-
公开(公告)号:US11804026B2
公开(公告)日:2023-10-31
申请号:US18047680
申请日:2022-10-19
Applicant: Aptiv Technologies Limited
Inventor: Weimeng Zhu , Yu Su , Christian Nunn
CPC classification number: G06V10/454 , G06N3/044 , G06N3/045 , G06N3/08 , G06T7/70 , G06T7/75 , G06V10/70 , G06V10/82
Abstract: A device for processing data sequences by means of a convolutional neural network is configured to carry out the following steps: receiving an input sequence comprising a plurality of data items captured over time using a sensor, each of said data items comprising a multi-dimensional representation of a scene, generating an output sequence representing the input sequence processed item-wise by the convolutional neural network, wherein generating the output sequence comprises: generating a grid-generation sequence based on a combination of the input sequence and an intermediate grid-generation sequence representing a past portion of the output sequence or the grid-generation sequence, generating a sampling grid on the basis of the grid-generation sequence, generating an intermediate output sequence by sampling from the past portion of the output sequence according to the sampling grid, and generating the output sequence based on a weighted combination of the intermediate output sequence and the input sequence.
-
公开(公告)号:US11604272B2
公开(公告)日:2023-03-14
申请号:US16904835
申请日:2020-06-18
Applicant: APTIV TECHNOLOGIES LIMITED
Inventor: Yu Su , Weimeng Zhu , Florian Kästner , Adrian Becker
IPC: G01S13/931 , G01S13/89
Abstract: A computer implemented method for object detection includes: determining a grid, the grid comprising a plurality of grid cells; determining, for a plurality of time steps, for each grid cell, a plurality of respective radar detection data, each radar detection data indicating a plurality of radar properties; determining, for each time step, a respective radar map indicating a pre-determined radar map property in each grid cell; converting the respective radar detection data of the plurality of grid cells for the plurality of time steps to a point representation of pre-determined first dimensions; converting the radar maps for the plurality of time steps to a map representation of pre-determined second dimensions, wherein the pre-determined first dimensions and the pre-determined second dimensions are at least partially identical; concatenating the point representation and the map representation to obtain concatenated data; and carrying out object detection based on the concatenated data.
-
公开(公告)号:US11521059B2
公开(公告)日:2022-12-06
申请号:US16373939
申请日:2019-04-03
Applicant: Aptiv Technologies Limited
Inventor: Weimeng Zhu , Yu Su , Christian Nunn
Abstract: A device for processing data sequences by means of a convolutional neural network is configured to carry out the following steps: receiving an input sequence comprising a plurality of data items captured over time using a sensor, each of said data items comprising a multi-dimensional representation of a scene, generating an output sequence representing the input sequence processed item-wise by the convolutional neural network, wherein generating the output sequence comprises: generating a grid-generation sequence based on a combination of the input sequence and an intermediate grid-generation sequence representing a past portion of the output sequence or the grid-generation sequence, generating a sampling grid on the basis of the grid-generation sequence, generating an intermediate output sequence by sampling from the past portion of the output sequence according to the sampling grid, and generating the output sequence based on a weighted combination of the intermediate output sequence and the input sequence.
-
公开(公告)号:US20220244383A1
公开(公告)日:2022-08-04
申请号:US17649666
申请日:2022-02-01
Applicant: Aptiv Technologies Limited
Inventor: Yu Su , Mirko Meuter
Abstract: Provided is a method for object detection in a surrounding of a vehicle using a deep neural network, comprising: inputting a first set of sensor-based data for a first Cartesian grid having a first spatial dimension and a first spatial resolution into a first branch of the deep neural network; inputting a second set of sensor-based data for a second Cartesian grid having a second spatial dimension and a second spatial resolution into a second branch of the deep neural network; providing an interaction between the first branch of the deep neural network and the second branch of the deep neural network at an intermediate stage of the deep neural network; and fusing a first output of the first branch of the deep neural network and a second output of the second branch of the deep neural network to detect the object in the surrounding of the vehicle.
-
公开(公告)号:US10977500B2
公开(公告)日:2021-04-13
申请号:US16369914
申请日:2019-03-29
Applicant: Aptiv Technologies Limited
Abstract: The present invention relates to a method and an image processing system for determining the color of a street marking, by: capturing an image of a street, detecting a street marking as a set of pixels provided by the image, wherein the pixels include at least two different pieces of color information, determining a color score for the street marking by comparing said at least two different pieces of color information, and determining the color of the street marking by comparing the color score to at least one threshold value.
-
公开(公告)号:US20190354779A1
公开(公告)日:2019-11-21
申请号:US16383858
申请日:2019-04-15
Applicant: Aptiv Technologies Limited
Abstract: A device for searching for a lane on which a vehicle can drive, wherein the device is configured to receive an image captured by a camera, the image showing an area in front of the vehicle, detect lane markings in the image, determine for each of the lane markings if the respective lane marking is of a first type indicating a road condition of a first type or a second type indicating a road condition of a second type, create lane candidates from the lane markings, divide the lane candidates into classes of lane candidates depending on the type of the lane markings of the lane candidates, and search the classes of lane candidates for a lane on which the vehicle can drive.
-
7.
公开(公告)号:US20190325306A1
公开(公告)日:2019-10-24
申请号:US16373939
申请日:2019-04-03
Applicant: Aptiv Technologies Limited
Inventor: Weimeng Zhu , Yu Su , Christian Nunn
IPC: G06N3/08
Abstract: A device for processing data sequences by means of a convolutional neural network is configured to carry out the following steps: receiving an input sequence comprising a plurality of data items captured over time using a sensor, each of said data items comprising a multi-dimensional representation of a scene, generating an output sequence representing the input sequence processed item-wise by the convolutional neural network, wherein generating the output sequence comprises: generating a grid-generation sequence based on a combination of the input sequence and an intermediate grid-generation sequence representing a past portion of the output sequence or the grid-generation sequence, generating a sampling grid on the basis of the grid-generation sequence, generating an intermediate output sequence by sampling from the past portion of the output sequence according to the sampling grid, and generating the output sequence based on a weighted combination of the intermediate output sequence and the input sequence.
-
公开(公告)号:US20220261670A1
公开(公告)日:2022-08-18
申请号:US17661913
申请日:2022-05-03
Applicant: Aptiv Technologies Limited
Inventor: Farzin G. Rajabizadeh , Narges Milani , Daniel Schugk , Lutz Roese-Koerner , Yu Su , Dennis Mueller
Abstract: A method of processing image data in a connectionist network comprises a plurality of units, wherein the method implements a multi-channel unit forming a respective one of the plurality of units, and wherein the method comprises: receiving, at the data input, a plurality of input picture elements representing an image acquired by means of a multi-channel image sensor, wherein the plurality of input picture elements comprise a first and at least a second portion of input picture elements, wherein the first portion of input picture elements represents a first channel of the image sensor and the second portion of input picture elements represents a second channel of the image sensor; processing of the first and at least second portion of input picture elements separately from each other; and outputting, at the data output, the processed first and second portions of input picture elements.
-
公开(公告)号:US20190318175A1
公开(公告)日:2019-10-17
申请号:US16369914
申请日:2019-03-29
Applicant: Aptiv Technologies Limited
Abstract: The present invention relates to a method and an image processing system for determining the color of a street marking, by: capturing an image of a street, detecting a street marking as a set of pixels provided by the image, wherein the pixels include at least two different pieces of color information, determining a color score for the street marking by comparing said at least two different pieces of color information, and determining the color of the street marking by comparing the color score to at least one threshold value.
-
10.
公开(公告)号:US11195038B2
公开(公告)日:2021-12-07
申请号:US16374138
申请日:2019-04-03
Applicant: Aptiv Technologies Limited
Inventor: Christian Nunn , Weimeng Zhu , Yu Su
IPC: G06K9/00
Abstract: A device for extracting dynamic information comprises a convolutional neural network, wherein the device is configured to receive a sequence of data blocks acquired over time, each of said data blocks comprising a multi-dimensional representation of a scene. The convolutional neural network is configured to receive the sequence as input and to output dynamic information on the scene in response, wherein the convolutional neural network comprises a plurality of modules, and wherein each of said modules is configured to carry out a specific processing task for extracting the dynamic information.
-
-
-
-
-
-
-
-
-