A system and method for egocentric-vision based future vehicle localization that include receiving at least one egocentric first person view image of a surrounding environment of a vehicle. The system and method also include encoding at least one past bounding box trajectory associated with at least one traffic participant that is captured within the at least one egocentric first person view image and encoding a dense optical flow of the egocentric first person view image associated with the at least one traffic participant. The system and method further include decoding at least one future bounding box associated with the at least one traffic participant based on a final hidden state of the at least one past bounding box trajectory encoding and the final hidden state of the dense optical flow encoding.

In one approach, a method includes: displaying, to a user of a first vehicle, image data obtained using a first field of view of a camera of the first vehicle, where the camera collects the image data for objects located outside of the first vehicle; detecting, by at least one processing device of the first vehicle, a second vehicle; determining, by the one processing device, whether the second vehicle is within a predetermined region relative to the first vehicle; and in response to determining that the second vehicle is within the predetermined region, displaying image data obtained using a second field of view of the camera.

An object detection apparatus 100 is an apparatus for detecting an object in a fish-eye image. The object detection apparatus 100 includes a normalized image acquisition unit 10 configured to acquire a normalized image obtained by normalizing a fish-eye image in which an object appears; a position detecting unit 20 configured to detect position coordinates of the object in the normalized image; and a determination unit 30 configured to determine a positional relationship between the object and the object detection apparatus 100 using the position coordinates of the object in the normalized image.

In one embodiment, a method includes receiving sensor data from a sensor array of a vehicle while traveling on a road. The method includes determining a confidence score for a classification of an object based on the sensor data. The method includes generating an object graphic corresponding to the object based on the confidence score for the classification of the object. The method includes retrieving, from one or more third-party systems, third-party data associated with an environment in proximity to the object and associated with the road based on the classification. The method includes generating an overlay graphic corresponding to the environment in proximity to the object based on the third-party data. The method includes providing for display the object graphic rendered in association with the overlay graphic corresponding to the environment.

The method for processing a depth map includes the following steps: S1: correcting an image of a target area that is collected by an image collection apparatus; S2: performing binocular matching on the image to obtain a depth map of the target area; and S3: acquiring a distribution of obstacles around an UAV according to the depth map. The method further includes: acquiring execution times of the foregoing steps before executing the steps; and establishing at least two threads and at least one ring queue according to the execution times of the steps, and executing the steps by the at least two threads to reduce a total execution time.

A method for a sensor-based and memory-based representation of a surroundings of a vehicle. The vehicle includes an imaging sensor for detecting the surroundings. The method includes: detecting a sequence of images; determining distance data on the basis of the detected images and/or of a distance sensor of the vehicle, the distance data comprising distances between the vehicle and objects in the surroundings of the vehicle; generating a three-dimensional structure of a surroundings model on the basis of the distance data; recognizing at least one object in the surroundings of the vehicle on the basis of the detected images, in particular by a neural network; loading a synthetic object model on the basis of the recognized object; adapting the generated three-dimensional structure of the surroundings model on the basis of the synthetic object model and on the basis of the distance data; and displaying the adapted surroundings model.

A system is comprised of one or more processors coupled to memory. The one or more processors are configured to receive image data based on an image captured using a camera of a vehicle and to utilize the image data as a basis of an input to a trained machine learning model to at least in part identify a distance of an object from the vehicle. The trained machine learning model has been trained using a training image and a correlated output of an emitting distance sensor.

A vehicle detection method includes obtaining a target image and depth information of each pixel in the target image, obtaining a distance value of a vehicle candidate area in the target image according to the target image and the depth information, and determining a detection model corresponding to the vehicle candidate area according to the distance value of the vehicle candidate area.

A vehicle detection method includes obtaining a target image and depth information of each pixel in the target image, obtaining a distance value of a vehicle candidate area in the target image according to the target image and the depth information, and determining a detection model corresponding to the vehicle candidate area according to the distance value of the vehicle candidate area.

Apparatuses, systems, and techniques are presented to determine distance for one or more objects. In at least one embodiment, a disparity network is trained to determine distance data from input stereoscopic images using a loss function that includes at least one of a gradient loss term and an occlusion loss term.