Abstract: A parameterization method for point cloud data, and a device includes performing semantic segmentation on a plurality of three-dimensional data points of point cloud data in a spatial coordinate system to obtain at least one semantic object, the at least one semantic object corresponds to a plurality of three-dimensional semantic data points. The parameterization method for point cloud data further includes performing parametric fitting on the plurality of three-dimensional semantic data points to obtain spatial geometric parameters of the at least one semantic object corresponding to the plurality of three-dimensional semantic data points in the spatial coordinate system. In addition, a map construction method is further provided. By adopting the above technical solution, the automation of constructing a semantic map based on the point cloud data can be achieved.

Abstract: Disclosed are a method and apparatus for modeling an object, a storage medium, and a vehicle control method. The method for modeling an object includes obtaining a point cloud associated with surroundings acquired by a perception system so that points corresponding to each part of the object are respectively recognized from the point cloud, and then corresponding representations are respectively generated for the object wholly and each part thereof. As such, the object is modeled and the modeling precision is improved, whereby a movement state of the object may be accurately determined.

Abstract: The present disclosure provides a method and apparatus for modeling an object. The method includes: obtaining a first group of point cloud frames associated with surroundings of a perception system; generating a hierarchical representation of an object in the surroundings based on the first group of point cloud frames; obtaining, from each frame of point cloud, a second group of point cloud frames associated with the surroundings, a group of points associated with the object and a first subset and a second subset of the group of points; determining, based on the first subset and the second subset, a first confidence and a second confidence; determining whether to change the hierarchical representation of the object based on the first confidence and the second confidence.

Abstract: A method includes: performing a voxelization processing on point cloud data to obtain a plurality of voxels, wherein the plurality of voxels correspond to a plurality of points in the point cloud data, and at least a portion of the plurality of voxels forms a voxel set; extracting a plurality of voxel features from the voxels in the voxel set; mapping the plurality of voxel features to a plurality of points included in the plurality of voxels, respectively, to obtain a plurality of point features of the plurality of points; and predicting, according to the plurality of point features, the target object.

Abstract: The present disclosure relates to a method for state estimation of a target comprising: obtaining an observation variable of the target at different moments through a plurality of sensors, wherein at least one observation variable is acquired by each sensor; determining a state variable of the target at different moments based on the observation variable; and optimizing the state variable of the target by minimizing a loss function. The loss function includes at least one of a position loss, an orientation loss, a velocity loss, a size loss, or a structural constraint of the target. The method of the present disclosure may obtain a sufficiently accurate state estimate. In addition, an apparatus, an electronic device, and a medium for state estimation of the target are also provided.

Abstract: The present disclosure relates to a method for target state estimation comprising: acquiring a data frame sequence corresponding to a plurality of moments; determining a first time window based on the data frame sequence, the first time window comprising a first set of state variables to be optimized; determining a plurality of second time windows within the first time window, wherein the number of data frames in each second time window is less than the number of data frames in the first time window, there is a repeated data frame in two adjacent second time windows, and there is a second set of state variables to be optimized; and optimizing the second set of state variables in the plurality of second time windows to obtain an optimized first set of state variables. The method of the present disclosure may obtain sufficiently accurate state estimation.

Abstract: The present disclosure provides a method and an apparatus for semantic segmentation of an image, capable of solving the problem in the related art associated with low speed and inefficiency in semantic segmentation of images. The method includes: receiving the image; performing semantic segmentation on the image to obtain an initial semantic segmentation result; and inputting image information containing the initial semantic segmentation result to a pre-trained convolutional neural network for semantic segmentation post-processing, so as to obtain a final semantic segmentation result. With the solutions of the present disclosure, the initial semantic segmentation result can be post-processed using the convolutional neural network, such that the speed and efficiency of the semantic segmentation of the image can be improved.

Abstract: The present disclosure relates to a method for training a neural network model and a method for generating an image. The method for training a neural network model includes: acquiring an image about a scene captured by a camera; determining a plurality of rays at least according to parameters of the camera when capturing the image; determining a plurality of sampling points according to a relative positional relationship between the rays and a point cloud, where the point cloud is associated with a part of the scene; determining color information of pixels of the image which correspond to the sampling points; and training the neural network model according to positions of the sampling points and the color information of the pixels.

Abstract: Disclosed are devices, systems and methods for processing an image. In one aspect a method includes receiving an image from a sensor array including an x-y array of pixels, each pixel in the x-y array of pixels having a value selected from one of three primary colors, based on a corresponding x-y value in a mask pattern. The method may further include generating a preprocessed image by performing preprocessing on the image. The method may further include performing perception on the preprocessed image to determine one or more outlines of physical objects.

Abstract: The present document relates to a method and an apparatus for generating a neural network. The method for generating a neural network according to the present document includes: training a plurality of neural networks for a plurality of performance parameters to obtain a plurality of parameter values for each performance parameter; training a plurality of neural network predictors based on the parameter values and the neural networks; and determining a target neural network using trained neural network predictors. According to the technique for generating a neural network herein, automatic searching for a network structure satisfying a preset constraint is enabled in a search space of network structures.

Abstract: The present disclosure provides a method and an apparatus for training a neural network and a computer server. The method includes: selecting automatically input data for which processing by the neural network fails, to obtain a set of data to be annotated; annotating the set of data to be annotated to obtain a new set of annotated data; acquiring a set of newly added annotated data containing the new set of annotated data, and determining a union of the set of newly added annotated data and a set of training sample data for training the neural network in a previous period as a set of training sample data for a current period; and training the neural network iteratively based on the set of training sample data for the current period, to obtain a neural network trained in the current period.

Abstract: A method and device for training a neural network are disclosed. The method comprises: selecting, by a training device, a teacher network performing the same functions of a student network; and iteratively training the student network and obtaining a target network, through aligning distributions of features between a first middle layer and a second middle layer corresponding to the same training sample data, so as to transfer knowledge of features of a middle layer of the teacher network to the student network.

Abstract: The present disclosure provides a method and an apparatus for creating an occupancy grid map, as well as a processing apparatus. The method includes: creating a current occupancy grid map based on a location of the vehicle and a previous occupancy grid map; and determining a current probability that each grid in the current occupancy grid map belongs to each of occupancy categories based on last environment perception information received from the sensors and updating an occupancy category to which each grid in the current occupancy grid map belongs based on the current probability that the grid belongs to each of the occupancy categories, in accordance with an asynchronous updating policy.

Abstract: Disclosed are devices, systems and methods for processing an image. In one aspect a method includes receiving an image from a sensor array including an x-y array of pixels, each pixel in the x-y array of pixels having a value selected from one of three primary colors, based on a corresponding x-y value in a mask pattern. The method may further include generating a preprocessed image by performing preprocessing on the image. The method may further include performing perception on the preprocessed image to determine one or more outlines of physical objects.

Abstract: The present disclosure relates to a traffic marker detection method and a training method for a traffic marker detection model, and relates to the technical field of intelligent transportation, and particularly to autonomous driving technology. The traffic marker detection method comprises acquiring a target image containing a traffic marker; and inputting the target image into a traffic marker detection model to obtain a detection mark corresponding to the traffic marker; wherein the detection mark comprises at least one of a detection point and a detection line for characterizing a position of the traffic marker in the target image.

Abstract: The present disclosure provides a method and an apparatus for sampling training data and a computer server. The method includes: inputting a video to a target detection model to obtain a detection result for each frame of image; inputting the detection results for all frames of images in the video to a target tracking model, to obtain a tracking result for each frame of image; and for each frame of image in the video: matching the detection result and the tracking result for the frame of image, and when the detection result and the tracking result for the frame of image are inconsistent with each other, determining the frame of image as a sample image to be marked, for which processing by the target detection model is not optimal.

Abstract: The present disclosure provides a method and an apparatus for creating an occupancy grid map, as well as a processing apparatus. The method includes: creating a current occupancy grid map based on a location of the vehicle and a previous occupancy grid map; and determining a current probability that each grid in the current occupancy grid map belongs to each of occupancy categories based on last environment perception information received from the sensors and updating an occupancy category to which each grid in the current occupancy grid map belongs based on the current probability that the grid belongs to each of the occupancy categories, in accordance with an asynchronous updating policy.

Abstract: A method and device for training a neural network are disclosed. The method comprises: selecting, by a training device, a teacher network performing the same functions of a student network; and iteratively training the student network and obtaining a target network, through aligning distributions of features between a first middle layer and a second middle layer corresponding to the same training sample data, so as to transfer knowledge of features of a middle layer of the teacher network to the student network.

Abstract: Embodiments of the present invention disclose a binocular image matching method, apparatus, device, and storage medium. The method comprises: performing target detection on a first image to obtain a first bounding box of a target in the first image; determining a second bounding box corresponding to the first bounding box in a second image; and obtaining a third bounding box of the target in the second image by regressing the second bounding box. The technical solutions realize accurate matching of targets in a binocular image without requiring performing target detection on both images in the binocular image, and then use a matching algorithm to match the targets detected in the two images, thereby greatly reducing the calculation overhead of target matching in the binocular image.

Abstract: The present disclosure provides a multiple target tracking method and apparatus, a calculating device and a storage medium, so as to solve the problem of inaccurate multiple target tracking in the prior art. The target tracking method comprises: obtaining a prediction box of a target in a current frame according to tracklets of one or more targets in historical frames; performing target detection on the current frame to obtain one or more detection boxes, wherein the detection boxes comprise a high-quality box and a medium-quality box; matching each prediction box with the detection box according to the similarity of the prediction box and the detection box; and in response to a prediction box being unmatched with the high-quality box but matched with the medium-quality box, determining that the target is in a tracking state in the current frame.