Using a read sensor to sense wrong-way driving. A method may include sensing, by a rear sensor of a vehicle, an environment of the vehicle to provide rear sensed information; processing the rear sensed information to provide at least one rear-sensed vehicle progress direction indications; generating or receiving at least one front-sensed vehicle progress direction indications; wherein the at least one front-sensed vehicle progress direction indications is generated by processing front-sensed information acquired during right-way progress; comparing at least one rear-sensed vehicle progress direction indications to the at least one front-sensed vehicle progress direction indications to determine whether the vehicle is wrong-way driving; and responding to the finding of the wrong-way driving.

Methods and systems for use of a reference image to detect a road obstacle are described. A computing device configured to control a vehicle, may be configured to receive, from an image-capture device, an image of a road on which the vehicle is travelling. The computing device may be configured to compare the image to a reference image; and identify a difference between the image and the reference image. Further, the computing device may be configured to determine a level of confidence for identification of the difference. Based on the difference and the level of confidence, the computing device may be configured to modify a control strategy associated with a driving behavior of the vehicle; and control the vehicle based on the modified control strategy.

A vehicle detection apparatus includes circuitry configured to cut out a region of interest in an image, calculate a composite feature quantity of the region of interest as a whole by dividing the region of interest into a plurality of divided regions, calculating a feature quantity of each of the divided regions, and combining the calculated feature quantities of the respective divided regions together, and perform, on the basis of the calculated composite feature quantity, filtering that discards the region of interest that is determinable as not being a wheel of a vehicle.

An imaging device includes a pixel unit where pixels including photoelectric converters are arranged, and a signal processing unit that processes signals from the pixel unit. The pixel unit includes a first region including pixels each driven in a first mode to read a signal based on charge generated by the photoelectric converters and a second region including the pixels each driven in a second mode to read more signals than in the first mode including a signal based on charges generated by a part of the photoelectric converters and a signal based on charges generated by the photoelectric converters. The signal processing unit calculates a first correction value based on an average of a first data group from the first region and a second correction value based on an average of a second data group from the second region by using the first correction value as an initial value.

An image processing apparatus includes a representative-distance calculator and a joining processor. The representative-distance calculator generates representative distance values of the basis of a distance image generated from a stereo image including an image of at least one object including a vehicle. The distance image includes distance values of pixels. The joining processor performs a joining process of joining a first image region and a second image region that are defined based on the image of the at least one object and disposed apart from each other in the stereo image. The joining processor performs a determining process of determining whether the first image region and the second image region each include an image of a side face of the vehicle. The joining processor performs the joining process when the first image region and the second image region each include the image of the side face of the vehicle.

Trajectory prediction may receiving a LiDAR image sequence including a set of LiDAR images and generating a LiDAR map, generating an interaction encoder result by feeding the LiDAR image sequence through an interaction encoder, generating a feature extractor result by feeding the LiDAR map through a feature extractor, generating a relation encoder result by feeding a past trajectory of a detected obstacle from the LiDAR image sequence, the interaction encoder result, and the feature extractor result through a relation encoder, generating an intention estimation result by feeding the relation encoder result through an intention estimator, generating a conditional generative model result by feeding the past trajectory of the detected obstacle, the intention estimation result, and a probability map through a conditional generative model encoder, and generating a trajectory prediction by feeding the relation encoder result, the past trajectory of the detected obstacle, and the conditional generative model result through a trajectory predictor.

The present invention discloses an around view synthesis system, including: a plurality of cameras each mounted in a vehicle to capture respective different areas around the vehicle; a boundary setting unit setting a synthesis boundary of images captured in an overlapping region where images captured by the plurality of cameras are overlapped; and an image synthesizer receiving the images captured by the plurality of cameras and synthesizing the received images according to the synthesis boundary set by the boundary setting unit.

A vehicle can use an image sensor to both detect objects and determine depth data associated with the environment the vehicle is traversing. The vehicle can capture image data and lidar data using the various sensors. The image data can be provided to a machine-learned model trained to output depth data of an environment. Such models may be trained, for example, by using lidar data and/or three-dimensional map data associated with a region in which training images and/or lidar data were captured as ground truth data. The autonomous vehicle can further process the depth data and generate additional data including localization data, three-dimensional bounding boxes, and relative depth data and use the depth data and/or the additional data to autonomously traverse the environment, provide calibration/validation for vehicle sensors, and the like.

A method of controlling a robot includes obtaining a first image and a second image of a plurality of objects, the first and second image being captured from different positions; obtaining, from the first and second images, a plurality of candidate positions corresponding to each of the plurality of objects, based on a capturing position of each of the first and second images and a direction to each of the plurality of objects from each capturing position; obtaining distance information between each capturing position and each of the plurality of objects in the first and second images by analyzing the first and second images; and identifying a position of each of the plurality of objects from among the plurality of candidate positions based on the distance information.

Provided is a method including emitting, with a laser light emitter disposed on a robot, a collimated laser beam projecting a light point on a surface opposite the laser light emitter; capturing, with each of at least two image sensors disposed on the robot, images of the projected light point; overlaying, with a processor of the robot, the images captured by the at least two image sensors to produce a superimposed image showing both captured images in a single image; determining, with the processor of the robot, a first distance between the projected light points in the superimposed image; and determining, with the processor, a second distance based on the first distance using a relationship that relates distance between light points with distance between the robot or a sensor thereof and the surface on which the collimated laser beam is projected.