Methods for creating a video

Abstract

The invention relates to a method for creating a video (V) with highlights of a vehicle (1) being driven, using a plurality of image capture units (2) provided on the vehicle, which continuously capture image data during the vehicle's (1) operation. According to the invention, it is provided that - a recording mode for creating the video (V) is activated when a predefined driving mode is selected or manually, - Image sequences with highlights of the driving operation are automatically selected depending on the detected signals of a driving dynamics and / or detected image data of an environment of the vehicle (1) and / or facial expressions and / or gestures of a driver of the vehicle (1) by means of evaluation by a trained pattern recognition procedure.

Description

[0001] The invention relates to a method for creating a video with highlights of a vehicle's driving operation with a plurality of vehicle-side image capture units which continuously capture image data during the vehicle's driving operation.

[0002] German patent DE 10 2019 132 716 A1 discloses a camera system, a transmission system, and an application of the camera system and the transmission system in motorsport. The camera system comprises two cameras designed for mounting on a vehicle for the continuous recording of real camera images, a computing unit, and a data connection, wherein the two cameras and the computing unit are connected to each other via the data connection for the purpose of transmitting the camera images, and the computing unit is designed and configured to generate an imaginary camera image from a drone perspective in real time from the camera images.

[0003] The invention is based on the objective of providing a method for creating a video with highlights of a vehicle's driving operation.

[0004] The problem is solved according to the invention by a method which has the features specified in claim 1.

[0005] Advantageous embodiments of the invention are the subject of the dependent claims.

[0006] A method for creating a video with highlights of a vehicle's operation with a plurality of vehicle-side image capture units, which continuously capture image data during the vehicle's operation, provides according to the invention that - a recording mode for creating the video is activated when a predefined driving mode is selected or manually, - Image sequences with highlights of the driving operation are automatically selected depending on recorded signals of driving dynamics and / or recorded image data of the vehicle's environment and / or facial expressions and / or gestures of the driver of the vehicle by means of evaluation by a trained pattern recognition procedure.

[0007] By applying the procedure, a driver of the vehicle is given the opportunity to relive the driving operation in the specified driving mode, in particular an off-road mode, a racing mode and a drift mode, especially the highlights of the driving operation with the correspondingly activated driving mode, that is, summarized specific moments and views.

[0008] The process uses components already present in the vehicle, so it is not necessary to make any design-related changes to the vehicle.

[0009] In one version of the procedure, it is possible to embed optical and / or acoustic content stored on the driver's mobile device and / or digital camera into the video created.

[0010] Exemplary embodiments of the invention are explained in more detail below with reference to a drawing.

[0011] This shows: Fig. 1 schematically a vehicle with a device for creating a video with highlights of a vehicle's operation.

[0012] The single figure shows, in an exemplary and highly simplified manner, a vehicle 1 with a device for creating a video V with highlights of a driving operation of vehicle 1.

[0013] The vehicle 1 has a number of image capture units 2, for example driver assistance cameras, a driver observation camera, a dashboard camera (a so-called dashcam), an underbody camera, a rear camera, a side camera, a so-called fisheye camera, etc.

[0014] Furthermore, the vehicle 1 has a sensor system 3 comprising a number of acceleration and / or position sensors, at least one steering angle sensor, a detection unit for recording the instantaneous vehicle speed, wheel speed sensors, etc. In particular, the sensor system 3 is designed to determine vehicle dynamics data.

[0015] It is generally known that vehicles 1, such as an off-road vehicle, also known as a sport utility vehicle (SUV), or a sports car, have various driving modes, which are intended in particular for use on routes outside of public roads and are designed to provide the driver of such a vehicle 1 with a special driving experience. These driving modes include, in particular, an off-road mode, a race mode, and a drift mode.

[0016] The following describes a method for creating a video V with highlights of a vehicle 1 driving operation with a plurality of vehicle-side image capture units 2, which continuously capture image data during the vehicle's driving operation.

[0017] The procedure provides that the vehicle's image capture units 2 capture image data, based on which highlights are automatically determined following a driving operation in a specified driving mode, which are compiled into a video V in order to present a driving experience, in particular to a driver of vehicle 1, in a re-experiential way.

[0018] Image recording using the image capture units 2 is activated by selecting the driving mode, for example, off-road mode, race mode, or drift mode. Alternatively, recording can also be activated manually by the driver of vehicle 1, so that the creation of such a video V is not limited to these driving modes.

[0019] In particular, to automatically select special events as highlights in the driving operation of the vehicle 1, recording units of the vehicle's existing sensor system 3 are used, based on whose recorded signals it is possible to draw conclusions about special driving maneuvers and / or situations.

[0020] Examples include signals recorded from a satellite-based positioning unit, acceleration and position sensors, signals relating to the intensity of braking and / or steering interventions, a determined driving speed, signals recorded from sensors of a driver assistance system, which in particular provide information about the immediate surroundings of vehicle 1 and objects detected in this surroundings, such as other vehicles when vehicle 1 is on a race track, signals relating to the coefficient of friction between vehicle wheels and the surface, and image data recorded from a driver observation camera.

[0021] If the off-road mode of vehicle 1 is activated, then, for example, driving over a crest and / or spinning the vehicle wheels and / or detecting particularly strong vibrations and / or detecting extreme gradient values ​​and / or a special facial expression and / or a special gesture of the driver and / or a passenger and / or a braking intervention can constitute a special event in the driving operation of vehicle 1.

[0022] If the race mode of vehicle 1 is activated, then, for example, a drift performed and / or strong acceleration and / or a braking maneuver performed and / or a special facial expression and / or a special gesture of the driver and / or the passenger and / or a detected S-curve as a special track section and / or an overtaking maneuver of vehicle 1 can represent a special event in the driving operation of vehicle 1.

[0023] Examples of an emotional event or events during the operation of vehicle 1 could be a particularly beautiful road, nature, surroundings and / or a sunset and / or the visibility of a rim of at least one of the vehicle wheels.

[0024] The detection of an event relevant as a highlight of driving operations is based on one or more pattern recognition methods, for example, a deep artificial neural convolutional network, whereby the pattern recognition method(s) have been previously trained with respect to acquired signals and data, in particular image data. For example, the pattern recognition method can be stored in a control unit 4 of the vehicle 1, which is connected to the image acquisition units 2 and the sensors 3.

[0025] A relevant event is then linked to specifically generated views of the image acquisition units 2 from vehicle 1 and / or on vehicle 1, such as: - a drive over a crest with a view based on image data captured by an image acquisition unit 2 designed as a fisheye camera, - a drift with a view of a rear vehicle wheel and / or a wheel arch, wherein this view is based on image data captured by an image capture unit 2 designed as an exterior mirror camera of the vehicle 1 and - an overtaking maneuver with combined views of an overtaken vehicle based on image data captured from a front camera and / or a side camera and / or a rear camera as image capture unit 2 or image capture units 2.

[0026] After the end of driving operation, particularly in one of the aforementioned driving modes, the generated image sequences—that is, the views created based on the captured image data—are automatically assembled in chronological order into a video V in control unit 4, containing the relevant events as highlights. The video V, or individual image sequences, are then overlaid with acoustic signals and / or music detected in the vehicle interior.

[0027] In one possible implementation, the driver of vehicle 1 has the option to adjust the video V to their liking at specific points in the image sequence. For example, the driver could subsequently be given the opportunity to influence certain properties of the video V, such as choosing a different view and / or a different image capture unit 2 and / or a different recording time.

[0028] In another version of the procedure, it is possible to apply certain effects and / or styles, such as sporty or elegant, and / or filters to video V.

[0029] Furthermore, the captured image data and / or the video V can be enriched with generative content, allowing vehicle 1 to be combined with elements from a previous scene. It is also possible to highlight specific elements, such as dirt and / or mud splashes, in the video V.

[0030] In particular, the pattern recognition method learns what the driver of vehicle 1 likes in order to be able to make suggestions, especially with regard to a selection of highlights and / or with regard to post-processing of video V, in order to personalize video V.

[0031] If the video V is created in a central computer unit connected to the vehicle 1 via data technology, to which the recorded signals and data, in particular image data, especially image sequences containing the highlights of the driving operation, are transmitted by means of the control unit 4, then the information for processing and individual design of the video V is transmitted to the central control unit 4.

[0032] The completed video V is made available to the driver of vehicle 1 via control unit 4 or the central computer unit for download to any mobile device, in particular a smartphone. For example, the download can be carried out using a vehicle-specific application program installed on the mobile device.

[0033] The procedure therefore provides that, based on an automated selection of particularly suitable image data and / or acoustic signals and / or music recorded in the vehicle interior, a video V with highlights in the driving operation of vehicle 1 is created based on recorded vehicle dynamics and / or object classification, for example, another overtaken vehicle, and / or scene detection, i.e., environment detection, and / or other measured variables which in combination represent a relevant event, in particular a highlight, in the driving operation of vehicle 1. Reference symbol list 1 vehicle 2 image capture units 3 Sensors 4 Control unit V Video QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] DE 10 2019 132 716 A1

[0002]

Claims

[1] Method for creating a video (V) with highlights of a driving operation of a vehicle (1) with a plurality of image recording units (2) present on the vehicle, which continuously record image data during the driving operation of the vehicle (1), characterized by , that - a recording mode for creating the video (V) is activated when a predefined driving mode is selected or manually, - Image sequences with highlights of the driving operation are automatically selected depending on detected signals of driving dynamics and / or detected image data of an environment of the vehicle (1) and / or a facial expression and / or a gesture of a driver of the vehicle (1) by means of evaluation by a trained pattern recognition method. [2] Method according to claim 1, characterized bythat the climax is / is detected as a spin of vehicle wheels and / or a vibration exceeding a threshold value and / or a gradient value exceeding a gradient threshold value and / or an overtaking maneuver of the vehicle (1). [3] Method according to claim 1 or 2, characterized by that the recorded selected image sequences are automatically assembled in their chronological order after the end of the driving operation and are accompanied by acoustic signals and / or music recorded in the vehicle interior. [4] Method according to one of the preceding claims, characterized by that the video (V) is created in the vehicle (1) and / or in a central computer unit linked to the vehicle (1) and is made available to the driver by means of an application program.

Images