Abstract: According to the embodiments of the present disclosure, a data processing method, an electronic device, and a non-transitory computer-readable storage medium are provided. The method includes: obtaining initial prompt information generated based on a user input at a client; encrypting the initial prompt information using a first key to obtain encrypted prompt information; retrieving, from a plurality of encrypted knowledge segments, at least one first encrypted knowledge segment matching the encrypted prompt information, wherein the plurality of encrypted knowledge segments are encrypted using the first key; and obtaining, using a machine learning model and based on the at least one first encrypted knowledge segment, a reply to the user input.

Abstract: Systems and methods for training multi-view 3D object detection frameworks are disclosed herein. In one example, a method includes the steps of predicting one or more predicted bounding boxes representing one or more objects within multi-view images using a decoder that considers (a) feature embeddings generated from image features from multi-view images, (b) geometric positional encodings that are associated with the feature embeddings, and (c) view-dependent queries, determining a viewpoint equivariance loss based on a comparison of the one or more predicted bounding boxes with one or more ground truth bounding boxes, and adjusting model weights of networks forming the multi-view 3D object detection framework based on the viewpoint equivariance loss.

Abstract: Systems and methods for enhanced end-to-end three-dimensional (3-D) object detection are disclosed that improve detecting objects in the 3D space from images, such as monocular camera image that may be captured during the operation the autonomous vehicles. For example, a vehicle can include a processor device detecting one or more objects in a 3D space by predicting 3D bounding boxes and predicting dense depth associated with target assignments. The target assignments correspond to the location of objects within an image of the 3D space of a surrounding environment for the vehicle. The vehicle can also include a controller device that receives the detection of the objects in the 3D space from the processor device and performs autonomous operations. The end-to-end 3D object detection techniques achieve a high level of object detection accuracy, with significant improvements compared to previous methods, due to the simplicity and optimization of its end-to-end functionality.

Abstract: Systems and methods are provided for implementing a multi-stage, ML model training process for autonomous or semi-autonomous driving. The multi-stage ML model training process comprises (1) 2D and 3D supervised losses during a synthetic data ML model training, (2) 2D supervised on real-world data, and (3) 3D self-supervised losses on real-world data. The improved ML training process may not rely on 3D object recognition with real-world 3D labeled data. Once the ML model is trained, in some examples, the trained ML model can implement an inference process to predict the 3D shape, size, and 6D pose of objects within a single image, operate at a category level, and eliminate the need for computer-aided design (CAD) models during inference.

Abstract: Embodiment of the disclosure provide a data access method and an apparatus, a device and a readable storage medium. The method includes: in response to a demand of processing the data resource generated in a target application, sending a data access authorization request for the data resource to a plurality of clients of the target application, the plurality of clients being associated with the data resource. The authorization information for the data access authorization request is received respectively from at least one of the plurality of clients. At least one access credential respectively corresponding to the at least one client is obtained based on the authorization information. The target data associated with the at least one client in the data resource is accessed with the at least one access credential to process the target data.

Abstract: A system for producing a depth map can include a processor and a memory. The memory can store a candidate depth production module and a depth map production module. The candidate depth production module can include instructions that cause the processor to: (1) identify, in a first image, an epipolar line associated with a pixel in a second image and (2) sample, from a first image feature set, a set of candidate depths for pixels along the epipolar line. The depth map production module can include instructions that cause the processor to: (1) determine a similarity measure between a feature, from a second image feature set, and a member of the set and (2) produce, from the second image, the depth map with a depth for the pixel being a depth associated with a member, of the set, having a greatest similarity measure.

Abstract: Systems, methods, and other embodiments described herein relate to deriving a geometric projection of an object shape using a normalized object reference frame (NORF) information and completing the object shape from the geometric projection through diffusion and triplanar processing. In one embodiment, a method includes estimating a NORF image and a NORF normal for an object from an image and noise using a NORF diffusion model, the object having incomplete data. The method also includes deriving a projection of the object from a point cloud using the NORF image and the NORF normal. The method also includes predicting a completed shape for the object from the projection and triplanar noise using a triplanar diffusion model.

Abstract: Systems and methods are provided for implementing a multi-stage, ML model training process for autonomous or semi-autonomous driving. The multi-stage ML model training process comprises (1) 2D and 3D supervised losses during a synthetic data ML model training, (2) 2D supervised on real-world data, and (3) 3D self-supervised losses on real-world data. The improved ML training process may not rely on 3D object recognition with real-world 3D labeled data. Once the ML model is trained, in some examples, the trained ML model can implement an inference process to predict the 3D shape, size, and 6D pose of objects within a single image, operate at a category level, and eliminate the need for computer-aided design (CAD) models during inference.

Abstract: Systems and methods are provided for generating amodal images from occlusions objects in input images. Examples herein include receiving a prompt selecting an object in an input image; applying the input image to a trained conditional generative model that generates an amodal image of the selected object based on the prompt and the input image; and outputting the amodal image.

Abstract: A method for data processing, a readable medium, and an electronic device are provided. The method for data processing includes: receiving a data processing task; determining target data corresponding to the data processing task and a target first key corresponding to the target data; decrypting the target first key according to the first session key via a target computing node to obtain the first key, and decrypting the target data based on the first key to obtain the data to be processed; and determining a data processing result according to a target model and the data to be processed. The target computing node is executed in a trusted execution environment. The target model is obtained by decrypting an encrypted target model based on a model key, the model key is stored in a key management service, and the key management service is executed in the trusted execution environment.

Abstract: A method for data acquisition, a device and a storage medium are provided. The method includes: determining a data identification intersection between databases of data providers, where the data identification intersection comprises data identifications that are same between the databases of the data providers; constructing a Bloom vector of a Bloom Filter according to the data identification intersection, and sending the Bloom vector to the data providers; receiving candidate data sent by the data providers, where the candidate data is data corresponding to a target data identification, and the target data identification is determined by the data providers from data identifications of respective databases through the Bloom Filter based on the Bloom vector; and selecting target data corresponding to the data identification intersection from the candidate data.

Abstract: Systems and methods for enhanced computer vision capabilities, particularly including 3D transformation equivariance, which may be applicable to autonomous vehicle operation are described. A vehicle may be equipped with an 3D transformation equivariance architecture for performing equivariance of image data in the 3D space for image analysis and computer vision functions. The 3D Transformation Equivariance system and method can be configured to replace Fourier positional embedding with spherical harmonics, ensuring equivariance to 3D rotations for the input embedding. Furthermore, the 3D Transformation Equivariance system can be designed with varying architectures that utilize equivariant self-attention and cross-attention modules that are tailer to the spherical harmonics embedding within the general architecture.

Abstract: Systems, methods, and other embodiments described herein relate to estimating scaled depth maps by sampling variational representations of an image using a learning model. In one embodiment, a method includes encoding data embeddings by a learning model to form conditioned latent representations using attention networks, the data embeddings including features about an image from a camera and calibration information about the camera. The method also includes computing a probability distribution of the conditioned latent representations by factoring scale priors. The method also includes sampling the probability distribution to generate variations for the data embeddings. The method also includes estimating scaled depth maps of a scene from the variations at different coordinates using the attention networks.

Abstract: A system for producing a depth map can include a processor and a memory. The memory can store a neural network. The neural network can include an encoding portion module, a multi-frame feature matching portion module, and a decoding portion module. The encoding portion module can include instructions that, when executed by the processor, cause the processor to encode an image to produce single-frame features. The multi-frame feature matching portion module can include instructions that, when executed by the processor, cause the processor to process the single-frame features to produce information. The decoding portion module can include instructions that, when executed by the processor, cause the processor to decode the information to produce the depth map. A first training dataset, used to train the multi-frame feature matching portion module, can be different from a second training dataset used to train the encoding portion module and the decoding portion module.

Abstract: A method for data processing, a readable medium, and an electronic device are provided. The method for data processing includes: receiving a data processing task; determining target data corresponding to the data processing task and a target first key corresponding to the target data; decrypting the target first key according to the first session key via a target computing node to obtain the first key, and decrypting the target data based on the first key to obtain the data to be processed; and determining a data processing result according to a target model and the data to be processed. The target computing node is executed in a trusted execution environment. The target model is obtained by decrypting an encrypted target model based on a model key, the model key is stored in a key management service, and the key management service is executed in the trusted execution environment.

Abstract: A method for data acquisition, a device and a storage medium are provided. The method includes: determining a data identification intersection between databases of data providers, where the data identification intersection comprises data identifications that are same between the databases of the data providers; constructing a Bloom vector of a Bloom Filter according to the data identification intersection, and sending the Bloom vector to the data providers; receiving candidate data sent by the data providers, where the candidate data is data corresponding to a target data identification, and the target data identification is determined by the data providers from data identifications of respective databases through the Bloom Filter based on the Bloom vector; and selecting target data corresponding to the data identification intersection from the candidate data.

Abstract: Systems, methods, and other embodiments described herein relate to estimating scaled depth maps by sampling variational representations of an image using a learning model. In one embodiment, a method includes encoding data embeddings by a learning model to form conditioned latent representations using attention networks, the data embeddings including features about an image from a camera and calibration information about the camera. The method also includes computing a probability distribution of the conditioned latent representations by factoring scale priors. The method also includes sampling the probability distribution to generate variations for the data embeddings. The method also includes estimating scaled depth maps of a scene from the variations at different coordinates using the attention networks.

Abstract: Systems, methods, and other embodiments described herein relate to augmenting image embeddings using derived geometries for estimating scaled depth. In one embodiment, a method includes generating a geometric viewing vector using pixel coordinates and intrinsic parameters about a camera for an image captured about a scene. The method also includes deriving geometric embeddings from the geometric viewing vector associated with the image for the camera. The method also includes computing a representation by augmenting image embeddings with the geometric embeddings, the image embeddings associated with visual characteristics about the image. The method also includes estimating a scaled depth of the image from the representation.

Abstract: Systems, methods, and other embodiments described herein relate to augmenting an image frame during training that enhances scene geometries and transformation capabilities for depth prediction. In one embodiment, a method includes generating rays with camera intrinsics to form a grid for an image frame. The method also includes injecting noise, by an encoder during training of a learning model, to individually perturb pixels within pixel boundaries for the rays, the pixel boundaries defined by the grid. The method also includes removing a subset of the rays randomly by the encoder and extract features from the rays. The method also includes comparing scaled depth estimates to a ground truth for a grid resolution using the features and adjust the learning model from the comparison.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for generating trajectory predictions for one or more agents in an environment. In one aspect, a method comprises: obtaining scene context data characterizing a scene in an environment at a current time point and generating a respective predicted future trajectory for each of a plurality of agents in the scene at the current time point by sampling a sequence of discrete motion tokens that defines a joint future trajectory for the plurality of agents using a trajectory prediction neural network that is conditioned on the scene context data.