction of the second guide post, and the photosensitive chip is disposed on the second base plate; the first driving mechanism is connected to the first base plate; and the second driving mechanism is connected to the second base plate

Abstract: The present disclosure relates to a method and apparatus for interacting with an application program, and an electronic device. The method includes: playing back multimedia content of a target object; and displaying, in a playback page for the multimedia content, an interaction component of the target object. The interactive component comprises: an integrated interactive card and action button. The interactive card is used for describing introductory content of the target object. The introductory content of the target object comprises: the icon of the target object, first basic information of the target object, and extension information of the target object. The action button is used for providing an entry to an interactive page of the target object.

Abstract: A camera structure includes a universal shaft, an outer gimbal support, an inner gimbal support, a gimbal carrier, a first driving mechanism, a second driving mechanism, and a camera module. The camera module is movably connected to the outer gimbal support, and the camera module is fixedly connected to the gimbal carrier. Two supporting portions of the universal shaft that are axially distributed along a first axis are hinged to the outer gimbal support, and two supporting portions of the universal shaft that are axially distributed along a second axis are hinged to the inner gimbal support. The first driving mechanism is configured to drive the inner gimbal support to rotate relative to the outer gimbal support along the first axis and/or the second axis. The second driving mechanism is configured to drive the gimbal carrier to rotate relative to the inner gimbal support along a third axis.

Abstract: A camera apparatus and an electronic device are disposed. The disclosed camera apparatus includes a supporting circuit board, an outer frame base, an inner frame base, a photosensitive chip, a lens holder, a lens, and a first driving mechanism, where the outer frame base is disposed on the supporting circuit board, the lens holder is disposed on the outer frame base, the lens is disposed on the lens holder, the supporting circuit board is electrically connected to the photosensitive chip via the outer frame base and the inner frame base sequentially; the inner frame base is movably disposed on the outer frame base, the photosensitive chip is disposed on the inner frame base, the photosensitive chip is arranged opposite to the lens, the first driving mechanism is configured to drive the inner frame base to move.

Abstract: Three dimensional object reconstruction for sensor simulation includes performing operations that include rendering, by a differential rendering engine, an object image from a target object model, and computing, by a loss function of the differential rendering engine, a loss based on a comparison of the object image with an actual image and a comparison of the target object model with a corresponding lidar point cloud. The operations further include updating the target object model by the differential rendering engine according to the loss, and rendering, after updating the target object model, a target object in a virtual world using the target object model.

Abstract: A camera module and an electronic device are provided. The camera module includes a housing including an accommodating cavity; a lens assembly movably disposed in the accommodating cavity; a circuit board connected to the lens assembly; a support plate. At least a part of the support plate is movably disposed in the accommodating cavity. The lens assembly and the circuit board are disposed on the support plate and move synchronously with the support plate. The camera module further includes a drive assembly disposed on the housing. At least a part of the drive assembly is connected to the support plate to drive the support plate to move.

Abstract: A camera module is disclosed and includes: a housing, internally provided with an accommodation cavity; a lens assembly, movably arranged in the accommodation cavity; a circuit board, connected to lens assembly; and a gimbal assembly, including: a support plate, where at least part of the support plate is movably arranged in the accommodation cavity, and the lens assembly and the circuit board are arranged on the support plate; a first driving assembly, where the first driving assembly is arranged on the housing, and at least part of the first driving assembly is connected to the support plate to drive the support plate to move in a first direction; and a second driving assembly, where the second driving assembly is arranged on the support plate, and at least part of the second driving assembly is connected to the support plate to drive the support plate to move in a second direction.

Abstract: A camera module and an electronic device. The camera module includes: a housing, an accommodating chamber being provided in the housing; a lens assembly, the lens assembly being movably disposed in the accommodating chamber; a first drive assembly, the first drive assembly being connected to the lens assembly so as to drive the lens assembly to move, and at least part of the first drive assembly being disposed on an inner side wall of the housing; a circuit board, the circuit board being connected to the lens assembly; and a second drive assembly, the second drive assembly being disposed in the housing, and at least part of the second drive assembly being connected to the circuit board so as to drive the circuit board to move.

Abstract: Real world object reconstruction and representation include performing operations that include sampling locations along a camera ray from a virtual camera to a target object to obtain a sample set of the locations along the camera ray. For each location of the at least a subset of the sample set, the operations include determining a position of the location with respect to the target object, executing, based on the position, a reflectance multilayer perceptron (MLP) model, to determine an albedo and material shininess for the location, and computing a radiance for the location and based on a viewing direction of the camera ray using the albedo and the material shininess. The operations further includes rendering a color value for the camera ray by compositing the radiance across the first sample set.

Abstract: A method for a protocol in digital asset trading includes step as follow. Trading information is exchanged between the first and the second nodes, in which a first digital currency belonging to the first node and a second digital currency belonging to the second node are expected to be exchanged, and prices of the first and second digital currencies are time-dependent. An atomic swap script and a smart contract are generated. A compensation fee to be paid to the second node by the first node is computed. A redeeming or refunding event is performed in response to the atomic swap script and the smart contract with paying the compensation fee to the second node, in which the compensation fee is dependent on the prices of the first and second digital currencies such that is time-dependent.

Abstract: A FCN-based MTS data classification method is disclosed, comprising: generating input conditions according to a parameter of a multivariate Gaussian model and the MTS data; establishing correspondence between the input conditions and data categories of the MTS data via learning ability of an artificial intelligence model; determining at least one corresponding current input conditions according to current MTS of a target object; and determining current data categories corresponding to the current input conditions through the correspondence, and determining data categories corresponding to the input conditions identical to the current input conditions in the correspondence as the current data categories. The parameters of the multivariate Gaussian model corresponding to the MTS data are served as the input conditions, so that the accuracy is guaranteed, while a training speed of an artificial intelligence model is greatly improved, and the higher the data set dimension, the more significant the improvement.

Abstract: Systems and methods for generating simulation data based on real-world dynamic objects are provided. A method includes obtaining two- and three-dimensional data descriptive of a dynamic object in the real world. The two- and three-dimensional information can be provided as an input to a machine-learned model to receive object model parameters descriptive of a pose and shape modification with respect to a three-dimensional template object model. The parameters can represent a three-dimensional dynamic object model indicative of an object pose and an object shape for the dynamic object. The method can be repeated on sequential two- and three-dimensional information to generate a sequence of object model parameters over time. Portions of a sequence of parameters can be stored as simulation data descriptive of a simulated trajectory of a unique dynamic object. The parameters can be evaluated by an objective function to refine the parameters and train the machine-learned model.

Abstract: A method for a protocol in digital asset trading includes step as follow. Trading information is exchanged between the first and the second nodes, in which a first digital currency belonging to the first node and a second digital currency belonging to the second node are expected to be exchanged, and prices of the first and second digital currencies are time-dependent. An atomic swap script and a smart contract are generated. A compensation fee to be paid to the second node by the first node is computed. A redeeming or refunding event is performed in response to the atomic swap script and the smart contract with paying the compensation fee to the second node, in which the compensation fee is dependent on the prices of the first and second digital currencies such that is time-dependent.

Abstract: The present disclosure provides a method, device and system for detecting a working state of a tower controller, the method includes: collecting voice data of a tower controller, and extracting a keyword from the voice data; acquiring a video image of the tower controller, and acquiring a gaze area of the tower controller from the video image; analyzing and detecting whether the tower controller has correctly accomplished an observation action according to the gaze area of the tower controller and the keyword. The present disclosure implements more efficient and accurate detection on the working state of the tower controller, and at the same time ensures the safety of an aircraft in an airport area and reduces a risk of colliding with other obstacles.