CN117221727A - Image stabilization control method for moving camera and related equipment - Google Patents
Image stabilization control method for moving camera and related equipment Download PDFInfo
- Publication number
- CN117221727A CN117221727A CN202311305755.8A CN202311305755A CN117221727A CN 117221727 A CN117221727 A CN 117221727A CN 202311305755 A CN202311305755 A CN 202311305755A CN 117221727 A CN117221727 A CN 117221727A
- Authority
- CN
- China
- Prior art keywords
- shake
- motion
- image stabilization
- stabilization control
- data
- 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.)
- Pending
Links
- 230000006641 stabilisation Effects 0.000 title claims abstract description 140
- 238000011105 stabilization Methods 0.000 title claims abstract description 140
- 238000000034 method Methods 0.000 title claims abstract description 72
- 230000033001 locomotion Effects 0.000 claims abstract description 445
- 230000007613 environmental effect Effects 0.000 claims description 22
- 238000004590 computer program Methods 0.000 claims description 11
- 230000015654 memory Effects 0.000 claims description 8
- 230000002265 prevention Effects 0.000 description 13
- 238000013528 artificial neural network Methods 0.000 description 8
- 238000013527 convolutional neural network Methods 0.000 description 8
- 230000000087 stabilizing effect Effects 0.000 description 6
- 230000009194 climbing Effects 0.000 description 4
- 230000006403 short-term memory Effects 0.000 description 4
- 230000009182 swimming Effects 0.000 description 4
- 230000001133 acceleration Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000001914 filtration Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000037237 body shape Effects 0.000 description 2
- 125000004122 cyclic group Chemical group 0.000 description 2
- 238000010191 image analysis Methods 0.000 description 2
- 238000003703 image analysis method Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 210000003205 muscle Anatomy 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 238000000611 regression analysis Methods 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 210000000689 upper leg Anatomy 0.000 description 2
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Landscapes
- Studio Devices (AREA)
- Adjustment Of Camera Lenses (AREA)
Abstract
The embodiment of the invention provides a method for controlling image stabilization of a motion camera, which comprises the following steps: when a target motion mode is entered, a first anti-shake parameter corresponding to the target motion mode is obtained, and a first image stabilization control is performed on the motion camera through the first anti-shake parameter; in the first image stabilization control process, acquiring motion environment data, lens shake data and picture data of the motion camera, and determining a second anti-shake parameter of the motion camera under the first image stabilization control according to the motion environment data, the lens shake data and the picture data; and performing second image stabilization control on the moving camera in the first image stabilization control process through the second anti-shake parameters. According to the invention, the shake data acquisition can be performed on the basis of the first shake-preventing parameter, so that the second shake-preventing parameter which is more suitable for the current motion environment is obtained, the image stabilization control of the motion camera is performed more accurately according to the second shake-preventing parameter, and the stability of the picture shot by the motion camera is improved.
Description
Technical Field
The invention relates to the field of camera control, in particular to a motion camera image stabilization control method and related equipment.
Background
With the wide application of technologies such as video shooting and remote control, the problem of shake of a moving camera becomes a key factor restricting the performance thereof. In practical application, the shake of the camera is affected by various factors, such as wind, light, terrain, personnel walking, and the like, and has strong randomness and time variability. In order to improve the stability of the photographed picture, real-time anti-shake control is required. However, in the conventional image stabilization control method, shake prediction is usually performed based on a simple vibration model, for example, only image shake is considered to perform image stabilization control, or only sensor data is considered to perform image stabilization control, and complex dynamic characteristics of an actual system cannot be accurately described, so that stability of images shot by a moving camera is low.
Disclosure of Invention
The embodiment of the invention provides a video stabilization control method of a moving camera, which aims to solve the problem that the stability of pictures shot by the moving camera is low because complex dynamic characteristics of an actual system cannot be accurately described in the prior art. The motion environment data and the lens shake data and the picture data of the motion camera are acquired in the first image stabilization control process, and the second shake prevention parameters of the motion camera under the first image stabilization control are determined according to the motion environment data, the lens shake data and the picture data, so that shake data acquisition can be performed on the basis of the first shake prevention parameters, the second shake prevention parameters which can be more suitable for the current motion environment are obtained, further, the motion camera is more accurately subjected to image stabilization control according to the second shake prevention parameters, and the picture stability shot by the motion camera is improved.
In a first aspect, an embodiment of the present invention provides a method for controlling image stabilization of a motion camera, where the method includes:
when a target motion mode is entered, a first anti-shake parameter corresponding to the target motion mode is obtained, and a first image stabilization control is performed on the motion camera through the first anti-shake parameter;
in the first image stabilization control process, acquiring motion environment data, lens shake data and picture data of the motion camera, and determining a second anti-shake parameter of the motion camera under the first image stabilization control according to the motion environment data, the lens shake data and the picture data;
and performing second image stabilization control on the moving camera in the first image stabilization control process through the second anti-shake parameter.
Optionally, when the target motion mode is entered, a first anti-shake parameter corresponding to the target motion mode is obtained, and before the first image stabilization control is performed on the motion camera by using the first anti-shake parameter, the method further includes:
acquiring body data and historical motion data of a user;
determining a motion type of the user based on the historical motion data, and determining a motion mode of the motion camera according to the motion type, wherein one motion type corresponds to one motion mode;
Determining a motion model corresponding to the motion type based on the motion type and the body data, wherein one motion type corresponds to one motion model;
and determining anti-shake parameters corresponding to each motion mode in the motion camera based on each motion model, wherein one motion mode corresponds to one anti-shake parameter.
Optionally, when the target motion mode is entered, acquiring a first anti-shake parameter corresponding to the target motion mode includes:
when a target motion mode is entered, traversing and searching are carried out in an anti-shake parameter table to obtain a first anti-shake parameter corresponding to the target motion mode, wherein the anti-shake parameter table comprises a corresponding relation between a motion model and the anti-shake parameter.
Optionally, the determining, according to the motion environment data, the lens shake data, and the frame data, a second anti-shake parameter of the motion camera under the first image stabilization control includes:
determining the environment shake amount of the moving camera under the first image stabilization control based on the moving environment data;
determining the lens shake amount of the moving camera under the first image stabilization control based on the lens shake data;
Determining the picture shake amount of the moving camera under the first image stabilization control based on the picture data;
and determining a second anti-shake parameter of the moving camera under the first image stabilization control based on the environment shake amount, the lens shake amount and the picture shake amount.
Optionally, the determining, based on the picture data, the picture shake amount of the motion camera under the first image stabilization control includes:
performing target recognition processing on the picture data to obtain a foreground target and a background target;
and determining the picture shake amount of the moving camera under the first image stabilization control based on the foreground target and the background target.
Optionally, the determining the second anti-shake parameter of the motion camera under the first image stabilization control based on the environmental shake amount, the lens shake amount, and the image shake amount includes:
determining a shake state of the moving camera at the last moment according to the environment shake amount, the lens shake amount and the picture shake amount;
predicting the shake state of the motion camera at the current moment according to the shake state of the previous moment, the environment shake amount at the current moment, the lens shake amount at the current moment and the picture shake amount at the current moment;
And determining a second anti-shake parameter of the motion camera under the first image stabilization control based on the shake state of the current moment.
Optionally, predicting the shake state of the motion camera at the current time according to the shake state of the previous time, the environmental shake amount at the current time, the lens shake amount at the current time, and the frame shake amount at the current time includes:
predicting the jitter state of the current moment based on the jitter state of the previous moment to obtain a predicted jitter state of the current moment;
obtaining an observation state at the current moment based on the environment shake amount at the current moment, the lens shake amount at the current moment and the picture shake amount at the current moment;
determining a Kalman gain at the current moment based on the predicted jitter state at the current moment and the overlapped part of the observed state at the current moment;
and calculating the jitter state of the motion camera at the current moment based on the predicted jitter state of the current moment, the observation state of the current moment and the Kalman gain of the current moment.
In a second aspect, an embodiment of the present invention further provides an image stabilization control apparatus for a moving camera, where the image stabilization control apparatus for a moving camera includes:
The first acquisition module is used for acquiring a first anti-shake parameter corresponding to a target motion mode when entering the target motion mode, and performing first image stabilization control on the motion camera through the first anti-shake parameter;
the processing module is used for acquiring motion environment data, lens shake data and picture data of the motion camera in the first image stabilization control process, and determining a second anti-shake parameter of the motion camera under the first image stabilization control according to the motion environment data, the lens shake data and the picture data;
and the control module is used for carrying out second image stabilization control on the moving camera in the first image stabilization control process through the second anti-shake parameter.
In a third aspect, an embodiment of the present invention provides an electronic device, including: the image stabilizing control method for the motion camera comprises a memory, a processor and a computer program which is stored in the memory and can run on the processor, wherein the steps in the image stabilizing control method for the motion camera are realized when the processor executes the computer program.
In a fourth aspect, an embodiment of the present invention provides a computer readable storage medium, where a computer program is stored, where the computer program when executed by a processor implements steps in a method for controlling image stabilization of a motion camera provided by the embodiment of the present invention.
In the embodiment of the invention, when a target motion mode is entered, a first anti-shake parameter corresponding to the target motion mode is obtained, and a first image stabilization control is performed on the motion camera through the first anti-shake parameter; in the first image stabilization control process, acquiring motion environment data, lens shake data and picture data of the motion camera, and determining a second anti-shake parameter of the motion camera under the first image stabilization control according to the motion environment data, the lens shake data and the picture data; and performing second image stabilization control on the moving camera in the first image stabilization control process through the second anti-shake parameter. The motion environment data and the lens shake data and the picture data of the motion camera are acquired in the first image stabilization control process, and the second shake prevention parameters of the motion camera under the first image stabilization control are determined according to the motion environment data, the lens shake data and the picture data, so that shake data acquisition can be performed on the basis of the first shake prevention parameters, the second shake prevention parameters which can be more suitable for the current motion environment are obtained, further, the motion camera is more accurately subjected to image stabilization control according to the second shake prevention parameters, and the picture stability shot by the motion camera is improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, 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 flowchart of a method for controlling image stabilization of a moving camera according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a motion camera image stabilization control device according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
As shown in fig. 1, fig. 1 is a flowchart of a method for controlling image stabilization of a moving camera according to an embodiment of the present invention. The image stabilization control method of the moving camera comprises the following steps:
101. when the target motion mode is entered, a first anti-shake parameter corresponding to the target motion mode is obtained, and the first image stabilization control is performed on the motion camera through the first anti-shake parameter.
In the embodiment of the invention, the motion camera may include a normal mode and a plurality of motion modes, the normal mode may take the motion camera as a normal camera, and the normal mode may perform or not perform the image stabilization control. The motion modes correspond to different motion types, each motion type can correspond to one motion mode, and corresponding first image stabilization control can be performed according to the corresponding motion type in the motion modes.
The user can switch modes according to a mode switching instruction preset on the moving camera, and the mode of the moving camera is switched between a normal mode and a plurality of moving modes. The triggering of the mode switching instruction may be performed by a physical key of the motion camera or an interface key of the motion camera. When the user switches to a certain motion mode and stays in the motion mode for a preset time, the motion mode can be determined to be a target motion mode, and then the motion camera is controlled to enter the target motion mode.
Among the plurality of motion modes, each motion mode corresponds to a first anti-shake parameter, the first anti-shake parameters can be obtained by sorting according to corresponding motion types, different motion types can understand different first anti-shake parameters, the motion types can include riding, jogging, mountain climbing, swimming, parachuting, long jump and the like, and the first anti-shake parameters can include first anti-shake parameters corresponding to riding, first anti-shake parameters corresponding to jogging, first anti-shake parameters corresponding to mountain climbing, first anti-shake parameters corresponding to swimming, first anti-shake parameters corresponding to parachuting, first anti-shake parameters corresponding to long jump and the like. It should be noted that the foregoing riding, jogging, mountain climbing, swimming, parachuting, and long jump are merely examples of the types of movement, the foregoing first anti-shake parameters corresponding to the riding movement, the first anti-shake parameters corresponding to the jogging movement, the first anti-shake parameters corresponding to the mountain climbing movement, the first anti-shake parameters corresponding to the swimming movement, the first anti-shake parameters corresponding to the parachuting movement, and the first anti-shake parameters corresponding to the long jump movement are also merely examples of the corresponding first anti-shake parameters, and the scope of the present application should not be limited thereto, and other types of movement and the first anti-shake parameters corresponding to other types of movement may be used.
After entering the target motion mode, the first anti-shake parameters corresponding to the target motion mode can be matched, corresponding control instructions are generated through the first anti-shake parameters, and the motion camera is controlled through the control instructions generated by the first anti-shake parameters, so that the motion camera can shoot images with stable pictures when a user performs corresponding motions.
In one possible embodiment, before entering the target motion mode, if it is detected that the user switches to a certain motion mode and stays in the motion mode for a preset time period, the motion condition of the motion camera and the motion rule of the motion mode within the preset time period are obtained, and if the motion condition of the motion camera within the preset time period accords with the motion rule of the motion mode, the motion mode is determined to be the target motion mode. In this way, the accuracy of the target movement pattern can be improved. When the motion condition of the motion camera within the preset time period does not accord with the motion rule of the motion type, prompting the user that the current motion mode of the motion camera does not accord with the current motion type of the user so that the user can switch to the corresponding motion mode. Further, when the motion condition of the motion camera within the preset duration does not accord with the motion rule of the motion type, the motion condition of the motion camera within the preset duration can be compared with the motion rule of each motion type to determine a corresponding candidate motion type, a corresponding candidate motion mode is determined according to the candidate motion type, and the user is prompted that the current motion mode of the motion camera does not accord with the current motion type of the user, so that the user can switch to the corresponding candidate motion mode.
102. In the first image stabilization control process, motion environment data, lens shake data and picture data of a motion camera are obtained, and a second anti-shake parameter of the motion camera under the first image stabilization control is determined according to the motion environment data, the lens shake data and the picture data.
In the embodiment of the invention, in the first image stabilization control process, data acquisition can be continuously performed on the motion environment of the user, the lens shake of the motion camera and the picture of the motion camera, so as to obtain the motion environment data of the user, the lens shake data of the motion camera and the picture data of the motion camera.
The above-mentioned movement environment data may include illumination data, wind speed and direction data, terrain flatness data, terrain gradient data, and the like. The lens shake data may include speed data, acceleration data, angular velocity data, and the like. The picture data may be a picture image photographed by a moving camera. The motion environment data can be acquired by an environment sensing sensor, and the lens shake data can be determined according to the inertial measurement unit of the camera, the gyroscope and other sensors.
After the motion environment data, the lens shake data and the picture data are obtained, a second anti-shake parameter of the motion camera under the first image stabilization control can be determined on the basis of the first anti-shake parameter. Specifically, a prediction model based on a neural network, such as a prediction model based on CNN (convolutional neural network), RNN (cyclic neural network), LSMT (long short-term memory network), may be established, and motion environment data, lens shake data, and picture data are processed through the prediction model to obtain a predicted shake amount, and the predicted shake amount is increased on the basis of the first anti-shake parameter to obtain a second anti-shake parameter.
103. And performing second image stabilization control on the moving camera in the first image stabilization control process through the second anti-shake parameters.
In the embodiment of the invention, after the second anti-shake parameter is obtained, a corresponding control instruction is generated through the second anti-shake parameter, and the motion camera is controlled through the control instruction generated by the second anti-shake parameter, so that the motion camera can shoot images with stable pictures when a user performs corresponding motion.
It should be noted that, the above-mentioned first anti-shake parameter may be understood as an anti-shake parameter common to the target motion mode, because the motion environments where different users are located are different, there is a difference in motion gesture between different users, the first anti-shake parameter is adopted to perform anti-shake only on a common feature level of the motion type, but not on a unique feature level of the user in the motion process, so the above-mentioned first anti-shake parameter may only limitedly improve the stability of the image of the motion camera in the motion process, and the above-mentioned second anti-shake parameter may be understood as an anti-shake parameter specific to the target motion mode, and considering that the motion environments where different users are located are different, there is a difference in motion gesture between different users, so that the second anti-shake parameter may perform anti-shake on a unique feature level of the user in the motion process, thereby further improving the stability of the image of the motion camera.
In the embodiment of the invention, when a target motion mode is entered, a first anti-shake parameter corresponding to the target motion mode is obtained, and a first image stabilization control is performed on the motion camera through the first anti-shake parameter; in the first image stabilization control process, acquiring motion environment data, lens shake data and picture data of the motion camera, and determining a second anti-shake parameter of the motion camera under the first image stabilization control according to the motion environment data, the lens shake data and the picture data; and performing second image stabilization control on the moving camera in the first image stabilization control process through the second anti-shake parameter. The motion environment data and the lens shake data and the picture data of the motion camera are acquired in the first image stabilization control process, and the second shake prevention parameters of the motion camera under the first image stabilization control are determined according to the motion environment data, the lens shake data and the picture data, so that shake data acquisition can be performed on the basis of the first shake prevention parameters, the second shake prevention parameters which can be more suitable for the current motion environment are obtained, further, the motion camera is more accurately subjected to image stabilization control according to the second shake prevention parameters, and the picture stability shot by the motion camera is improved.
Optionally, when entering the target motion mode, acquiring a first anti-shake parameter corresponding to the target motion mode, and before the step of performing first image stabilization control on the motion camera through the first anti-shake parameter, acquiring body data and historical motion data of a user; determining a motion type of a user based on the historical motion data, and determining a motion mode of a motion camera according to the motion type, wherein one motion type corresponds to one motion mode; determining a motion model corresponding to the motion type based on the motion type and the body data, wherein one motion type corresponds to one motion model; and determining anti-shake parameters corresponding to each motion mode in the motion camera based on each motion model, wherein one motion mode corresponds to one anti-shake parameter.
In the embodiment of the present invention, the body data may include data of a body shape (including height, arm length, thigh length, etc.), body mass, muscle strength, etc. of the user, and the body data of the body shape, body mass, muscle strength, etc. of the user may be obtained through uploading by the user or through a wearable device.
The historical motion data can comprise motion type, motion terrain, motion gesture, motion duration and the like, and can be acquired according to a historical record of the wearable device or the motion camera.
After the body data and the historical motion data of the user are obtained, the motion type commonly used by the user can be determined according to the motion type in the historical motion data, and the corresponding motion mode is set in the motion camera according to the motion type commonly used by the user.
Meanwhile, a motion model of the user for each motion type can be established according to the body data and the historical motion data of the user. It will be appreciated that the exercise model is different for different users due to the different physical and historical exercise data of different users, e.g., the thigh lengths of the two users are different, and the stride and frequency of the user during running are different, and thus the corresponding exercise models are different.
The motion conditions of a user under different motion types can be simulated through the motion model, the model of the motion camera is fixed in the motion model corresponding to the user according to the motion conditions under different motion types, the shaking conditions of the motion camera under different motion types are determined according to the motion conditions under different motion types, and the shaking conditions are eliminated to serve as targets to calculate the first shaking prevention parameters under the corresponding motion types. Specifically, when the motion conditions under different motion types are obtained, the first anti-shake parameter under the corresponding motion type can be calculated according to the optical anti-shake method or the electronic anti-shake method.
After the first anti-shake parameters of the user under each motion type are obtained, the first anti-shake parameters are associated with corresponding motion modes, and after the user selects a target motion mode, the corresponding first anti-shake parameters can be found according to the target motion mode, so that the first image stabilization control can be performed on the motion camera through the first anti-shake parameters.
Optionally, in the step of obtaining the first anti-shake parameter corresponding to the target motion mode when entering the target motion mode, the traversal search may be performed in the anti-shake parameter table when entering the target motion mode, so as to obtain the first anti-shake parameter corresponding to the target motion mode, where the anti-shake parameter table includes a correspondence between the motion model and the anti-shake parameter.
In the embodiment of the invention, the motion mode and the first anti-shake parameter may be associated through an anti-shake parameter table, in the anti-shake parameter table, one motion mode corresponds to one first anti-shake parameter, and after entering the target motion mode, the first anti-shake parameter corresponding to the target motion mode may be found out in the anti-shake parameter table.
Optionally, in the step of determining the second anti-shake parameter of the motion camera under the first image stabilization control according to the motion environment data, the lens shake data and the picture data, the amount of environmental shake of the motion camera under the first image stabilization control may be determined based on the motion environment data; determining the lens shake amount of the moving camera under the first image stabilization control based on the lens shake data; determining the picture shake amount of the moving camera under the first image stabilization control based on the picture data; and determining a second anti-shake parameter of the moving camera under the first image stabilization control based on the environment shake amount, the lens shake amount and the picture shake amount.
In the embodiment of the invention, the motion environment data can be analyzed and processed through regression analysis, machine learning or other statistical methods to obtain the environment shake quantity of the motion camera under the first image stabilization control. Specifically, environmental features related to environmental shake can be extracted from the environmental data, and classified regression is performed on the environmental features to obtain the environmental shake amount of the motion camera under the first image stabilization control. The objective of the above classification regression is the amount of environmental jitter. The above-mentioned classification regression may be a linear regression. It will be appreciated that the more extreme the ambient light, e.g., the stronger or weaker, the greater the amount of ambient shake can be determined; in the wind speed and wind direction data, the larger the included angle between the wind direction and the movement direction of the user is, the larger the wind band is, the larger the environment shaking amount can be determined; the smaller the topographic flatness is, the uneven sports topographic is illustrated, and the more pits are, the larger the environment shaking amount can be determined; the larger the terrain gradient, the larger the motion terrain height difference is, and the larger the environment shaking amount can be determined.
Similarly, the lens shake data may be analyzed by regression analysis, machine learning or other statistical methods to obtain the lens shake amount of the moving camera under the first image stabilization control. It can be understood that the greater the movement speed of the user in the movement process, the greater the speed of the moving camera, the greater the lens shake amount can be determined; the larger the acceleration of the user in the motion process is, the larger the acceleration of the motion camera is, the larger the lens shake amount can be determined; the larger the turning or turning speed of the user in the movement process is, the larger the angular speed of the movement camera is, and the larger the lens shake amount can be determined.
For the picture data, image analysis may be performed on the picture data to obtain a picture shake amount of the moving camera under the first image stabilization control. The image analysis may be an image analysis method based on image characteristics or may be a pixel-based image analysis method.
After the environment shake amount, the lens shake amount and the picture shake amount are obtained, a second anti-shake parameter of the moving camera under the first image stabilization control can be determined on the basis of the first anti-shake parameter. Specifically, a prediction model based on a neural network, such as a prediction model based on CNN (convolutional neural network), RNN (cyclic neural network), LSMT (long short-term memory network), may be established, and motion environment data, lens shake data, and picture data are processed through the prediction model to obtain a predicted shake amount, and the predicted shake amount is increased on the basis of the first anti-shake parameter to obtain a second anti-shake parameter.
Optionally, in the step of determining the image shake amount of the moving camera under the first image stabilization control based on the image data, target recognition processing may be performed on the image data to obtain a foreground target and a background target; based on the foreground object and the background object, the image shake amount of the moving camera under the first image stabilizing control is determined.
In the embodiment of the present invention, the foreground object may be a moving object, the background object may be a stationary object, the moving object may be a movable object such as a person, a vehicle, an animal, etc., and the stationary object may be a stationary object such as a road surface, a tree, a lawn, a sky, etc.
Specifically, the frame data at least comprises two continuous frame images, the two continuous frame images are subjected to object recognition processing to obtain a first foreground object and a first background object of a front frame image, and a second foreground object and a second background object of a rear frame image, one first foreground object corresponds to one second foreground object, one first background object corresponds to one second background object, a first motion vector between the first foreground object and the corresponding second foreground object is calculated, a second motion vector between the first foreground object and the corresponding second foreground object is calculated, wherein the first motion vector comprises a relative motion vector between a motion vector of a user and the foreground object, the second motion vector comprises a relative motion vector between the motion vector of the user and the background object, and the second motion vector is a motion vector of the user because the background object is static. In general, the cause of the picture shake originates from a motion vector of a user, and thus, the picture shake amount can be calculated by the following equation:
Wherein u is as described above (t-1,t) The amount of screen shake between the preceding screen image t-1 and the following screen image t is represented by N, M, a, and a, respectively, the number of foreground objects, the number of background objects, and the number of background objects, respectively i A first motion vector representing an ith foreground object, b j A second motion vector representing a jth foreground object, wherein Δt represents a time interval between the preceding picture image t-1 and the following picture image t, and k is 1 With k above 2 Is a priori value of k 1 With k above 2 Associated with the number of foreground objects N and the number of background objects M.
The above-mentioned environmental jitter amount may be the environmental jitter amount sequence V from the time of entering the target motion mode to the current time t ＝{v 1 ,v 2 ,…,v t-1 ,v t The lens shake amount may be a sequence R of lens shake amounts from when the target motion mode is entered to the current time t ＝{r 1 ,r 2 ,…,r t-1 ,r t The above-mentioned picture jitter amount may be a picture jitter amount sequence U from when the target motion mode is entered to the current time t ＝{u 1 ,u 2 ,…,u t-1 ,u t }. In the predicted motion camera environment shake amount sequence V t ＝{v 1 ,v 2 ,…,v t-1 ,v t Sequence of lens shake amount R t ＝{r 1 ,r 2 ,…,r t-1 ,r t Sequence of picture jitter amounts U t ＝{u 1 ,u 2 ,…,u t-1 ,u t After } can be at the first preventionAnd determining a second anti-shake parameter of the motion camera under the first image stabilization control on the basis of the shake parameter. Specifically, a prediction model based on a neural network, such as a prediction model based on CNN (convolutional neural network), RNN (recurrent neural network), LSMT (long short-term memory network), may be established, and the environmental jitter amount sequence V is determined by the prediction model t ＝{v 1 ,v 2 ,…,v t-1 ,v t Sequence of lens shake amount R t ＝{r 1 ,r 2 ,…,r t-1 ,r t Sequence of picture jitter amounts U t ＝{u 1 ,u 2 ,…,u t-1 ,u t And processing to obtain a predicted jitter amount, and adding the predicted jitter amount to the first anti-jitter parameter to obtain a second anti-jitter parameter.
Optionally, in the step of determining the second anti-shake parameter of the motion camera under the first image stabilization control based on the environmental shake amount, the lens shake amount, and the picture shake amount, a shake state of the motion camera at a previous time may be determined according to the environmental shake amount, the lens shake amount, and the picture shake amount; predicting the shake state of the motion camera at the current moment according to the shake state at the last moment, the environment shake amount at the current moment, the lens shake amount at the current moment and the picture shake amount at the current moment; and determining a second anti-shake parameter of the motion camera under the first image stabilization control based on the shake state at the current moment.
In the embodiment of the invention, the sequence V can be based on the environment jitter amount t ＝{v 1 ,v 2 ,…,v t-1 ,v t Sequence of lens shake amount R t ＝{r 1 ,r 2 ,…,r t-1 ,r t Sequence of picture jitter amounts U t ＝{u 1 ,u 2 ,…,u t-1 ,u t Determining the jitter state s of the moving camera at the previous moment t-1 Based on the jitter state s at the previous time t-1 The amount of environmental shake v at the present moment t The lens shake amount r at the current moment t Picture shake amount u at current time t Predicting a jitter state s of a motion camera at a current moment t . In particular, it is possible toKalman filtering algorithm is applied to environment jitter amount sequence V t-1 ＝{v 1 ,v 2 ,…,v t-1 Sequence of lens shake amount R t-1 ＝{r 1 ,r 2 ,…,r t-1 Sequence of picture jitter amounts U t-1 ＝{u 1 ,u 2 ,…,u t-1 Filtering to obtain jitter state s of the motion camera at the last moment t-1 Based on the jitter state s of the last time instant t-1 The amount of environmental shake v at the present moment t The lens shake amount r at the current moment t Picture shake amount u at current time t Predicting the jitter state s of a motion camera at the current moment by using a Kalman filtering algorithm t 。
At the time of predicting the jitter state s of the motion camera at the current time t And then, determining a second anti-shake parameter of the motion camera under the first image stabilization control on the basis of the first anti-shake parameter. Specifically, a prediction model based on a neural network, such as a prediction model based on CNN (convolutional neural network), RNN (recurrent neural network), LSMT (long short-term memory network), may be established, and the jitter state s at the current time is determined by the prediction model t Processing is carried out to obtain a predicted jitter amount, and the predicted jitter amount is increased on the basis of the first anti-jitter parameter to obtain a second anti-jitter parameter. The jitter state sequence S can also be processed by a prediction model t ＝{s 1 ,s 2 ,…,s t-1 ,s t And processing to obtain a predicted jitter amount, and adding the predicted jitter amount to the first anti-jitter parameter to obtain a second anti-jitter parameter.
Optionally, in the step of predicting the shake state of the motion camera at the current time according to the shake state of the previous time, the environmental shake amount at the current time, the lens shake amount at the current time and the picture shake amount at the current time, the shake state at the current time may be predicted based on the shake state at the previous time, so as to obtain a predicted shake state at the current time; obtaining an observation state at the current moment based on the environment shake amount at the current moment, the lens shake amount at the current moment and the picture shake amount at the current moment; determining a Kalman gain at the current moment based on the overlapped part of the predicted jitter state at the current moment and the observed state at the current moment; and calculating the jitter state of the motion camera at the current moment based on the predicted jitter state at the current moment, the observation state at the current moment and the Kalman gain at the current moment.
In the embodiment of the invention, the state prediction part of the Kalman filter can be used for prediction. Specifically, the jitter state s of the previous time can be used t-1 As an input value, a predicted jitter state ps at the current time is obtained by a state prediction section of a Kalman filter t . Then the environment shake amount, the lens shake amount and the picture shake amount can be processed by using the observation part of the Kalman filter to obtain the observation state qs at the current moment t . The observation portion of the kalman filter takes into account the actual influence of the amount of ambient shake, the amount of lens shake, and the amount of picture shake on camera shake. Then, the jitter state ps is predicted based on the current time t Observation state qs at current time t Overlapping portions zs of (2) t Determining a Kalman gain w at the current moment; predicting jitter state ps based on current time t Observation state qs at current time t And the Kalman gain w at the current moment is calculated to obtain the jitter state s of the motion camera at the current moment t 。
The above-described predictive jitter state ps t Can pass through S t-1 ＝{s 1 ,s 2 ,…,s t-1 Expected μ of } 0 Sum covariance delta 0 Obtained for the observation state qs t Thus can pass through the overlapping portion zs t Sum covariance delta 1 Representation is made, since μ is expected 0 Sum covariance delta 0 Can pass through S t-1 ＝{s 1 ,s 2 ,…,s t-1 Obtained, covariance delta 1 May be shifted by the ambient jitter amount sequence V t ＝{v 1 ,v 2 ,…,v t-1 ,v t Sequence of lens shake amount R t ＝{r 1 ,r 2 ,…,r t-1 ,r t Sequence of picture jitter amounts U t ＝{u 1 ,u 2 ,…,u t-1 ,u t Obtained, kalman gain w=δ 0 (δ 0 +δ 1 ) -1 ，μ 1 ＝w(zs t -μ 0 )，δ 1 ＝wδ 0 And then the jitter state s of the motion camera at the current moment can be solved t ＝(μ 1 ，δ 1 )。
In the embodiment of the invention, the influence of the environment shake quantity, the lens shake quantity and the picture shake quantity is considered, so that the shake state at the current moment is more accurate, and the accuracy of the second anti-shake parameter is further improved.
As shown in fig. 2, an embodiment of the present invention provides a motion camera image stabilization control apparatus, including:
the first obtaining module 201 is configured to obtain a first anti-shake parameter corresponding to a target motion mode when entering the target motion mode, and perform a first image stabilization control on the motion camera through the first anti-shake parameter;
the processing module 202 is configured to obtain motion environment data and lens shake data and picture data of the motion camera during the first image stabilization control process, and determine a second anti-shake parameter of the motion camera under the first image stabilization control according to the motion environment data, the lens shake data and the picture data;
and the control module 203 is configured to perform a second image stabilization control on the moving camera in the first image stabilization control process through the second anti-shake parameter.
Optionally, the apparatus further includes:
the second acquisition module is used for acquiring body data and historical motion data of a user;
the first determining module is used for determining the motion type of the user based on the historical motion data and determining the motion mode of the motion camera according to the motion type, wherein one motion type corresponds to one motion mode;
a second determining module, configured to determine a motion model corresponding to the motion type, one of the motion types corresponding to one of the motion models, based on the motion type and the body data;
and the third determining module is used for determining anti-shake parameters corresponding to each motion mode in the motion camera based on each motion model, wherein one motion mode corresponds to one anti-shake parameter.
Optionally, when the first obtaining module 201 is further configured to enter a target motion mode, performing traversal search in an anti-shake parameter table to obtain a first anti-shake parameter corresponding to the target motion mode, where the anti-shake parameter table includes a correspondence between a motion model and the anti-shake parameter.
Optionally, the processing module 202 is further configured to determine an amount of environmental shake of the motion camera under the first image stabilization control based on the motion environment data; determining the lens shake amount of the moving camera under the first image stabilization control based on the lens shake data; determining the picture shake amount of the moving camera under the first image stabilization control based on the picture data; and determining a second anti-shake parameter of the moving camera under the first image stabilization control based on the environment shake amount, the lens shake amount and the picture shake amount.
Optionally, the processing module 202 is further configured to perform object recognition processing on the frame data to obtain a foreground object and a background object; and determining the picture shake amount of the moving camera under the first image stabilization control based on the foreground target and the background target.
Optionally, the processing module 202 is further configured to determine a shake status of the motion camera at a previous moment according to the environmental shake amount, the lens shake amount, and the frame shake amount; predicting the shake state of the motion camera at the current moment according to the shake state of the previous moment, the environment shake amount at the current moment, the lens shake amount at the current moment and the picture shake amount at the current moment; and determining a second anti-shake parameter of the motion camera under the first image stabilization control based on the shake state of the current moment.
Optionally, the processing module 202 is further configured to predict a jitter state at a current time based on the jitter state at the previous time, to obtain a predicted jitter state at the current time; obtaining an observation state at the current moment based on the environment shake amount at the current moment, the lens shake amount at the current moment and the picture shake amount at the current moment; determining a Kalman gain at the current moment based on the predicted jitter state at the current moment and the overlapped part of the observed state at the current moment; and calculating the jitter state of the motion camera at the current moment based on the predicted jitter state of the current moment, the observation state of the current moment and the Kalman gain of the current moment.
The image stabilizing control device for the moving camera provided by the embodiment of the invention can realize all the processes realized by the image stabilizing control method for the moving camera in the embodiment of the method, and can achieve the same beneficial effects. In order to avoid repetition, a description thereof is omitted.
Referring to fig. 3, fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present invention, as shown in fig. 3, including: a memory 302, a processor 301, and a computer program stored on the memory 302 and executable on the processor 301 for a motion camera image stabilization control method, wherein:
the processor 301 is configured to call a computer program stored in the memory 302, and perform the following steps:
when a target motion mode is entered, a first anti-shake parameter corresponding to the target motion mode is obtained, and a first image stabilization control is performed on the motion camera through the first anti-shake parameter;
in the first image stabilization control process, acquiring motion environment data, lens shake data and picture data of the motion camera, and determining a second anti-shake parameter of the motion camera under the first image stabilization control according to the motion environment data, the lens shake data and the picture data;
And performing second image stabilization control on the moving camera in the first image stabilization control process through the second anti-shake parameter.
Optionally, when the target motion mode is entered, a first anti-shake parameter corresponding to the target motion mode is obtained, and before the first image stabilization control is performed on the motion camera by using the first anti-shake parameter, the method executed by the processor 301 further includes:
acquiring body data and historical motion data of a user;
determining a motion type of the user based on the historical motion data, and determining a motion mode of the motion camera according to the motion type, wherein one motion type corresponds to one motion mode;
determining a motion model corresponding to the motion type based on the motion type and the body data, wherein one motion type corresponds to one motion model;
and determining anti-shake parameters corresponding to each motion mode in the motion camera based on each motion model, wherein one motion mode corresponds to one anti-shake parameter.
Optionally, when the entering the target motion mode performed by the processor 301, acquiring a first anti-shake parameter corresponding to the target motion mode includes:
When a target motion mode is entered, traversing and searching are carried out in an anti-shake parameter table to obtain a first anti-shake parameter corresponding to the target motion mode, wherein the anti-shake parameter table comprises a corresponding relation between a motion model and the anti-shake parameter.
Optionally, the determining, by the processor 301, the second anti-shake parameter of the motion camera under the first image stabilization control according to the motion environment data, the lens shake data, and the picture data includes:
determining the environment shake amount of the moving camera under the first image stabilization control based on the moving environment data;
determining the lens shake amount of the moving camera under the first image stabilization control based on the lens shake data;
determining the picture shake amount of the moving camera under the first image stabilization control based on the picture data;
and determining a second anti-shake parameter of the moving camera under the first image stabilization control based on the environment shake amount, the lens shake amount and the picture shake amount.
Optionally, the determining, by the processor 301, the amount of frame shake of the motion camera under the first image stabilization control based on the frame data includes:
performing target recognition processing on the picture data to obtain a foreground target and a background target;
And determining the picture shake amount of the moving camera under the first image stabilization control based on the foreground target and the background target.
Optionally, the determining, by the processor 301, the second anti-shake parameter of the motion camera under the first image stabilization control based on the environmental shake amount, the lens shake amount, and the frame shake amount includes:
determining a shake state of the moving camera at the last moment according to the environment shake amount, the lens shake amount and the picture shake amount;
predicting the shake state of the motion camera at the current moment according to the shake state of the previous moment, the environment shake amount at the current moment, the lens shake amount at the current moment and the picture shake amount at the current moment;
and determining a second anti-shake parameter of the motion camera under the first image stabilization control based on the shake state of the current moment.
Optionally, the predicting, by the processor 301, the shake state of the motion camera at the current time according to the shake state at the previous time, the environmental shake amount at the current time, the lens shake amount at the current time, and the frame shake amount at the current time includes:
Predicting the jitter state of the current moment based on the jitter state of the previous moment to obtain a predicted jitter state of the current moment;
obtaining an observation state at the current moment based on the environment shake amount at the current moment, the lens shake amount at the current moment and the picture shake amount at the current moment;
determining a Kalman gain at the current moment based on the predicted jitter state at the current moment and the overlapped part of the observed state at the current moment;
and calculating the jitter state of the motion camera at the current moment based on the predicted jitter state of the current moment, the observation state of the current moment and the Kalman gain of the current moment.
The electronic equipment provided by the embodiment of the invention can realize each process realized by the image stabilizing control method of the moving camera in the embodiment of the method, and can achieve the same beneficial effects. In order to avoid repetition, a description thereof is omitted.
The embodiment of the invention also provides a computer readable storage medium, on which a computer program is stored, which when executed by a processor, implements each process of the motion camera image stabilization control method or the application-side motion camera image stabilization control method provided by the embodiment of the invention, and can achieve the same technical effects, and in order to avoid repetition, the description is omitted here.
Those skilled in the art will appreciate that the processes implementing all or part of the methods of the above embodiments may be implemented by a computer program for instructing relevant hardware, and the program may be stored in a computer readable storage medium, and the program may include the processes of the embodiments of the methods as above when executed. The storage medium may be a magnetic disk, an optical disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM) or the like.
The foregoing disclosure is illustrative of the present invention and is not to be construed as limiting the scope of the invention, which is defined by the appended claims.
Claims (10)
1. The image stabilization control method for the motion camera is characterized by comprising the following steps of:
when a target motion mode is entered, a first anti-shake parameter corresponding to the target motion mode is obtained, and a first image stabilization control is performed on the motion camera through the first anti-shake parameter;
in the first image stabilization control process, acquiring motion environment data, lens shake data and picture data of the motion camera, and determining a second anti-shake parameter of the motion camera under the first image stabilization control according to the motion environment data, the lens shake data and the picture data;
And performing second image stabilization control on the moving camera in the first image stabilization control process through the second anti-shake parameter.
2. The method of claim 1, wherein when the target motion mode is entered, acquiring a first anti-shake parameter corresponding to the target motion mode, and before performing a first image stabilization control on the motion camera by using the first anti-shake parameter, the method further comprises:
acquiring body data and historical motion data of a user;
determining a motion type of the user based on the historical motion data, and determining a motion mode of the motion camera according to the motion type, wherein one motion type corresponds to one motion mode;
determining a motion model corresponding to the motion type based on the motion type and the body data, wherein one motion type corresponds to one motion model;
and determining anti-shake parameters corresponding to each motion mode in the motion camera based on each motion model, wherein one motion mode corresponds to one anti-shake parameter.
3. The method of claim 2, wherein the obtaining the first anti-shake parameter corresponding to the target motion mode when the target motion mode is entered comprises:
When a target motion mode is entered, traversing and searching are carried out in an anti-shake parameter table to obtain a first anti-shake parameter corresponding to the target motion mode, wherein the anti-shake parameter table comprises a corresponding relation between a motion model and the anti-shake parameter.
4. A method as claimed in any one of claims 1 to 3, wherein said determining a second anti-shake parameter of the motion camera under first image stabilization control from the motion environment data, the lens shake data, and the picture data comprises:
determining the environment shake amount of the moving camera under the first image stabilization control based on the moving environment data;
determining the lens shake amount of the moving camera under the first image stabilization control based on the lens shake data;
determining the picture shake amount of the moving camera under the first image stabilization control based on the picture data;
and determining a second anti-shake parameter of the moving camera under the first image stabilization control based on the environment shake amount, the lens shake amount and the picture shake amount.
5. The method of claim 4, wherein determining an amount of picture shake of the motion camera under a first stabilization control based on the picture data comprises:
Performing target recognition processing on the picture data to obtain a foreground target and a background target;
and determining the picture shake amount of the moving camera under the first image stabilization control based on the foreground target and the background target.
6. The method of claim 4, wherein the determining a second anti-shake parameter of the motion camera under a first image stabilization control based on the amount of ambient shake, the amount of lens shake, and the amount of picture shake comprises:
determining a shake state of the moving camera at the last moment according to the environment shake amount, the lens shake amount and the picture shake amount;
predicting the shake state of the motion camera at the current moment according to the shake state of the previous moment, the environment shake amount at the current moment, the lens shake amount at the current moment and the picture shake amount at the current moment;
and determining a second anti-shake parameter of the motion camera under the first image stabilization control based on the shake state of the current moment.
7. The method of claim 6, wherein predicting the motion camera's jitter state at the current time based on the previous time's jitter state, the current time's environmental jitter amount, the current time's lens jitter amount, and the current time's picture jitter amount, comprises:
Predicting the jitter state of the current moment based on the jitter state of the previous moment to obtain a predicted jitter state of the current moment;
obtaining an observation state at the current moment based on the environment shake amount at the current moment, the lens shake amount at the current moment and the picture shake amount at the current moment;
determining a Kalman gain at the current moment based on the predicted jitter state at the current moment and the overlapped part of the observed state at the current moment;
and calculating the jitter state of the motion camera at the current moment based on the predicted jitter state of the current moment, the observation state of the current moment and the Kalman gain of the current moment.
8. The utility model provides a motion camera steady image controlling means which characterized in that, motion camera steady image controlling means includes:
the first acquisition module is used for acquiring a first anti-shake parameter corresponding to a target motion mode when entering the target motion mode, and performing first image stabilization control on the motion camera through the first anti-shake parameter;
the processing module is used for acquiring motion environment data, lens shake data and picture data of the motion camera in the first image stabilization control process, and determining a second anti-shake parameter of the motion camera under the first image stabilization control according to the motion environment data, the lens shake data and the picture data;
And the control module is used for carrying out second image stabilization control on the moving camera in the first image stabilization control process through the second anti-shake parameter.
9. An electronic device, comprising: a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps in the motion camera image stabilization control method according to any one of claims 1 to 7 when the computer program is executed.
10. A computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, which when executed by a processor, implements the steps in the moving camera image stabilization control method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311305755.8A CN117221727A (en) | 2023-10-09 | 2023-10-09 | Image stabilization control method for moving camera and related equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311305755.8A CN117221727A (en) | 2023-10-09 | 2023-10-09 | Image stabilization control method for moving camera and related equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117221727A true CN117221727A (en) | 2023-12-12 |
Family
ID=89051117
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311305755.8A Pending CN117221727A (en) | 2023-10-09 | 2023-10-09 | Image stabilization control method for moving camera and related equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117221727A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112203018A (en) * | 2020-09-30 | 2021-01-08 | 钟竞 | Camera anti-shake self-adaptive adjustment method and system based on artificial intelligence |
| CN113395454A (en) * | 2021-07-06 | 2021-09-14 | Oppo广东移动通信有限公司 | Anti-shake method and device for image shooting, terminal and readable storage medium |
| CN115278070A (en) * | 2022-07-23 | 2022-11-01 | 宁波市杭州湾大桥发展有限公司 | Bridge deck monitoring video anti-shaking method and system, storage medium and intelligent terminal |
| CN115604583A (en) * | 2021-07-09 | 2023-01-13 | Oppo广东移动通信有限公司（Cn） | Image processing method, image processor, electronic device, and storage medium |
| WO2023284591A1 (en) * | 2021-07-12 | 2023-01-19 | 华为技术有限公司 | Video capture method and apparatus, electronic device, and storage medium |
| CN116266418A (en) * | 2021-12-15 | 2023-06-20 | 北京小米移动软件有限公司 | Motion recognition method, motion recognition device, electronic equipment and storage medium |
-
2023
- 2023-10-09 CN CN202311305755.8A patent/CN117221727A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112203018A (en) * | 2020-09-30 | 2021-01-08 | 钟竞 | Camera anti-shake self-adaptive adjustment method and system based on artificial intelligence |
| CN113395454A (en) * | 2021-07-06 | 2021-09-14 | Oppo广东移动通信有限公司 | Anti-shake method and device for image shooting, terminal and readable storage medium |
| CN115604583A (en) * | 2021-07-09 | 2023-01-13 | Oppo广东移动通信有限公司（Cn） | Image processing method, image processor, electronic device, and storage medium |
| WO2023284591A1 (en) * | 2021-07-12 | 2023-01-19 | 华为技术有限公司 | Video capture method and apparatus, electronic device, and storage medium |
| CN116266418A (en) * | 2021-12-15 | 2023-06-20 | 北京小米移动软件有限公司 | Motion recognition method, motion recognition device, electronic equipment and storage medium |
| CN115278070A (en) * | 2022-07-23 | 2022-11-01 | 宁波市杭州湾大桥发展有限公司 | Bridge deck monitoring video anti-shaking method and system, storage medium and intelligent terminal |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105654512B (en) | A kind of method for tracking target and device | |
| CN109685797B (en) | Bone point detection method, device, processing equipment and storage medium | |
| JP4553384B2 (en) | Imaging apparatus and control method therefor, computer program, and storage medium | |
| US11394870B2 (en) | Main subject determining apparatus, image capturing apparatus, main subject determining method, and storage medium | |
| US10705408B2 (en) | Electronic device to autofocus on objects of interest within field-of-view of electronic device | |
| CN113066001A (en) | Image processing method and related equipment | |
| CN112119627A (en) | Target following method and device based on holder, holder and computer storage medium | |
| CN106998425A (en) | Vibration testing equipment, method for detecting vibration and picture pick-up device | |
| CN112184767A (en) | Method, device, equipment and storage medium for tracking moving object track | |
| CN116958584B (en) | Key point detection method, regression model training method and device and electronic equipment | |
| JP2019149621A (en) | Information processing device, information processing method, and program | |
| CN113438409B (en) | Delay calibration method, delay calibration device, computer equipment and storage medium | |
| CN103810460B (en) | Object tracking method and object tracking device | |
| CN117221727A (en) | Image stabilization control method for moving camera and related equipment | |
| CN109302562B (en) | Video-based target detection tracking method, system, medium, and device | |
| US9883105B2 (en) | Image processing apparatus and control method thereof | |
| CN114296539B (en) | Direction prediction method, virtual reality device and non-transitory computer readable medium | |
| CN113497889B (en) | Object tracking method and device under motion shooting condition and storage medium | |
| KR102361029B1 (en) | Method, device and non-transitory computer-readable recording medium for suggesting contents about golf | |
| CN110602487B (en) | Video image jitter detection method based on TSN (time delay network) | |
| CN115037869A (en) | Automatic focusing method and device, electronic equipment and computer readable storage medium | |
| Kainz et al. | Estimating the Height of a Person from a Video Sequence | |
| JP5152144B2 (en) | Image processing device | |
| JP2580516B2 (en) | Real-time three-dimensional motion measuring apparatus and method | |
| JP2018205944A (en) | Moving image reproduction system |
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 |