CN-115376083-B - Vehicle photographing method, device and system and vehicle

CN115376083BCN 115376083 BCN115376083 BCN 115376083BCN-115376083-B

Abstract

The application discloses a vehicle photographing method, device and system and a vehicle. The method comprises the steps of responding to a wake-up instruction of a target object, obtaining an environment image of an environment where the target object is located through a camera of a vehicle, wherein the environment image comprises environment information of the environment and the target object, obtaining feature information of the target object in the environment image, inquiring a sample picture from a database, wherein similarity between attribute information contained in the sample picture and the environment information and similarity between the attribute information and the feature information exceed a first threshold value, the attribute information comprises the environment information and the feature information, photographing the target object according to the sample picture to obtain the target picture, receiving target scores of the target object on the target picture, and learning photographing preference of the target object according to the target scores. The application solves the technical problems that the vehicle in the prior art can not shoot the user, the user is required to automatically adjust the position according to the shooting result in the shooting process, and the shooting preference of the user can not be learned.

Inventors

CHI XIAOMING
CHEN BO
TANG YONGJUN

Assignees

中国第一汽车股份有限公司

Dates

Publication Date: 20260512
Application Date: 20220816

Claims (8)

1. A method of photographing a vehicle, comprising: Responding to a wake-up instruction of a target object, and acquiring an environment image of an environment where the target object is located through a camera of a vehicle, wherein the environment image comprises environment information of the environment and the target object; Acquiring characteristic information of a target object in the environment image, and inquiring a sample picture from a database, wherein the similarity between attribute information contained in the sample picture and the environment information and the similarity between the attribute information and the characteristic information exceed a first threshold value, and the attribute information comprises the environment information and the characteristic information; photographing the target object according to the sample picture to obtain a target picture; receiving a target score of the target object on the target picture, and learning photographing preference of the target object according to the target score; The method comprises the steps of establishing a three-dimensional model according to environment information, determining corresponding imaging effects of a target object under different photographing parameters in the three-dimensional model according to photographing parameters of sample pictures to obtain a plurality of preview pictures, wherein the photographing parameters of the sample pictures at least comprise first position information of a person in the sample pictures, gesture information of the person in the sample pictures and photographing angles of the sample pictures; The method further comprises the steps of downloading the evaluation model when initializing a user program, collecting target scores of the target pictures, and updating the evaluation model according to the target scores, wherein the updated evaluation model is used for providing advice information with multiple dimensions for the target object, and the advice information comprises advice information used for adjusting the shooting angle, advice information used for adjusting the matching degree of the gesture of the target object and the background or the environment, and advice information used for adjusting the position of the target object in the background.
2. The method of claim 1, wherein prior to querying the sample graph from the database, the method further comprises: Acquiring a material picture and a corresponding picture score, wherein the material picture is a picture uploaded in advance, the picture score is a score of the material picture according to a preset standard, and the picture score is used for quantifying the degree that the picture accords with the aesthetic standard; training an evaluation model according to the material pictures and the picture scores, wherein the evaluation model is used for scoring the pictures; and scoring the material pictures in the material library according to the evaluation model, and storing the pictures with scores exceeding a second threshold value in a sample gallery.
3. The method of claim 2, wherein photographing the target object according to the photographing parameters in the first picture comprises: The photographing parameters of the first picture are obtained, wherein the photographing parameters of the first picture at least comprise second position information of a person in the first picture, gesture information of the person in the first picture and photographing angles of the first picture; and generating prompt information for adjusting the position of the target object or adjusting the position of the camera according to the photographing parameters of the first picture so as to photograph the target object.
4. A method according to claim 3, wherein adjusting the position of the camera or the position of the target object comprises: adjusting the shooting angle of the camera according to the shooting parameters of the first picture; Generating a shooting suggestion according to the prompt information or the second position information of the person in the first picture and the gesture information of the person in the first picture, and playing the voice information generated according to the shooting suggestion.
5. The method of claim 3, wherein the adjusting the position of the camera or the position of the target object is performed in accordance with a priority of a rotation angle of the camera, a horizontal movement position of the vehicle, and a horizontal movement position of the target object.
6. An apparatus for photographing a vehicle, comprising: The acquisition module is used for responding to a wake-up instruction of a target object and acquiring an environment image of an environment where the target object is located through a camera of a vehicle, wherein the environment image comprises environment information of the environment and the target object; The query module is used for acquiring the characteristic information of the target object in the environment image and querying a sample picture from a database, wherein the similarity between attribute information contained in the sample picture and the environment information and the similarity between the attribute information and the characteristic information exceed a first threshold value, and the attribute information comprises the environment information and the characteristic information; The photographing module is used for photographing the target object according to the sample picture to obtain a target picture; the receiving module is used for receiving the target score of the target object on the target picture and learning the photographing preference of the target object according to the target score; The photographing module is further used for establishing a three-dimensional model according to the environment information, determining corresponding imaging effects of the target object under different photographing parameters in the three-dimensional model according to photographing parameters of the sample picture to obtain a plurality of preview pictures, wherein the photographing parameters of the sample picture at least comprise first position information of a person in the sample picture, gesture information of the person in the sample picture and photographing angles of the sample picture; The vehicle photographing device is further used for downloading the evaluation model when a user program is initialized, collecting target scores of the target pictures, and updating the evaluation model according to the target scores, wherein the updated evaluation model is used for providing multiple-dimension suggestion information for the target object, the suggestion information comprises suggestion information used for adjusting the photographing angle, suggestion information used for adjusting matching degree of the gesture of the target object and the background or the environment, and suggestion information used for adjusting the position of the target object in the background.
7. A vehicle photographing system is characterized by comprising a vehicle, a camera, a voice interaction unit, a three-dimensional simulation unit, an evaluation unit and a processing unit, wherein, The vehicle is used for responding to a wake-up instruction of a target object and starting a photographing function of the vehicle; The camera is arranged on the vehicle and used for shooting the target object to obtain an environment image of the environment where the target object is located; the three-dimensional simulation unit is used for establishing a three-dimensional model according to the environmental information in the environmental image; the voice interaction unit is used for notifying the target object of a shooting suggestion, wherein the shooting suggestion comprises shooting position information and shooting gesture information; The evaluation unit is used for scoring the pictures shot by the camera; The processing unit is used for acquiring an environment image of an environment where the target object is located, wherein the environment image comprises environment information of the environment and the target object, acquiring characteristic information of the target object in the environment image, inquiring a sample picture from a database, wherein the similarity of attribute information contained in the sample picture, the environment information and the characteristic information exceeds a first threshold value, the attribute information comprises the environment information and the characteristic information; The vehicle photographing system is further used for establishing a three-dimensional model according to the environment information, determining corresponding imaging effects of the target object under different photographing parameters in the three-dimensional model according to photographing parameters of the sample picture to obtain a plurality of preview pictures, wherein the photographing parameters of the sample picture at least comprise first position information of a person in the sample picture, gesture information of the person in the sample picture and photographing angles of the sample picture; And updating the evaluation model according to the target score, wherein the updated evaluation model is used for providing advice information with multiple dimensions for the target object, and the advice information comprises advice information for adjusting the shooting angle, advice information for adjusting the matching degree of the gesture of the target object and the background or the environment, and advice information for adjusting the position of the target object in the background.
8. A vehicle characterized in that a controller of the vehicle is adapted to perform a method of photographing the vehicle according to any one of claims 1 to 5.

Description

Vehicle photographing method, device and system and vehicle Technical Field The application relates to the field of artificial intelligence, in particular to a vehicle photographing method, device and system and a vehicle. Background Along with the progress of technology, people have more and more diversified requirements for recording life of the people, and along with wider application of automobiles in life of the people, the photographing function of the automobile can become a supplement for photographing in life of the people. For example, when we want the vehicle to take a picture automatically for us while traveling, and offer us a picture advice such as where to stand, what pose to put, what light to select, etc. The prior art does not have better self-timer at present, and especially the vehicle does not possess this function at all, and the cell-phone is although can carry out the self-timer, but the self-timer mode of cell-phone is unsuitable for carrying out the long-distance self-timer, and it is restricted to take full-length photograph, and in addition, present self-timer, the user need to take a photograph to accomplish and look over shooting effect, goes to look over again after according to shooting effect adjustment position, and the process is complicated and extravagant a large amount of time. In view of the above problems, no effective solution has been proposed at present. Disclosure of Invention The embodiment of the application provides a vehicle photographing method, device and system and a vehicle, which at least solve the technical problems that the vehicle in the prior art cannot photograph a user, the user is required to adjust the position according to the photographing result in the photographing process, and the photographing preference of the user cannot be learned. According to one aspect of the embodiment of the application, a vehicle photographing method is provided, which comprises the steps of responding to a wake-up instruction of a target object, obtaining an environment image of an environment where the target object is located through a camera of the vehicle, wherein the environment image comprises environment information of the environment and the target object, obtaining characteristic information of the target object in the environment image, inquiring a sample picture from a database, wherein similarity between attribute information contained in the sample picture and the environment information and similarity between the attribute information and the characteristic information exceed a first threshold value, photographing the target object according to the sample picture to obtain the target picture, receiving target scores of the target object on the target picture, and learning photographing preference of the target object according to the target scores. Optionally, before inquiring the sample picture from the database, the method further comprises the steps of obtaining a material picture and a corresponding picture score, wherein the material picture is a picture uploaded in advance, the picture score is a score of the material picture according to a preset standard, the picture score is used for quantifying the degree that the picture accords with the aesthetic standard, training an evaluation model according to the material picture and the picture score, wherein the evaluation model is used for scoring the picture, scoring the material picture in the material database according to the evaluation model, and storing the picture with the score exceeding a second threshold in the sample picture library. The method comprises the steps of establishing a three-dimensional model according to environment information, determining corresponding imaging effects of a target object under different photographing parameters in the three-dimensional model according to photographing parameters of sample pictures, and obtaining a plurality of preview pictures, wherein the photographing parameters of the sample pictures at least comprise first position information of characters in the sample pictures, gesture information of the characters in the sample pictures and photographing angles of the sample pictures, scoring the preview pictures according to an evaluation model, selecting a picture corresponding to the highest score from scoring results as a first picture, and photographing the target object according to photographing parameters in the first picture. Optionally, photographing the target object according to photographing parameters in the first picture, wherein the photographing parameters of the first picture at least comprise second position information of the person in the first picture, posture information of the person in the first picture and photographing angles of the first picture, and generating prompt information for adjusting the position of the target object or adjusting the position of a camera according to the photographing parameters of the first pi