CN113567149B - User shooting intention recognition method, device and equipment - Google Patents
User shooting intention recognition method, device and equipment Download PDFInfo
- Publication number
- CN113567149B CN113567149B CN202110845857.3A CN202110845857A CN113567149B CN 113567149 B CN113567149 B CN 113567149B CN 202110845857 A CN202110845857 A CN 202110845857A CN 113567149 B CN113567149 B CN 113567149B
- Authority
- CN
- China
- Prior art keywords
- shooting
- images
- image
- position information
- intention
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000001514 detection method Methods 0.000 claims abstract description 116
- 230000006378 damage Effects 0.000 claims description 94
- 238000012216 screening Methods 0.000 claims description 21
- 230000009466 transformation Effects 0.000 claims description 10
- 238000004364 calculation method Methods 0.000 claims description 5
- 238000004422 calculation algorithm Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 6
- 230000008859 change Effects 0.000 description 4
- 230000003902 lesion Effects 0.000 description 4
- 238000010801 machine learning Methods 0.000 description 4
- 230000000007 visual effect Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 210000004899 c-terminal region Anatomy 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000000087 stabilizing effect Effects 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M17/00—Testing of vehicles
- G01M17/007—Wheeled or endless-tracked vehicles
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the specification provides a method, a device and equipment for identifying shooting intention of a user, wherein in the method for identifying the shooting intention of the user, a group of shooting images of a vehicle are acquired, and the group of shooting images comprises at least two shooting images. Position information and attitude information of the photographing apparatus at the time of photographing each of the group of photographed images are determined. The components of the vehicle in each photographed image are detected, thereby obtaining the component detection result of each photographed image. The position information, the posture information, and the component detection result are input into a user photographing intention recognition model to recognize photographing intentions of the respective photographed images.
Description
The application relates to a divisional application of an application patent application with the application number 201811023190.3 of 201811023190.3, named as a method, a device and equipment for identifying shooting intention of a user, which is filed on the 09 th and 03 th of 2018.
Technical Field
One or more embodiments of the present disclosure relate to the field of computer technologies, and in particular, to a method, an apparatus, and a device for identifying a shooting intention of a user.
Background
In the conventional technology, the damage assessment process of the vehicle can be as follows: a series of captured images taken by a user for a vehicle to be damaged is acquired. The series of captured images are input into a damage detection model, thereby obtaining a damage detection result of the vehicle. However, when the series of photographed images are photographed, the user may move the position of the camera or change the posture of the camera, and the photographed images may not include a lesion during the movement of the position or change of the posture of the user.
Therefore, it is necessary to provide a scheme for identifying the shooting intention of the user so that the accuracy of the damage detection result can be improved when the damage of the vehicle is detected based on the shooting intention of the user.
Disclosure of Invention
One or more embodiments of the present disclosure describe a method, an apparatus, and a device for identifying a user's shooting intention, which can identify the user's shooting intention, so that when detecting a damage to a vehicle based on the user's shooting intention, the accuracy of the damage detection result can be improved.
In a first aspect, a method for identifying a shooting intention of a user is provided, including:
Acquiring a group of photographed images of the vehicles; the group of shot images comprises at least two shot images;
determining position information and posture information of the photographing device when photographing each photographed image in the group of photographed images;
detecting the parts of the vehicle in each photographed image, thereby obtaining a part detection result of each photographed image;
and inputting the position information, the gesture information and the component detection result into a user shooting intention recognition model to recognize the shooting intention of each shooting image.
In a second aspect, there is provided an apparatus for recognizing a photographing intention of a user, including:
an acquisition unit configured to acquire a group of photographed images of a vehicle; the group of shot images comprises at least two shot images;
A determining unit configured to determine position information and posture information of a photographing apparatus when photographing each of the group of photographed images acquired by the acquiring unit;
A detection unit configured to detect a component of the vehicle in each of the captured images acquired by the acquisition unit, thereby obtaining a component detection result of each of the captured images;
And a recognition unit configured to input the position information, the posture information, and the component detection result detected by the detection unit, which are determined by the determination unit, into a user photographing intention recognition model to recognize a photographing intention of each of the photographed images.
In a third aspect, there is provided an apparatus for recognizing a user's photographing intention, comprising:
A memory;
One or more processors; and
One or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, which when executed by the processors implement the steps of:
Acquiring a group of photographed images of the vehicles; the group of shot images comprises at least two shot images;
determining position information and posture information of the photographing device when photographing each photographed image in the group of photographed images;
detecting the parts of the vehicle in each photographed image, thereby obtaining a part detection result of each photographed image;
and inputting the position information, the gesture information and the component detection result into a user shooting intention recognition model to recognize the shooting intention of each shooting image.
According to the method, the device and the equipment for identifying the shooting intention of the user, a group of shooting images of the vehicle are obtained, and the group of shooting images comprises at least two shooting images. Position information and attitude information of the photographing apparatus at the time of photographing each of the group of photographed images are determined. The components of the vehicle in each photographed image are detected, thereby obtaining the component detection result of each photographed image. The position information, the posture information, and the component detection result are input into a user photographing intention recognition model to recognize photographing intentions of the respective photographed images. Thereby, recognition of the user's photographing intention is achieved. In addition, after the shooting intention of the user is identified, the damage of the vehicle can be detected based on the shooting intention of the user, so that the accuracy of the damage detection result can be improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present description, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present description, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic diagram of a vehicle damage assessment system provided herein;
FIG. 2 is a flowchart of a method for identifying a user's shooting intention according to an embodiment of the present disclosure;
fig. 3 is a flowchart of a method for determining position and posture information of a photographing apparatus provided in the present specification;
FIG. 4 is a flow chart of a method for detecting damage to a vehicle according to the present disclosure;
FIG. 5 is a flowchart of another method for detecting damage to a vehicle according to the present disclosure;
Fig. 6 is a schematic diagram of a device for identifying a shooting intention of a user according to an embodiment of the present disclosure;
fig. 7 is a schematic diagram of a device for identifying a user's shooting intention according to an embodiment of the present disclosure.
Detailed Description
The following describes the scheme provided in the present specification with reference to the drawings.
The method for identifying the shooting intention of the user according to one or more embodiments of the present disclosure may be applied to the vehicle damage assessment system 10 shown in fig. 1, where in fig. 1, the vehicle damage assessment system 10 may include: the user captures the intent recognition module 102 and the vehicle damage detection module 104.
The user photographing intention recognition module 102 is used to acquire a series of photographed images of the vehicle photographed by the data acquirer (including C-terminal user, loss fighter of insurance company, etc.) through the photographing apparatus. In this specification, the series of captured images may also be referred to as a group of captured images. The set of captured images may correspond to a case. The above-described group of captured images generally have continuity, and the captured images having continuity may have overlapping (overlapping) areas therebetween. After the above-described set of captured images is acquired, position information and attitude information of the capturing device at the time of capturing each of the set of captured images may be determined. And detecting the parts of the vehicle in each photographed image to obtain a part detection result of each photographed image. Finally, the shooting intention of each shooting image is identified according to the position information, the gesture information and the component detection result when the shooting device shoots each shooting image.
The vehicle damage detection module 104 is configured to determine a damage detection result of the vehicle according to each photographed image and the photographing intention of each photographed image. In one implementation, the initial damage detection result of the vehicle determined based on each captured image may be adjusted with the capture intent of each captured image displayed as a weight. In another implementation, each captured image and the capturing intention of each captured image may also be input into the damage detection model to determine the damage detection result of the vehicle. In still another implementation, the key captured images may be screened based on information such as the capturing intention of each captured image. And then determining a damage detection result of the vehicle based on the key shot image.
The vehicle damage detection module 104 provided in the present disclosure can improve accuracy of damage detection results when detecting damage to a vehicle based on a shooting intention of a shooting image. In addition, the vehicle damage detection module 104 can save computing resources and improve damage assessment efficiency of the vehicle damage assessment system 10 when detecting damage to the vehicle based on the key captured image.
It should be appreciated that the vehicle damage assessment system 10 shown in fig. 1 is merely illustrative and that the system according to embodiments of the present disclosure is not limited to the configuration shown in fig. 1. For example, a shunt module may be further included in fig. 1, which is not limited in this specification.
Fig. 2 is a flowchart of a method for identifying a shooting intention of a user according to an embodiment of the present disclosure. The subject of execution of the method may be a device with processing capabilities: the server or system or module, for example, may capture the intent recognition module 102 for the user in FIG. 1. As shown in fig. 2, the method specifically may include:
step 202, a set of captured images of a vehicle is acquired.
The set of captured images may be obtained by a data collector (including the C-terminal user, and the nominated person of the insurance company, etc.) through the capture device when capturing the damaged portion of the vehicle. Since there may be a plurality of damaged portions of the vehicle, the data collector may move the position of the photographing apparatus or change the posture of the photographing apparatus during photographing, so that each photographed image of the set of photographed images has continuity. The captured images having continuity may have overlapping (overlapping) areas therebetween. In this specification, a group of photographed images may include at least two photographed images.
In step 204, position information and pose information of the photographing apparatus at the time of photographing each of the group of photographed images are determined.
In this specification, the positional information of the photographing apparatus may refer to three-dimensional coordinates of the photographing apparatus in a three-dimensional space, and the posture information of the photographing apparatus may refer to information of a rotation angle or the like of the photographing apparatus.
In one implementation, the above-described position information and attitude information may be determined in a two-by-two combination. There may be a region of coincidence between the combined captured images. In one example, the method for determining the position information and the posture information may be as shown in fig. 3. In fig. 3, the determining method may include the steps of:
step 302, selecting two shot images with overlapping areas from a group of shot images.
The two captured images here may constitute one of the combinations described above. It should be noted that the two photographed images may be adjacent to each other, for example, the 1 st image and the 2 nd image; the 1 st image and the 3 rd image may not be adjacent, for example, as long as there is a region of overlap between the two captured images.
And step 304, extracting key points on the two photographed images respectively.
The keypoints herein may have corresponding positional information in the captured image, such as two-dimensional coordinates: (x, y). In addition, the key points may correspond to a physical object (e.g., a wheel), so that it may have visual characteristic information such as color, texture, angle, etc.
And 306, performing feature matching on the key points on the two shot images, so as to obtain a plurality of groups of successfully matched target key points.
The feature matching may refer to comparing visual feature information of two key points respectively located on two photographed images. It is understood that a set of target keypoints that are successfully matched may refer to two keypoints that are identical in the corresponding physical object.
When the target key points which are successfully matched pairwise are multiple groups, multiple groups of target key points can be formed.
Step 308, determining a transformation relationship between the two photographed images according to the position information of the multiple groups of target key points in the two photographed images.
The transformation relationship here may be, for example: "move from left to right", "rotate left by x degrees" and "rotate right by x degrees" etc. In one example, the position change condition of the corresponding real object can be determined by comparing two-dimensional coordinates of two target key points in each group on the two photographed images. If the corresponding object moves from the middle position of one image to the left position of the other image, the transformation relationship between the two photographed images may be: "move left to right".
Step 310, estimating the position information of the multiple groups of target key points in the three-dimensional space according to the position information and the transformation relation.
Here, the position information of the multiple sets of target key points in the three-dimensional space may be estimated by combining a binocular positioning algorithm or a three-dimensional reconstruction algorithm. It should be noted that, the position information of the target key point in the three-dimensional space is the position information of the corresponding object. Because the real objects corresponding to the two target key points in one group are the same, the position information of the two target key points in the three-dimensional space is the same. In one example, the location information of a set of target keypoints in three-dimensional space may be represented as: three-dimensional coordinates: (x, y, z).
In step 312, the position information of the multiple sets of target key points in each captured image and the position information in the three-dimensional space are subjected to re-projection calculation to determine the position information and the posture information of the capturing device when capturing the two captured images.
For example, the position information of multiple groups of target key points in one shooting image and the position information in the three-dimensional space can be subjected to re-projection calculation to determine the position information and the posture information of the shooting device when the shooting image is shot; then, according to the method, the position information and the posture information of the shooting equipment when shooting another shooting image are determined.
The above steps 302-312 are repeatedly performed until the position information and the posture information of the photographing apparatus at the time of photographing each photographed image are determined.
Returning to fig. 2, fig. 2 may further include the steps of:
Step 206, detecting the components of the vehicle in each photographed image, thereby obtaining the component detection result of each photographed image.
Specifically, the detection of the components of the vehicle in each captured image may be performed according to an object detection algorithm. The target detection algorithms herein may include, but are not limited to, fast (Faster) -area based convolutional neural networks (Region-based Convolutional Neural Network, RCNN), area based full convolutional networks (Region-based Fully Convolutional Network, RFCN), single-shot multiple bounding box detectors (Single Shot MultiBox Detector, SSD), and YOLO, among others.
Step 208, the position information, the gesture information, and the component detection result are input into the user photographing intention recognition model to recognize the photographing intention of each photographed image.
The user shooting intention recognition model may be obtained by training a machine learning algorithm according to a motion track of the shooting device when shooting a plurality of groups of shooting images and a component detection result of the plurality of groups of shooting images. The motion trajectory of the photographing apparatus when photographing a group of photographed images may refer to a sequence in which position information and posture information of the photographing apparatus when photographing each of the group of photographed images are combined.
It should be noted that, the principle of the user shooting intention recognition model recognizing the user shooting intention may be explained as follows: first, based on the movement locus of the photographing apparatus, the movement condition of the photographing apparatus, for example, a backward movement of 1 meter or the like, may be determined. Then, by combining the component detection results in the photographed image, the following photographing intention can be obtained: "near the door", "far from the right rear tire" and "stable at the head", etc.
Of course, in practical application, after the shooting intention of the user is identified, the motion trail of the shooting device can be further combined to expand the shooting intention. For example, the extended shooting intention of the user may be: shooting the whole car map, shooting the part map, and shooting the damage detail map. It will be appreciated that the expanded user's capture intent is typically relative to a plurality of captured images.
In summary, the above embodiments of the present disclosure can quickly and accurately identify the shooting intention of the user by using the pre-constructed user shooting intention identification model.
After the shooting intention of each shooting image is recognized, the damage of the vehicle can be detected based on the shooting intention, so that the accuracy of the damage detection result can be improved.
Fig. 4 is a flowchart of a method for detecting damage to a vehicle provided in the present specification. As shown in fig. 4, the method may include the steps of:
steps 402 to 408 are the same as steps 202 to 208.
Step 410, determining a damage detection result of the vehicle according to each photographed image and the photographing intention of each photographed image.
In one implementation, the determining process of the damage detection result of the vehicle may be: and inputting each photographed image into a damage detection model to obtain an initial damage detection result corresponding to each photographed image. And determining the weight of the corresponding initial damage detection result according to the shooting intention of each shooting image. And determining the damage detection result of the vehicle according to the initial damage detection result and the corresponding weight.
For example, when the shooting intention of the user is recognized: when shooting a specific part repeatedly from different positions and angles in close proximity, the probability of damage on the part tends to be large, and therefore the damage detection result on the part can be given a large weight. Conversely, when the shooting intention of the user is identified: when moving from one position to another position relatively far away and then staying for detail image shooting, a new lesion is usually shot, and a greater weight can be given to a lesion detection result obtained at that time. The images shot during the movement process may not contain damage, if damage is detected from the shot images, the damage is likely to belong to false detection, and the corresponding weight can be reduced.
In summary, by combining the weights described above, more accurate damage detection results can be obtained.
In another implementation, the determining process of the damage detection result of the vehicle may be: and inputting each photographed image and the photographing intention of each photographed image into the damage detection model to determine a damage detection result of the vehicle. The captured image and the captured intention are input as features to the damage detection model.
The damage detection model in the present specification may output a damage detection result of the corresponding vehicle based on the input captured image. It can be obtained by training using a large number of calibrated captured images including lesions. In addition, the damage detection result determined in the present specification may include at least one of the following information: a damaged member, a damaged position, a damaged type, a damaged degree, and the like.
The above-described embodiments of the present specification can improve the accuracy of the damage detection result by inputting the weight displayed as the shooting intention of the user or as the feature to the damage detection model.
Fig. 5 is a flowchart of another method for detecting damage to a vehicle according to the present disclosure. As shown in fig. 5, the method may include the steps of:
Steps 502 to 508 are the same as steps 202 to 208.
Step 510, screening each photographed image according to the predefined screening rule and the photographing intention of each photographed image, thereby obtaining the key photographed image.
Here, the predefined screening rules may be: "repeatedly taking images of the same part from a plurality of different distances and angles" or "after a movement, stabilizing the image of the camera shooting details", etc. Specifically, whether the shooting intention of each shooting image meets the rule can be judged, and if so, the shooting intention of each shooting image can be screened as a key shooting image; if not, the captured image may be ignored. For example, a captured image having a high probability of being discriminated as "a photographer is capturing a damage" may be selected as a key captured image, while a captured image having a low probability of being discriminated as "a photographer is capturing a damage" may be ignored.
It will be appreciated that the captured images that generally meet the above-described rules are highly likely to be injuries to the data collector when capturing the vehicle.
In other embodiments, each captured image and/or the captured intent and/or visual features and/or component detection results of each captured image may also be input into the machine learning model to screen key captured images.
Step 512, inputting the key shot image into the damage detection model to determine the damage detection result of the vehicle.
According to the embodiment of the specification, the method for screening the key shooting images and inputting the key shooting images into the damage identification model can avoid processing a large amount of invalid data, so that the computing resources can be saved, and the damage detection efficiency of the vehicle can be improved.
Corresponding to the above method for identifying a user shooting intention, an embodiment of the present disclosure further provides an apparatus for identifying a user shooting intention, as shown in fig. 6, where the apparatus may include:
An acquisition unit 602 for acquiring a group of photographed images of the vehicle. The group of photographed images includes at least two photographed images.
A determination unit 604 for determining positional information and attitude information of the photographing apparatus at the time of photographing each of the group of photographed images acquired by the photographing acquisition unit 602.
A detecting unit 606 for detecting the component of the vehicle in each of the captured images acquired by the acquiring unit 602, thereby obtaining a component detection result of each of the captured images.
And a recognition unit 608 for inputting the position information, the posture information, and the component detection result detected by the detection unit 606, which are determined by the determination unit 604, into the user photographing intention recognition model to recognize the photographing intention of each photographed image.
The user shooting intention recognition model is obtained by training a machine learning algorithm according to the motion trail of shooting equipment when shooting a plurality of groups of shooting images and the detection result of the parts of the plurality of groups of shooting images. Wherein the motion trail of the shooting device when shooting a group of shooting images is determined according to the position information and the gesture information of the shooting device when shooting each shooting image in the group of shooting images.
The determining unit 604 may specifically be configured to:
two shot images with overlapping areas are selected from a group of shot images.
And extracting key points on the two photographed images respectively.
And performing feature matching on the key points on the two shooting images so as to obtain a plurality of groups of successfully matched target key points.
And determining the position information and the posture information of the shooting equipment when shooting the two shooting images according to the position information of the multiple groups of target key points in the two shooting images.
The above steps are repeatedly executed until the position information and the posture information of the photographing apparatus at the time of photographing each photographed image are determined.
The determining unit 604 may be further specifically configured to:
And determining the transformation relation between the two shooting images according to the position information of the multiple groups of target key points in the two shooting images.
And estimating the position information of the multiple groups of target key points in the three-dimensional space according to the position information and the transformation relation.
And carrying out re-projection calculation on the position information of the multiple groups of target key points in each shooting image and the position information in the three-dimensional space so as to determine the position information and the posture information of the shooting equipment when shooting two shooting images.
Optionally, the determining unit 604 is further configured to determine a damage detection result of the vehicle according to each captured image and the capturing intention of each captured image.
The determining unit 604 may specifically be configured to:
And inputting each photographed image into a damage detection model to obtain an initial damage detection result corresponding to each photographed image.
And determining the weight of the corresponding initial damage detection result according to the shooting intention of each shooting image.
And determining the damage detection result of the vehicle according to the initial damage detection result and the corresponding weight.
Optionally, the determining unit 604 may be further specifically configured to:
and inputting each photographed image and the photographing intention of each photographed image into the damage detection model to determine a damage detection result of the vehicle.
Optionally, the apparatus may further include:
the first filtering unit 610 is configured to filter each of the captured images according to a predefined filtering rule and a capturing intention of each of the captured images, so as to obtain a key captured image.
The determining unit 604 is further configured to input the key captured images screened by the first screening unit 610 into a damage detection model, so as to determine a damage detection result of the vehicle.
Optionally, the apparatus may further include:
And a second screening unit 612 for inputting each photographed image and/or photographing intention and/or visual characteristics and/or component detection results of each photographed image into the machine learning model to screen the key photographed images.
The determining unit 604 is further configured to input the key captured images screened by the second screening unit 612 into a damage detection model, so as to determine a damage detection result of the vehicle.
The functions of the functional modules of the apparatus in the foregoing embodiments of the present disclosure may be implemented by the steps of the foregoing method embodiments, so that the specific working process of the apparatus provided in one embodiment of the present disclosure is not repeated herein.
In the recognition apparatus for a user's photographing intention provided in one embodiment of the present specification, the acquisition unit 602 acquires a group of photographed images of a vehicle. The determination unit 604 determines position information and attitude information of the photographing device at the time of photographing each of the group of photographed images. The detection unit 606 detects the components of the vehicle in each of the captured images, thereby obtaining the component detection results of each of the captured images. The recognition unit 608 inputs the position information, the posture information, and the component detection result into the user photographing intention recognition model to recognize the photographing intention of each photographed image. Thus, recognition of the user's photographing intention can be achieved.
The recognition device for a user's shooting intention provided in one embodiment of the present disclosure may be a sub-module or sub-unit of the user's shooting intention recognition module 102 in fig. 1.
Corresponding to the above method for identifying a user shooting intention, an embodiment of the present disclosure further provides an apparatus for identifying a user shooting intention, as shown in fig. 7, where the apparatus may include: memory 702, one or more processors 704, and one or more programs. Wherein the one or more programs are stored in the memory 702 and configured to be executed by the one or more processors 704, the programs when executed by the processor 704 performing the steps of:
a set of captured images of a vehicle is acquired. The group of photographed images includes at least two photographed images.
Position information and attitude information of the photographing apparatus at the time of photographing each of the group of photographed images are determined.
The components of the vehicle in each photographed image are detected, thereby obtaining the component detection result of each photographed image.
The position information, the posture information, and the component detection result are input into a user photographing intention recognition model to recognize photographing intentions of the respective photographed images.
The device for identifying the shooting intention of the user provided by the embodiment of the specification can identify the shooting intention of the user.
Fig. 7 shows an example in which the user capturing intention recognition device provided in the embodiment of the present disclosure is a server. In practical application, the device may also be a terminal, which is not limited in this specification.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for the apparatus embodiments, since they are substantially similar to the method embodiments, the description is relatively simple, and reference is made to the description of the method embodiments in part.
The steps of a method or algorithm described in connection with the disclosure herein may be embodied in hardware, or may be embodied in software instructions executed by a processor. The software instructions may be comprised of corresponding software modules that may be stored in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor such the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor. The processor and the storage medium may reside in an ASIC. In addition, the ASIC may reside in a server. The processor and the storage medium may reside as discrete components in a server.
Those skilled in the art will appreciate that in one or more of the examples described above, the functions described in the present invention may be implemented in hardware, software, firmware, or any combination thereof. When implemented in software, these functions may be stored on or transmitted over as one or more instructions or code on a computer-readable medium. Computer-readable media includes both computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. A storage media may be any available media that can be accessed by a general purpose or special purpose computer.
The foregoing describes specific embodiments of the present disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
The foregoing detailed description of the embodiments has further described the objects, technical solutions and advantages of the present specification, and it should be understood that the foregoing description is only a detailed description of the embodiments of the present specification, and is not intended to limit the scope of the present specification, but any modifications, equivalents, improvements, etc. made on the basis of the technical solutions of the present specification should be included in the scope of the present specification.
Claims (20)
1. A method for identifying a user's photographing intention, comprising:
Acquiring a group of shooting images;
Acquiring position information and posture information of a shooting device when shooting each shooting image in the group of shooting images; the shooting equipment forms a motion track of the shooting equipment according to the position information and the posture information when shooting each shooting image;
performing target detection on each shooting image to obtain target detection results corresponding to each shooting image;
And identifying the shooting intention corresponding to each shooting image based on the position information, the gesture information and the target detection result corresponding to each shooting image.
2. The method of claim 1, wherein the acquiring position information and pose information of the photographing device when photographing each of the set of photographed images comprises:
Extracting key points from any two shooting images with overlapping areas in the group of shooting images respectively;
Performing feature matching on the key points on the two shooting images so as to obtain a plurality of groups of successfully matched target key points;
And determining the position information and the posture information of the shooting equipment when shooting the two shooting images according to the position information of the target key points which are successfully matched in the two shooting images.
3. The method according to claim 2, wherein the determining the position information and the posture information of the photographing device when photographing the two photographed images according to the position information of the target key points in the two photographed images, which are successfully matched, includes:
determining a transformation relation between the two shooting images according to the position information of the target key points successfully matched in the plurality of groups in the two shooting images;
estimating the position information of the plurality of groups of successfully matched target key points in a three-dimensional space according to the position information and the transformation relation;
And carrying out re-projection calculation on the position information of the target key points successfully matched in the plurality of groups in each shooting image and the position information in the three-dimensional space so as to determine the position information and the posture information of shooting equipment when shooting the two shooting images.
4. A method for identifying a user's photographing intention, comprising:
Acquiring a set of photographed images of a vehicle;
Acquiring position information and posture information of a shooting device when shooting each shooting image in the group of shooting images; the shooting equipment forms a motion track of the shooting equipment according to the position information and the posture information when shooting each shooting image;
performing component detection on each photographed image to obtain a component detection result corresponding to each photographed image;
And identifying the shooting intention corresponding to each shooting image based on the position information, the gesture information and the detection result of the corresponding component of each shooting image.
5. A vehicle damage identification method, comprising:
Acquiring a set of photographed images of a vehicle;
Identifying respective corresponding shooting intents of each shooting image in the group of shooting images; the shooting intention is determined based on a motion trail of the shooting device when shooting each shooting image and a target detection result corresponding to each shooting image; the motion trail is formed based on position information and posture information of the shooting equipment when shooting the shooting images;
And determining a damage detection result of the vehicle according to each photographed image and the corresponding photographing intention.
6. The method of claim 5, wherein the determining the damage detection result of the vehicle according to the respective photographed images and the respective corresponding photographed intents comprises:
determining the weight corresponding to each shooting image according to the shooting intention corresponding to each shooting image;
and determining a damage detection result of the vehicle by using a damage detection model based on the photographed images and the corresponding weights.
7. The method of claim 5, wherein the determining the damage detection result of the vehicle according to the respective captured images and the respective corresponding capturing intents comprises:
and inputting the photographed images and the corresponding photographing intentions into a damage detection model to determine a damage detection result of the vehicle.
8. The method of claim 5, wherein the determining the damage detection result of the vehicle according to the respective photographed images and the respective corresponding photographed intents comprises:
Screening key shooting images from the shooting images at least based on shooting intents corresponding to the shooting images;
And inputting the key shooting images into a damage detection model to determine a damage detection result of the vehicle.
9. The method of claim 8, wherein the screening key captured images from the respective captured images comprises:
screening key shooting images from the shooting images according to a predefined screening rule and shooting intentions corresponding to the shooting images; wherein the screening rule includes a matching condition.
10. The method of claim 8, wherein the screening key captured images from the respective captured images comprises:
and inputting at least each shooting image and the corresponding shooting intention into an image screening model, and taking the shooting image output by the image screening model as the key shooting image.
11. An apparatus for recognizing a user's photographing intention, comprising:
an acquisition unit configured to acquire a group of captured images;
the acquisition unit is further used for acquiring position information and posture information of the shooting equipment when shooting each shooting image in the group of shooting images; the shooting equipment forms a motion track of the shooting equipment according to the position information and the posture information when shooting each shooting image;
The target detection unit is used for carrying out target detection on each shooting image to obtain target detection results corresponding to each shooting image;
And the identification unit is used for identifying the shooting intention corresponding to each shooting image based on the position information, the gesture information and the target detection result corresponding to each shooting image.
12. The device of claim 11, wherein the acquisition unit is specifically configured to:
Extracting key points from any two shooting images with overlapping areas in the group of shooting images respectively;
Performing feature matching on the key points on the two shooting images so as to obtain a plurality of groups of successfully matched target key points;
And determining the position information and the posture information of the shooting equipment when shooting the two shooting images according to the position information of the target key points which are successfully matched in the two shooting images.
13. The apparatus of claim 12, wherein the obtaining unit is further specifically configured to:
determining a transformation relation between the two shooting images according to the position information of the target key points successfully matched in the plurality of groups in the two shooting images;
estimating the position information of the plurality of groups of successfully matched target key points in a three-dimensional space according to the position information and the transformation relation;
And carrying out re-projection calculation on the position information of the target key points successfully matched in the plurality of groups in each shooting image and the position information in the three-dimensional space so as to determine the position information and the posture information of shooting equipment when shooting the two shooting images.
14. An apparatus for recognizing a user's photographing intention, comprising:
an acquisition unit configured to acquire a set of photographed images of a vehicle;
the acquisition unit is further used for acquiring position information and posture information of the shooting equipment when shooting each shooting image in the group of shooting images; the shooting equipment forms a motion track of the shooting equipment according to the position information and the posture information when shooting each shooting image;
the component detection unit is used for carrying out component detection on each photographed image to obtain a component detection result corresponding to each photographed image;
And the identification unit is used for identifying the shooting intention corresponding to each shooting image based on the position information, the gesture information and the detection result of the corresponding component of each shooting image.
15. A vehicle damage identification device, comprising:
an acquisition unit configured to acquire a set of photographed images of a vehicle;
An identifying unit, configured to identify a shooting intention corresponding to each of the shooting images in the group of shooting images; the shooting intention is determined based on a motion trail of the shooting device when shooting each shooting image and a target detection result corresponding to each shooting image; the motion trail is formed based on position information and posture information of the shooting equipment when shooting the shooting images;
and the determining unit is used for determining the damage detection result of the vehicle according to the photographed images and the corresponding photographing intentions.
16. The apparatus of claim 15, wherein the determining unit is specifically configured to:
determining the weight corresponding to each shooting image according to the shooting intention corresponding to each shooting image;
and determining a damage detection result of the vehicle by using a damage detection model based on the photographed images and the corresponding weights.
17. The apparatus of claim 15, wherein the determining unit is specifically configured to:
and inputting the photographed images and the corresponding photographing intentions into a damage detection model to determine a damage detection result of the vehicle.
18. The apparatus of claim 15, wherein the determining unit is specifically configured to:
Screening key shooting images from the shooting images at least based on shooting intents corresponding to the shooting images;
And inputting the key shooting images into a damage detection model to determine a damage detection result of the vehicle.
19. The apparatus of claim 18, wherein the determining unit is further specifically configured to:
screening key shooting images from the shooting images according to a predefined screening rule and shooting intentions corresponding to the shooting images; wherein the screening rule includes a matching condition.
20. The apparatus of claim 18, wherein the determining unit is further specifically configured to:
and inputting at least each shooting image and the corresponding shooting intention into an image screening model, and taking the shooting image output by the image screening model as the key shooting image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110845857.3A CN113567149B (en) | 2018-09-03 | 2018-09-03 | User shooting intention recognition method, device and equipment |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110845857.3A CN113567149B (en) | 2018-09-03 | 2018-09-03 | User shooting intention recognition method, device and equipment |
| CN201811023190.3A CN110567728B (en) | 2018-09-03 | 2018-09-03 | Method, device and equipment for identifying shooting intention of user |
Related Parent Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811023190.3A Division CN110567728B (en) | 2018-09-03 | 2018-09-03 | Method, device and equipment for identifying shooting intention of user |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113567149A CN113567149A (en) | 2021-10-29 |
| CN113567149B true CN113567149B (en) | 2024-05-31 |
Family
ID=68772359
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811023190.3A Active CN110567728B (en) | 2018-09-03 | 2018-09-03 | Method, device and equipment for identifying shooting intention of user |
| CN202110845857.3A Active CN113567149B (en) | 2018-09-03 | 2018-09-03 | User shooting intention recognition method, device and equipment |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811023190.3A Active CN110567728B (en) | 2018-09-03 | 2018-09-03 | Method, device and equipment for identifying shooting intention of user |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN110567728B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101467153A (en) * | 2006-06-08 | 2009-06-24 | 创伤科学有限责任公司 | Method and apparatus for obtaining photogrammetric data to estimate impact severity |
| CN103759716A (en) * | 2014-01-14 | 2014-04-30 | 清华大学 | Dynamic target position and attitude measurement method based on monocular vision at tail end of mechanical arm |
| CN104567725A (en) * | 2014-12-17 | 2015-04-29 | 苏州华兴致远电子科技有限公司 | Vehicle operation fault detection system and method |
| CN107194323A (en) * | 2017-04-28 | 2017-09-22 | 阿里巴巴集团控股有限公司 | Car damage identification image acquiring method, device, server and terminal device |
| KR101798988B1 (en) * | 2017-07-12 | 2017-11-17 | (주)잼시큐리티시스템 | Under vehicle monitoring system |
| CN107368776A (en) * | 2017-04-28 | 2017-11-21 | 阿里巴巴集团控股有限公司 | Car damage identification image acquiring method, device, server and terminal device |
| CN107884419A (en) * | 2017-11-08 | 2018-04-06 | 安吉汽车物流股份有限公司 | Automobile chassis automatic checkout equipment, automobile intelligent detecting system |
Family Cites Families (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100430690C (en) * | 2006-12-19 | 2008-11-05 | 南京航空航天大学 | Method for making three-dimensional measurement of objects utilizing single digital camera to freely shoot |
| CN102221358B (en) * | 2011-03-23 | 2012-12-12 | 中国人民解放军国防科学技术大学 | Monocular visual positioning method based on inverse perspective projection transformation |
| KR101354157B1 (en) * | 2012-08-17 | 2014-01-23 | 영남대학교 산학협력단 | Shock sensing device for vehicle and method for controlling therefof |
| CN103175512B (en) * | 2013-03-08 | 2014-12-31 | 中国人民解放军国防科学技术大学 | Shooting measurement method of attitude of tail end of boom of concrete pump truck |
| US10319035B2 (en) * | 2013-10-11 | 2019-06-11 | Ccc Information Services | Image capturing and automatic labeling system |
| CN107403424B (en) * | 2017-04-11 | 2020-09-18 | 阿里巴巴集团控股有限公司 | Vehicle loss assessment method and device based on image and electronic equipment |
| CN107392218B (en) * | 2017-04-11 | 2020-08-04 | 创新先进技术有限公司 | Vehicle loss assessment method and device based on image and electronic equipment |
| CN107358596B (en) * | 2017-04-11 | 2020-09-18 | 阿里巴巴集团控股有限公司 | Vehicle loss assessment method and device based on image, electronic equipment and system |
| CN107167139A (en) * | 2017-05-24 | 2017-09-15 | 广东工业大学 | A kind of Intelligent Mobile Robot vision positioning air navigation aid and system |
| CN108090838B (en) * | 2017-11-21 | 2020-09-29 | 阿里巴巴集团控股有限公司 | Method, device, server, client and system for identifying damaged part of vehicle |
| CN108237992B (en) * | 2017-12-18 | 2020-02-21 | 北京车和家信息技术有限公司 | Vehicle body detection method and vehicle |
-
2018
- 2018-09-03 CN CN201811023190.3A patent/CN110567728B/en active Active
- 2018-09-03 CN CN202110845857.3A patent/CN113567149B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101467153A (en) * | 2006-06-08 | 2009-06-24 | 创伤科学有限责任公司 | Method and apparatus for obtaining photogrammetric data to estimate impact severity |
| CN103759716A (en) * | 2014-01-14 | 2014-04-30 | 清华大学 | Dynamic target position and attitude measurement method based on monocular vision at tail end of mechanical arm |
| CN104567725A (en) * | 2014-12-17 | 2015-04-29 | 苏州华兴致远电子科技有限公司 | Vehicle operation fault detection system and method |
| CN107194323A (en) * | 2017-04-28 | 2017-09-22 | 阿里巴巴集团控股有限公司 | Car damage identification image acquiring method, device, server and terminal device |
| CN107368776A (en) * | 2017-04-28 | 2017-11-21 | 阿里巴巴集团控股有限公司 | Car damage identification image acquiring method, device, server and terminal device |
| KR101798988B1 (en) * | 2017-07-12 | 2017-11-17 | (주)잼시큐리티시스템 | Under vehicle monitoring system |
| CN107884419A (en) * | 2017-11-08 | 2018-04-06 | 安吉汽车物流股份有限公司 | Automobile chassis automatic checkout equipment, automobile intelligent detecting system |
Non-Patent Citations (1)
| Title |
|---|
| 基于彩色图像处理的高速列车快速检测方法;胡广胜 等;MNCN -;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110567728B (en) | 2021-08-20 |
| CN110567728A (en) | 2019-12-13 |
| CN113567149A (en) | 2021-10-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109711243B (en) | Static three-dimensional face in-vivo detection method based on deep learning | |
| CN108960211B (en) | Multi-target human body posture detection method and system | |
| US9177404B2 (en) | Systems and methods of merging multiple maps for computer vision based tracking | |
| US11205276B2 (en) | Object tracking method, object tracking device, electronic device and storage medium | |
| KR101722803B1 (en) | Method, computer program, and device for hybrid tracking of real-time representations of objects in image sequence | |
| WO2012086821A1 (en) | Positioning apparatus and positioning method | |
| WO2014126671A1 (en) | Camera aided motion direction and speed estimation | |
| KR20120048370A (en) | Object pose recognition apparatus and method using the same | |
| CN106981077A (en) | Infrared image and visible light image registration method based on DCE and LSS | |
| CN109271908B (en) | Vehicle loss detection method, device and equipment | |
| Psyllos et al. | M-SIFT: A new method for Vehicle Logo Recognition | |
| CN113610967B (en) | Three-dimensional point detection method, three-dimensional point detection device, electronic equipment and storage medium | |
| JP6577424B2 (en) | Action recognition device and action recognition method | |
| CN111951211B (en) | Target detection method, device and computer readable storage medium | |
| CN112884804A (en) | Action object tracking method and related equipment | |
| CN112070035A (en) | Target tracking method and device based on video stream and storage medium | |
| CN113567149B (en) | User shooting intention recognition method, device and equipment | |
| CN114373203A (en) | Picture archiving method and device, terminal equipment and computer readable storage medium | |
| CN111723610B (en) | Image recognition method, device and equipment | |
| WO2017042852A1 (en) | Object recognition appratus, object recognition method and storage medium | |
| JP6946912B2 (en) | Estimator program, estimator, and estimation method | |
| Ge et al. | Template-based object detection through partial shape matching and boundary verification | |
| CN111767940A (en) | Target object identification method, device, equipment and storage medium | |
| Del Bimbo et al. | Local shape estimation from a single keypoint | |
| CN117726656B (en) | Target tracking method, device, system and medium based on super-resolution image |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |