公开(公告)号：US20210350185A1

公开(公告)日：2021-11-11

申请号：US17383465

申请日：2021-07-23

Applicant: Cognata Ltd.

Inventor： Dan ATSMON ,  Eran ASA ,  Ehud SPIEGEL

IPC: G06K9/62 ,  G06T7/55 ,  G06T7/579 ,  G05D1/00 ,  G05D1/02 ,  G06K9/00 ,  G06N3/04 ,  G06N3/08 ,  G06T3/00

Abstract: A system for creating synthetic data for testing an autonomous system, comprising at least one hardware processor adapted to execute a code for: using a machine learning model to compute a plurality of depth maps based on a plurality of real signals captured simultaneously from a common physical scene, each of the plurality of real signals are captured by one of a plurality of sensors, each of the plurality of computed depth maps qualifies one of the plurality of real signals; applying a point of view transformation to the plurality of real signals and the plurality of depth maps, to produce synthetic data simulating a possible signal captured from the common physical scene by a target sensor in an identified position relative to the plurality of sensors; and providing the synthetic data to at least one testing engine to test an autonomous system comprising the target sensor.

公开(公告)号：US20230306680A1

公开(公告)日：2023-09-28

申请号：US18202970

申请日：2023-05-29

Applicant: Cognata Ltd.

Inventor： Dan ATSMON ,  Eran ASA ,  Ehud SPIEGEL

IPC: G06T15/10 ,  G06T7/55 ,  G06T7/579 ,  G05D1/00 ,  G05D1/02 ,  G06N3/04 ,  G06N3/08 ,  G06T3/00 ,  G06V20/56 ,  G06F18/21 ,  G06F18/24 ,  G06F18/28 ,  G06F18/214

CPC classification number: G06T15/10 ,  G06T7/55 ,  G06T7/579 ,  G05D1/0088 ,  G05D1/0246 ,  G06N3/04 ,  G06N3/08 ,  G06T3/0018 ,  G06V20/56 ,  G06F18/217 ,  G06F18/24 ,  G06F18/28 ,  G06F18/2148 ,  G05D2201/0213 ,  G06T2207/10028 ,  G06T2207/20081 ,  G06T2207/20084 ,  G06T2207/30252 ,  G06V2201/07

Abstract: A system for creating synthetic data for testing an autonomous system, comprising at least one hardware processor adapted to execute a code for: using a machine learning model to compute a plurality of depth maps based on a plurality of real signals captured simultaneously from a common physical scene, each of the plurality of real signals are captured by one of a plurality of sensors, each of the plurality of computed depth maps qualifies one of the plurality of real signals; applying a point of view transformation to the plurality of real signals and the plurality of depth maps, to produce synthetic data simulating a possible signal captured from the common physical scene by a target sensor in an identified position relative to the plurality of sensors; and providing the synthetic data to at least one testing engine to test an autonomous system comprising the target sensor.

公开(公告)号：US20210312244A1

公开(公告)日：2021-10-07

申请号：US17286526

申请日：2019-10-15

Applicant: Cognata Ltd.

Inventor： Dan ATSMON ,  Eran ASA ,  Ehud SPIEGEL

IPC: G06K9/62 ,  G06N20/00 ,  G06N5/04 ,  G06T7/00

Abstract: A method for training a model for generating simulation data for training an autonomous driving agent, comprising: analyzing real data, collected from a driving environment, to identify a plurality of environment classes, a plurality of moving agent classes, and a plurality of movement pattern classes; generating a training environment, according to one environment class; and in at least one training iteration: generating, by a simulation generation model, a simulated driving environment according to the training environment and according to a plurality of generated training agents, each associated with one of the plurality of agent classes and one of the plurality of movement pattern classes; collecting simulated driving data from the simulated environment; and modifying at least one model parameter of the simulation generation model to minimize a difference between a simulation statistical fingerprint, computed using the simulated driving data, and a real statistical fingerprint, computed using the real data.

公开(公告)号：US20200210779A1

公开(公告)日：2020-07-02

申请号：US16594200

申请日：2019-10-07

Applicant: Cognata Ltd.

Inventor： Dan ATSMON ,  Eran ASA ,  Ehud SPIEGEL

IPC: G06K9/62 ,  G06T7/579 ,  G06T3/00 ,  G06T7/55 ,  G06K9/00 ,  G05D1/02 ,  G05D1/00 ,  G06N3/08 ,  G06N3/04

Abstract: A system for creating synthetic data for testing an autonomous system, comprising at least one hardware processor adapted to execute a code for: using a machine learning model to compute a plurality of depth maps based on a plurality of real signals captured simultaneously from a common physical scene, each of the plurality of real signals are captured by one of a plurality of sensors, each of the plurality of computed depth maps qualifies one of the plurality of real signals; applying a point of view transformation to the plurality of real signals and the plurality of depth maps, to produce synthetic data simulating a possible signal captured from the common physical scene by a target sensor in an identified position relative to the plurality of sensors; and providing the synthetic data to at least one testing engine to test an autonomous system comprising the target sensor.

公开(公告)号：US20230202511A1

公开(公告)日：2023-06-29

申请号：US17926598

申请日：2021-05-27

Applicant: Cognata Ltd.

Inventor： Dan ATSMON ,  Ehud SPIEGEL

IPC: B60W60/00

CPC classification number: B60W60/001 ,  B60W2554/4049 ,  B60W2420/42

Abstract: A system for generating simulated driving scenarios, comprising at least one hardware processor adapted for generating a plurality of simulated driving scenarios, each generated by providing a plurality of input driving objects to a machine learning model, where the machine learning model is trained using another machine learning model, trained to compute a classification indicative of a likelihood that a simulated driving scenario produced by the machine learning model comprises an interesting driving scenario.

公开(公告)号：US20230316789A1

公开(公告)日：2023-10-05

申请号：US18022556

申请日：2021-09-14

Applicant: Cognata Ltd.

Inventor： Ilan TSAFRIR ,  Guy TSAFRIR ,  Ehud SPIEGEL ,  Dan ATSMON

IPC: G06V20/70 ,  G06V20/58 ,  G06V10/77

CPC classification number: G06V20/70 ,  G06V20/58 ,  G06V10/7715

Abstract: There is provided a method for annotating digital images for training a machine learning model, comprising: generating, from digital images and a plurality of dense depth maps, each associated with one of the digital images, an aligned three-dimensional stacked scene representation of a scene, where the digital images are captured by sensor(s) at the scene, and where each point in the three-dimensional stacked scene is associated with a stability score indicative of a likelihood the point is associated with a static object of the scene, removing from the three-dimensional stacked scene unstable points to produce a static three-dimensional stacked scene, detecting in at least one of the digital images static object(s) according to the static three-dimensional stacked scene, and classifying and annotating the static object(s). The machine learning model may be trained on the images annotated with a ground truth of the static object(s).

公开(公告)号：US20220188579A1

公开(公告)日：2022-06-16

申请号：US17687720

申请日：2022-03-07

Applicant: Cognata Ltd.

Inventor： Dan ATSMON ,  Eran ASA ,  Ehud SPIEGEL

IPC: G06K9/62 ,  G06N20/00 ,  G06N5/04 ,  G06T7/00

Abstract: A method for training a model for generating simulation data for training an autonomous driving agent, comprising: analyzing real data, collected from a driving environment, to identify a plurality of environment classes, a plurality of moving agent classes, and a plurality of movement pattern classes; generating a training environment, according to one environment class; and in at least one training iteration: generating, by a simulation generation model, a simulated driving environment according to the training environment and according to a plurality of generated training agents, each associated with one of the plurality of agent classes and one of the plurality of movement pattern classes; collecting simulated driving data from the simulated environment; and modifying at least one model parameter of the simulation generation model to minimize a difference between a simulation statistical fingerprint, computed using the simulated driving data, and a real statistical fingerprint, computed using the real data.