A car lens offset detection method and a car lens offset detection system are provided. The first image capturing device and the second image capturing device are disposed on a car, and the method includes: capturing a first image with use of a first image capturing device and capturing a second image with use of a second image capturing device; obtaining a plurality of first feature points from the first image according to a plurality of first predetermined positions and obtaining a plurality of second feature points from the second image according to a plurality of second predetermined positions; comparing feature values of the first feature points with feature values of the second feature points and determining whether the first feature points match the second feature points; in response to the first feature points not matching the second feature points, performing a calibration and warning operation.

A vehicle detecting device includes: a camera that includes an image sensor and a first filter, the image sensor including a plurality of imaging elements including a first imaging element group and a second imaging element group, the first filter keeping an amount of light entering the first imaging element group lower than an amount of light entering the second imaging element group; an image generating unit that generates a first image based on information obtained from the first imaging element group and a second image based on information obtained from the second imaging element group; and a detecting unit that detects a front vehicle based on the first image and the second image.

An object detection device configured to be mounted on a vehicle includes an object extraction unit that is configured to extract a point group that is a set of points representing a part of an object as the object, a neighboring object extraction unit that is configured to extract a neighboring object in an XY-plane of a world coordinate system, wherein the neighboring object is the object which is the closest to the vehicle, a coordinate transformation unit that is configured to transform coordinates of the neighboring object in the world coordinate system into coordinates of the neighboring object in an image captured by a camera, and a person determination unit that is configured to perform a person detection process in which it is determined whether or not the neighboring object is a person on the coordinates of the neighboring object.

A system and method for predicting driving actions based on intent-aware driving models that include receiving at least one image of a driving scene of an ego vehicle. The system and method also include analyzing the at least one image to detect and track dynamic objects located within the driving scene and to detect and identify driving scene characteristics associated with the driving scene and processing an ego-thing graph associated with the dynamic objects and an ego-stuff graph associated with the driving scene characteristics. The system and method further include predicting a driver stimulus action based on a fusion of representations of the ego-thing graph and the ego-stuff graph and a driver intention action based on an intention representation associated with driving intentions of a driver of the ego vehicle.

Systems and methods for three-dimensional object detection are disclosed herein. One embodiment inputs, to a neural network, a two-dimensional label associated with an object to produce a Normalized-Object-Coordinate-Space (NOCS) image and a shape vector, the shape vector mapping to a continuously traversable coordinate shape space (CSS); decodes the NOCS image and the shape vector to an object model in the CSS; back-projects, in a frustum, the NOCS image to a LIDAR point cloud; identifies correspondences between the LIDAR point cloud and the object model to estimate an affine transformation between the LIDAR point cloud and the object model; iteratively refines the affine transformation using a differentiable SDF renderer; extracts automatically a three-dimensional label for the object based, at least in part, on the iteratively refined affine transformation; and performs three-dimensional object detection of the object based, at least in part, on the extracted three-dimensional label for the object.

Techniques and systems are provided for determining one or more sizes of one or more objects. For example, a bounding region identifying a first object detected in an image can be obtained. A map including map points can also be obtained. The map points correspond to one or more reference locations in a three-dimensional space. The bounding region identifying the first object can be associated with at least one map point of the map points included in the map. Using the bounding region and the at least one map point, an estimated three-dimensional position and an estimated size of the first object detected in the image can be determined. In some examples, other information can be used to estimate the estimated three-dimensional position and an estimated size of the first object, such as radar information and/or other information.

The present disclosure relates to an XR device and a method for controlling the same, and more particularly, is applicable to a 5G communication technology field, a robot technology field, an autonomous technology field and an artificial intelligence (AI) technology field. The method for controlling an XR device of a vehicle includes acquiring a camera view by capturing an image in front of the vehicle; acquiring position information of the vehicle by detecting a position of the vehicle, acquiring movement information of the vehicle by detecting movement of the vehicle, and providing navigation of an augmented reality (AR) mode displaying at least one virtual object for guiding a path by overlapping the at least one virtual object on the camera view based on at least the position information of the vehicle or the movement information of the vehicle.

A system and a method for detecting a pile of material by an autonomous machine. A 3D point cloud method includes obtaining a 3D point cloud indicative of an environment having a material pile, performing a ground surface estimation on the point cloud to identify non-ground points, grouping the non-ground points into clusters based on a proximity metric, creating a normalized height histogram for each of the clusters, comparing the normalized height histogram of each cluster to a generalized pile histogram, and identifying a cluster as a pile based on the similarity between the normalized height histogram and the generalized pile histogram. A 2D image method includes obtaining a 2D image from an imaging device, calibrating the imaging device with respect to a coordinate frame of the machine, and autonomously detecting an image of a material pile in the two-dimensional image using a trained deep-learning neural network.

A vehicle and a control method thereof are provided. The vehicle may include a camera configured to acquire a front image of the vehicle; a storage configured to store an outline image of a reference license plate, and store location information including a position of a front vehicle based on a size of the outline image of the reference license plate, and a controller configured to recognize a license plate included in the front image, and determine a position of the front vehicle based on the license plate and the location information.

A control method for a mobile object that automatically moves includes: causing a sensor disposed on the mobile object to detect a target object multiple times to acquire detection results of respective detection processes performed by the sensor as a point group; estimating a position and an attitude of the target object for each detection process based on the point group; calculating reliability of accuracy in an estimation result of the position and the attitude of the target object for each detection process; selecting, from among estimation results of the respective detection processes, an estimation result used for setting an approach path to a target position at which a predetermined position and attitude with respect to the target object are achieved based on the reliability for each detection process; setting the approach path based on the selected estimation result; and causing the mobile object to move along the approach path.