Video generation method and device for realizing picture rotation, and vehicle

Abstract

The invention provides a video generation method and device for realizing picture rotation and a vehicle, which are used for generating a continuous video similar to video shooting by using rotation of a mobile device by using videos shot by vehicle-mounted cameras mounted in multiple directions on the vehicle. The video generation method for realizing picture rotation comprises the following steps: in response to a vehicle-mounted camera video synthesis request of a user, obtaining initial videos collected by a plurality of vehicle-mounted cameras; the plurality of vehicle-mounted cameras are respectively mounted in different directions of a vehicle and are used for covering a panoramic view angle around the vehicle; sequentially extracting a video sequence with a preset duration from the plurality of initial videos according to a preset sequence; combining the plurality of video sequences based on a time sequence to generate a continuous video forming a picture rotation effect; wherein the starting moment of the video sequence extracted from the initial video collected by the next vehicle-mounted camera is the last moment of the video sequence extracted from the initial video collected by the previous vehicle-mounted camera.

Description

Technical Field

[0001] The present application relates to the field of vehicle intelligent control, and specifically to a video generation method, device and vehicle for realizing picture rotation. Background Art

[0002] With the development of vehicle intelligence, in-vehicle camera systems have become an important part of modern vehicles. Generally, multiple cameras installed on a vehicle are mainly used for generating 360° images of the vehicle. This 360° image system simultaneously captures videos through multiple cameras and stitches these videos together to generate a panoramic image or video around the vehicle. This panoramic image system provides great convenience to the driver in scenarios such as vehicle parking and low-speed driving, helping them better observe the environment around the vehicle and avoid dangerous situations such as collisions.

[0003] However, the existing 360° image system mainly focuses on static panoramic display, and its functions are relatively single. During driving, users may need a richer visual experience, such as generating a moving effect similar to video shooting with a rotating mobile device using the videos captured by multiple in-vehicle cameras. This need has not been fully met in the prior art. The existing in-vehicle camera system cannot achieve a dynamic picture rotation effect similar to that of mobile device shooting, and users cannot obtain a continuous and smooth dynamic panoramic video through the in-vehicle camera system, which limits the functions and user experience of the in-vehicle camera system.

[0004] Therefore, a new technical solution is needed that can generate a continuous video with a picture rotation effect using the videos captured by in-vehicle cameras, thereby providing users with a richer and more dynamic visual experience. Summary of the Invention

[0005] The present invention provides a video generation method, device and vehicle for realizing picture rotation, which is used to generate a continuous video similar to video shooting with a rotating mobile device using the videos captured by in-vehicle cameras installed in multiple orientations on the vehicle.

[0006] The technical solution of the present invention is as follows: The present application provides a video generation method for realizing picture rotation, including: In response to a user's request for in-vehicle camera video synthesis, obtaining initial videos captured by multiple in-vehicle cameras; the multiple in-vehicle cameras are respectively installed in different orientations on the vehicle to cover the panoramic view around the vehicle; Sequentially extracting video sequences of a predetermined duration from the multiple initial videos in a preset order; Combining the multiple video sequences in chronological order to generate a continuous video with a picture rotation effect; Among them, the starting moment of the video sequence extracted from the initial video collected by the latter vehicle-mounted camera is the last moment of the video sequence extracted from the initial video collected by the former vehicle-mounted camera.

[0007] Preferably, the initial videos collected by multiple vehicle-mounted cameras are historical video data specified by the user.

[0008] Preferably, the initial video is video data collected in real time by the vehicle-mounted camera at the moment when the user initiates a video synthesis request.

[0009] Preferably, before the step of combining the multiple video sequences in chronological order to generate a continuous video with a picture rotation effect, the method further includes: Performing frame rate and resolution processing on all the extracted video sequences so that all the video sequences have the same frame rate and the same resolution.

[0010] Preferably, when the predetermined duration corresponds to N frame images and the number of vehicle-mounted cameras is M, the step of sequentially extracting video sequences of the predetermined duration from the multiple initial videos in a preset order includes: Taking one of the M vehicle-mounted cameras as the starting camera, and extracting the first N frame images to the Nth frame image of the initial video collected by the starting camera to form the first video sequence; In a clockwise or counterclockwise order, extracting the Nth frame image to the 2Nth frame image of the initial video collected by the next vehicle-mounted camera after the starting camera to form the second video sequence; and repeating the above steps until the video sequences are extracted from the initial videos collected by the remaining M - 2 vehicle-mounted cameras are completed, obtaining M video sequences; Among them, for the ith video sequence among the M video sequences, its starting frame image is the (i - 1)*Nth frame image in the ith initial video, and the last frame image is the i*nth frame image in the ith initial video.

[0011] Preferably, one vehicle-mounted camera selected by the user from multiple vehicle-mounted cameras is used as the starting camera.

[0012] Preferably, the method further includes: Adding video effects to the generated continuous video according to user settings.

[0013] Preferably, the vehicle-mounted camera video synthesis request carries the total video duration of the continuous video expected to be generated by the user, and the predetermined duration is the ratio of the total video duration to the number of vehicle-mounted cameras.

[0014] This application also provides a video generation device for realizing picture rotation, including: An initial video acquisition module, configured to acquire initial videos collected by a plurality of on-vehicle cameras in response to a user's request for synthesizing on-vehicle camera videos; the plurality of on-vehicle cameras are respectively installed at different positions of the vehicle to cover a panoramic view around the vehicle; A video sequence extraction module, configured to sequentially extract video sequences with a preset duration from the plurality of initial videos in a preset order; A continuous video generation module, configured to combine the plurality of video sequences in chronological order to generate a continuous video with a picture rotation effect; Wherein, the start time of the video sequence extracted from the initial video collected by the latter on-vehicle camera is the end time of the video sequence extracted from the initial video collected by the former on-vehicle camera.

[0015] This application also provides a vehicle, including the above-mentioned video generation device for realizing picture rotation.

[0016] The beneficial effects of the present invention are: By screening and processing the video data of multiple on-vehicle cameras according to certain rules, the initial videos collected by the multiple on-vehicle cameras can be fused into a new continuous video, solving the problem that the existing one-key video generation technology cannot be applied to the on-vehicle multi-camera scenario, and realizing the one-key video generation of the video data of multiple on-vehicle cameras in the on-vehicle scenario. At the same time, based on the screening and processing of the video data of multiple on-vehicle cameras, the initial videos of the multiple on-vehicle cameras are fused into a complete continuous video, and the finally generated target video can have a 360-degree dynamic perspective, making the target video obtained by one-key video generation more attractive. Description of the Drawings

[0017] Figure 1 It is a flowchart of the video generation method for realizing picture rotation in the embodiment of this application; Figure 2 It is a structural block diagram of the video generation device for realizing picture rotation in the embodiment of this application. Detailed Embodiments

[0018] The following further elaborates on the method of the present invention in combination with embodiments and the description of the drawings. This embodiment is implemented on the premise of the technical solution of the present invention, and gives detailed implementation manners and specific operation processes, but the protection scope of the present invention is not limited to the following embodiments.

[0019] Refer to Figure 1 , the embodiment of this application provides a video generation method for realizing picture rotation, including: S101, in response to a user's request for synthesizing on-vehicle camera videos, acquire initial videos collected by a plurality of on-vehicle cameras; the plurality of on-vehicle cameras are respectively installed at different positions of the vehicle to cover a panoramic view around the vehicle; S102. Extract video sequences of a preset duration from the multiple initial videos in sequence according to a preset order. S103. Combine the multiple video sequences in chronological order to generate a continuous video forming a picture rotation effect. Among them, the start time of the video sequence extracted from the initial video captured by the latter vehicle-mounted camera is the end time of the video sequence extracted from the initial video captured by the former vehicle-mounted camera.

[0020] Among them, the specific layout details of the multiple vehicle-mounted cameras shall be subject to the actual vehicle layout positions. For example, the horizontal field of view angle of each vehicle-mounted camera is not less than 100°, the frame rate of each vehicle-mounted camera is not less than 30fps, the resolution of each vehicle-mounted camera is not less than 1080P, and all vehicle-mounted cameras can cover a 360-degree range around the vehicle, while ensuring the clarity of the final synthesized video and the resolution meets the user's requirements. For example, in the embodiment of the present application, it is assumed that the vehicle is equipped with four vehicle-mounted cameras, which are respectively installed at the front, rear, left, and right of the vehicle to cover the panoramic view around the vehicle. Each camera has the ability to collect high-definition video and is connected to the central processing unit of the vehicle-mounted system through the vehicle-mounted network.

[0021] During driving, the user initiates a "vehicle-mounted camera video synthesis" request through the touch screen or voice assistant of the vehicle-mounted system. The user can choose to use the real-time collected video data or select the existing historical video data from the storage module of the vehicle-mounted system as the initial video. For example, the user can select "real-time shooting" or "select from historical videos" through the vehicle-mounted system interface.

[0022] In step S101, if the user selects "real-time shooting", all vehicle-mounted cameras will be started in real time to start collecting video data; if the user selects "select from historical videos", the user-specified historical video data will be retrieved from the storage module.

[0023] In step S102, the preset order is preset according to the installation position order of the vehicle-mounted cameras, such as clockwise order or counterclockwise order; after determining that the starting camera is the front camera, for example, the video sequences are extracted in sequence from "front camera → right camera → rear camera → left camera" in the clockwise direction.

[0024] Among them, in the embodiment of the present application, the starting camera can be a preset camera; it can also be determined by the user's input.

[0025] In step S102, the predetermined duration can be determined by the total length of the continuous video input by the user through the in-vehicle system interface. For example, the user inputs that "the total duration of the continuous video is 20 seconds". According to the total duration of the continuous video input by the user, a video sequence with a predetermined duration (i.e., 5S) is extracted from each initial video.

[0026] In S103, during the process of combining to generate a continuous video, the start time of the video sequence extracted from the initial video captured by the rear in-vehicle camera is the last time of the video sequence extracted from the initial video captured by the previous in-vehicle camera. For example: The duration of the video sequence extracted by the front camera is from 0 to 5 seconds; The duration of the video sequence extracted by the right camera is from 5 to 10 seconds; The duration of the video sequence extracted by the rear camera is from 10 to 15 seconds; The duration of the video sequence extracted by the left camera is from 15 to 20 seconds.

[0027] In this way, the generated continuous video visually presents an effect similar to that of rotating shooting with a mobile device, and the picture smoothly transitions between cameras in different orientations.

[0028] The generated continuous video is stored in the memory of the in-vehicle system, and the user can browse and playback it at any time through the screen of the in-vehicle system.

[0029] In the embodiment of the present application, considering that the hardware parameters of in-vehicle cameras in different orientations are inconsistent, therefore, before the step of combining multiple video sequences based on the time sequence to generate a continuous video with a picture rotation effect, this video generation method in the embodiment of the present application further includes: S104, perform frame rate and resolution processing on all the extracted video sequences to make the frame rates of all video sequences the same and make the resolutions of all video sequences the same.

[0030] In the embodiment of the present application, when performing frame rate and resolution processing on all video sequences, for example, select the minimum frame rate Z and the minimum resolution Y among all the video sequences as the standard processing parameters. Finally, the frame rate of all video sequences after being processed by the normalization algorithm is Z, and the resolution is Y, realizing the unification of the frame rates and resolutions of all video sequences and avoiding the video picture incoordination caused by inconsistent video sequence parameters.

[0031] Videos captured by different cameras may have different frame rates. In order to make the frame rates of all video sequences consistent, it is necessary to adjust the frame rates. The specific method is as follows: Select a target frame rate, which can usually be the highest frame rate among all cameras or a preset common frame rate (e.g., 30fps or 60fps). For example, if the frame rates of four cameras are 30fps, 24fps, 30fps, and 25fps respectively, 30fps can be selected as the target frame rate.

[0032] For video sequences with a frame rate lower than the target frame rate, increase the number of frames through interpolation methods to reach the target frame rate. Interpolation methods include, for example, linear interpolation, bicubic interpolation, or more advanced motion compensation interpolation. For example, for a video sequence with a frame rate of 24fps, its frame rate needs to be increased to 30fps through interpolation methods.

[0033] For video sequences with a frame rate higher than the target frame rate, reduce the number of frames through frame dropping methods to reach the target frame rate.

[0034] Videos captured by different cameras may have different resolutions. To make the resolutions of all video sequences consistent, the resolutions need to be adjusted. The specific methods are as follows: Select a target resolution, which can usually be the highest resolution among all cameras or a preset common resolution (e.g., 1920x1080 or 1280x720). For example, if the resolutions of four cameras are 1920x1080, 1280x720, 1920x1080, and 1280x720 respectively, 1920x1080 can be selected as the target resolution.

[0035] For video sequences with a resolution lower than the target resolution, increase the resolution through upscaling methods. Upscaling methods include bilinear interpolation, bicubic interpolation, or more advanced super-resolution algorithms. For example, for a video sequence with a resolution of 1280x720, its resolution needs to be increased to 1920x1080 through upscaling methods.

[0036] For video sequences with a resolution higher than the target resolution, reduce the resolution through downscaling methods. For example, for a video sequence with a resolution of 1920x1080, if the target resolution is 1280x720, its resolution needs to be reduced to 1280x720 through downscaling methods.

[0037] After processing the frame rate and resolution of all video sequences, the frame rates of all video sequences are the same and the resolutions of all video sequences are made the same. Assume that in the embodiment of the present application, a predetermined duration corresponds to N frame images. When the number of vehicle-mounted cameras is M, the step S102 of sequentially extracting video sequences of the predetermined duration from a plurality of the initial videos in a preset order includes: S1021, taking one of the M vehicle-mounted cameras as the starting camera, extracting the first frame image to the Nth frame image of the initial video collected by the starting camera to form the first video sequence; S1022, in a clockwise or counterclockwise order, extracting the Nth frame image to the 2Nth frame image of the initial video collected by the next vehicle-mounted camera after the starting camera to form the second video sequence; and repeating the above steps until the video sequences are extracted from the initial videos collected by the remaining M - 2 vehicle-mounted cameras are completed, obtaining M video sequences; Among them, for the i-th video sequence among the M video sequences, its starting frame image is the (i - 1)*Nth frame image in the i-th initial video, and the last frame image is the i*nth frame image in the i-th initial video.

[0038] In the embodiment of the present application, after generating the continuous video, the method further includes: S105, adding video special effects to the generated continuous video according to user settings. These special effects include but are not limited to: color adjustment, transition effects, dynamic effects, text or graphic overlay, and audio effects, etc.

[0039] In the embodiment of the present application, the user can transmit the generated continuous video to a mobile device or other external devices through the wireless communication module (such as Wi-Fi, Bluetooth, etc.) of the vehicle-mounted system, facilitating the user to share it with others.

[0040] Assume that when the user is parking and wants to record the situation around the vehicle, the "real-time shooting" mode is selected. The vehicle-mounted system starts all cameras and collects a 1-minute video. The user sets the total duration of the continuous video to 40s through the vehicle-mounted system interface, and it can be determined that the predetermined duration of each video sequence is 10 seconds. The system sequentially extracts 10-second video sequences from four cameras in a clockwise direction and combines them. The final generated continuous video has a duration of 40 seconds, and the user can clearly see the smooth rotation of the images of the surrounding environment of the vehicle between different orientations, as if the user is holding a mobile device and shooting around the vehicle.

[0041] Referring to Figure 2 , the embodiment of the present application further provides a video generation device for realizing picture rotation, including: The initial video acquisition module 201 is configured to acquire initial videos collected by multiple vehicle-mounted cameras in response to a user's request for synthesizing vehicle-mounted camera videos; the multiple vehicle-mounted cameras are respectively installed at different positions of the vehicle and are used to cover the panoramic view around the vehicle. The video sequence extraction module 202 is configured to sequentially extract video sequences of a preset duration from the multiple initial videos in a preset order. The continuous video generation module 203 is configured to combine the multiple video sequences in chronological order to generate a continuous video with a picture rotation effect. Wherein, the start time of the video sequence extracted from the initial video collected by the latter vehicle-mounted camera is the end time of the video sequence extracted from the initial video collected by the former vehicle-mounted camera.

[0042] The present application also provides a vehicle, including the above video generation device for realizing picture rotation.

[0043] By screening and processing the video data of multiple vehicle-mounted cameras according to certain rules, the initial videos collected by multiple vehicle-mounted cameras can be fused into a new continuous video, solving the problem that the existing one-key video generation technology cannot be applied to the vehicle-mounted multi-camera scenario, and realizing the one-key video generation of the video data of multiple vehicle-mounted cameras in the vehicle-mounted scenario. At the same time, based on the screening and processing of the video data of multiple vehicle-mounted cameras, the initial videos of multiple vehicle-mounted cameras are fused into a complete continuous video, and the finally generated target video can have a 360-degree dynamic perspective, making the target video obtained by one-key video generation more attractive.

[0044] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts among the various embodiments can be referred to each other.

[0045] Although the preferred embodiments of the embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications once they know the basic creative concepts. Therefore, the appended claims are intended to be construed to include the preferred embodiments and all changes and modifications falling within the scope of the embodiments of the present invention.

[0046] It should also be noted that in this text, terms such as "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings. This is for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or component referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation on the present invention. In addition, relative terms such as "first" and "second" are used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations, nor can they be construed as indicating or implying relative importance. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or terminal device comprising a series of elements does not include those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, article or terminal device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the existence of additional identical elements in the process, method, article or terminal device comprising the element.

[0047] The technical solutions provided by the present invention have been introduced in detail above. Specific examples are used in this text to elaborate on the principles and implementation manners of the present invention. The description of the above embodiments is only for helping to understand the present invention, and the content of this specification should not be construed as a limitation on the present invention. At the same time, for those of ordinary skill in the art, according to the present invention, there will be various changes in the specific implementation manners and application scopes. It is not necessary and impossible to enumerate all the implementation manners here, and the obvious changes or variations derived therefrom are still within the protection scope of the present invention.

Claims

1. A video generation method for realizing screen rotation, characterized in that include: In response to a user's vehicle camera video synthesis request, obtaining initial videos captured by multiple vehicle cameras; Multiple on-board cameras are installed at different positions of the vehicle to cover a panoramic view around the vehicle; Extracting video sequences of a predetermined length from the plurality of initial videos in sequence according to a preset order; Combining the plurality of video sequences based on a time sequence to generate a continuous video with a picture rotation effect; The starting moment of the video sequence extracted from the initial video captured by the latter vehicle-mounted camera is the last moment of the video sequence extracted from the initial video captured by the previous vehicle-mounted camera.

2. The video generation method for realizing screen rotation according to claim 1, wherein The initial video collected by multiple vehicle-mounted cameras is historical video data specified by the user.

3. The video generation method for realizing screen rotation according to claim 1, characterized in that, The initial video is video data collected in real time by the vehicle-mounted camera when the user initiates a video synthesis request.

4. The video generation method for realizing screen rotation according to claim 1, characterized in that Before the step of combining the plurality of video sequences based on the time sequence to generate a continuous video with a picture rotation effect, the method further comprises: The frame rates and resolutions of all extracted video sequences are processed to make the frame rates of all video sequences the same and the resolutions of all video sequences the same.

5. The video generation method for realizing screen rotation according to claim 4, wherein The predetermined time length corresponds to N frames of images. When the number of vehicle-mounted cameras is M, the step of extracting video sequences of the predetermined time length from the plurality of initial videos in a preset order comprises: Taking one of the M vehicle-mounted cameras as a starting camera, extracting the first frame image to the Nth frame image of the initial video captured by the starting camera to form a first video sequence; Extract the Nth frame image to the 2Nth frame image of the initial video captured by the next vehicle-mounted camera after the starting camera in a clockwise or counterclockwise order to form a second video sequence; and repeat the above steps until the video sequences are extracted from the initial videos captured by the remaining M-2 vehicle-mounted cameras to obtain M video sequences; Among them, for the i-th video sequence in the M video sequences, its starting frame image is the (i-1)*N-th frame image in the i-th initial video, and the last frame image is the i*n-th frame image in the i-th initial video.

6. The video generation method for realizing screen rotation according to claim 1, wherein A vehicle camera selected by a user from multiple vehicle cameras is used as a starting camera.

7. The video generation method for realizing screen rotation according to claim 1, characterized in that, The method further comprises: According to user settings, video effects are added to the generated continuous video.

8. The video generation method for realizing screen rotation according to claim 1, wherein, The vehicle-mounted camera video synthesis request carries the total video duration of the continuous video that the user expects to generate, and the predetermined duration is the ratio of the total video duration to the vehicle-mounted camera.

9. A video generation device for realizing screen rotation, characterized in that, include: An initial video acquisition module, used to respond to a user's vehicle camera video synthesis request and acquire initial videos captured by multiple vehicle cameras; Multiple on-board cameras are installed at different positions of the vehicle to cover a panoramic view around the vehicle; A video sequence extraction module, used to extract video sequences of a predetermined length from the plurality of initial videos in sequence according to a preset order; A continuous video generation module, used for combining the plurality of video sequences based on a time sequence to generate a continuous video with a picture rotation effect; Among them, the starting moment of the video sequence extracted from the initial video collected by the latter vehicle-mounted camera is the last moment of the video sequence extracted from the initial video collected by the former vehicle-mounted camera.

10. A vehicle, characterized in that, It includes the video generation device for realizing picture rotation described in claim 9.

Images