Method, system, device and medium for processing human eye focus and deviation distance

Abstract

The present invention relates to a method, system, device and medium for processing the focus and deviation distance of human eyes, and the method comprises: obtaining the real-time position of the human eye gaze point; processing the object at the position of the human eye gaze point; simulating the focus time of the human eye during the processing to adapt to the focus time of different people. The present invention can improve the rendering reality of the focus point, reduce eye fatigue and reduce the sense of dizziness; can simulate the natural observation state of the human eye to avoid excessive eye tension, thereby causing dizziness; while blurring the picture, the fineness of the blurred part of the picture is reduced, thereby reducing the software's demand for computer performance; finely modeling and rendering the focused object to improve the realism of the virtual scene.

Description

Technical Field

[0001] The present invention belongs to the field of virtual vision, and in particular relates to a method, system, device and medium for processing the focusing and deviation distances of human eyes. Background Art

[0002] Using VR helmets to build three-dimensional virtual scenes is becoming more and more widely used in industries such as games, shopping, and movies. However, when using VR helmets, wearers often experience dizziness. By increasing the resolution of VR imaging and improving the image quality, the dizziness of the wearer can be alleviated to a certain extent, but it cannot be eradicated.

[0003] The human eye is similar to an adjustable focal length lens. When observing objects at different distances, the human eye uses eye muscles to quickly adjust the focal length of the eyeball so that the target object can be seen clearly. During the entire observation process, the human eye cannot focus on all objects in the field of view at the same time, but can only accurately focus on objects in a small range in the field of view, resulting in the so-called "depth of field", that is, the focused object is very clear, and the farther away from the object, the blurrier the picture people feel. This sense of depth of field can effectively ensure that the human eye is only sensitive to a small range of scenes at a time, and is not disturbed by the complex information of a large range of scenes. It is in a relaxed and natural observation state, and people will not feel dizzy even if they observe for a long time.

[0004] Traditional VR imaging is different from the natural observation state of the human eye. Since VR software does not know that the human eye will observe a certain distance and direction in the scene, in order to ensure that people can clearly observe all the images in the scene when observing the scene, traditional VR will render the entire picture clearly when imaging. As a result, there is no so-called "depth of field" in the picture. When observing, the human eye will see all the contents of the picture at the same time. A large amount of picture information will cause great pressure on the human eye. In addition, due to the size and volume of VR glasses, when wearing VR glasses, the human eye is very close to the imaging screen, and the human eye's own focus adjustment ability is greatly limited. The human eye cannot distinguish between the primary and secondary targets in the picture, which leads to eye tension during observation and excessive concentration in a short period of time, which will soon cause dizziness.

[0005] Therefore, in VR helmets, simulating the "depth of field" or "blurring" effect of natural observation process, so that the human eye can focus on the target object during observation, is the key to solving the problem of dizziness. At present, there is a method to directly blur the entire screen, that is, to make the clarity of the center of the picture always the highest, and gradually reduce the clarity of the edges to simulate the feeling of human eye observation. However, this method can only blur the picture away from the neutral position, and cannot blur the picture away from the focusing distance of the human eye; and this method assumes that the wearer's eyes are always looking straight at the picture, which does not conform to the actual observation situation of the wearer, so the improvement is limited. Summary of the invention

[0006] In order to overcome the problems existing in the prior art, the present invention provides a method, system, device and medium for processing the focusing and deviation distance of the human eye, which are used to overcome the existing defects.

[0007] A method for processing the focus and deviation distance of human eyes, used in virtual reality equipment, comprising the steps of:

[0008] S1. Get the real-time human eye gaze point position;

[0009] S2. Processing the object at the position of the human eye's gaze point;

[0010] S3. During the processing, the focusing time of human eyes is simulated to adapt to the focusing time of different people.

[0011] According to the above aspects and any possible implementation manner, an implementation manner is further provided, wherein S1 includes:

[0012] S11. Capture an image of a person's eyes and obtain the pupil center position through image processing;

[0013] S12. Taking the corneal reflection point as the base point of the relative position between the photographing device and the eyeball;

[0014] S13. The sight line vector coordinates can be obtained according to the pupil center, thereby determining the gaze point of the human eye.

[0015] According to the above aspects and any possible implementation manner, an implementation manner is further provided, wherein the S11 specifically includes:

[0016] A1. Input adjacent sequence of images;

[0017] A2. Determine whether the adjacent sequence of images is a bright pupil and dark pupil image, if so, proceed to the next step, otherwise return to A1;

[0018] A3. Under given filtering conditions, filter out the image content that has nothing to do with the bright pupil and dark pupil features;

[0019] A4. Perform multiple continuous filtering analyses on the image features of the bright pupil and the dark pupil to accurately determine the pupil center position;

[0020] A5. Perform positioning check on the eye area where the center of the pupil is located. If it is the eye area, the positioning check passes, and A61 and A62 are performed at the same time. Otherwise, the positioning check is performed to determine whether the number of times exceeds the maximum number. If the number exceeds the maximum number, return to A1, otherwise adjust the filter threshold and then return to A3;

[0021] A61 uses ellipse fitting to locate the pupil, pupil and cornea in adjacent sequence images;

[0022] A62 uses a centroid algorithm to calculate the eye area to obtain the position of the pupil, pupil, and cornea;

[0023] A63. Compare the sizes of the eyeball, pupil, and cornea positions obtained by A62 and A61 to see if they satisfy the preset relationship. If so, set a threshold to obtain the line of sight evaluation and extract the line of sight feature vector. Otherwise, return to A1.

[0024] According to the above aspects and any possible implementation, a further implementation is provided, wherein the extracted sight feature vector as follows:

[0025]

[0026] in: is the vector from pupil center to corneal reflection; is the ratio of the major and minor axes of the pupil ellipse, is the long axis of the pupil, is the short axis of the pupil; θ is the angle between the long axis of the pupil ellipse and the vertical direction; is the position of the pupil center in the image; is the position of the corneal reflection in the image.

[0027] As described above, the aspects and any possible implementation methods further provide an implementation method, which also includes: the accuracy of the human eye's gaze point is improved through pupil motion compensation, centroid positioning corneal reflection center solution, edge filtering and head position compensation.

[0028] According to the aspects described above and any possible implementation method, an implementation method is further provided, wherein processing the object at the position of the human eye's gaze point includes accurately rendering the object at the human eye's focusing distance and blurring the object at a distance that deviates from the focusing distance.

[0029] According to the aspects described above and any possible implementation method, an implementation method is further provided, wherein the focusing time is determined by the eye and head rotation speed, the moving direction vector between observation points and the zoom rate.

[0030] The present invention also provides a system for processing the focus and deviation distance of human eyes, wherein the system implements the method described and comprises the following modules:

[0031] An acquisition module is used to obtain the real-time gaze point position of human eyes;

[0032] A processing module, used for processing the object at the position of the human eye gaze point;

[0033] The simulation module is used to simulate the focusing time of human eyes during the processing to adapt to the focusing time of different people.

[0034] The present invention further provides an electronic device, comprising:

[0035] A memory storing executable instructions;

[0036] A processor is used to execute the executable instructions in the memory to implement the method.

[0037] The present invention also provides a computer storage medium, on which a computer program is stored, and the computer program is executed by a processor to implement the method described.

[0038] Beneficial effects of the present invention

[0039] The method for processing the focus and deviation distance of human eyes of the present invention is used in virtual reality equipment, and the method comprises: obtaining the real-time position of the human eye gaze point; processing the object at the position of the human eye gaze point; simulating the focus time of the human eye during the processing to adapt to the focus time of different people. The present invention can improve the rendering reality of the focus point, reduce eye fatigue and reduce the sense of dizziness; can simulate the natural observation state of the human eye to avoid excessive eye tension, thereby causing dizziness; while blurring the picture, the fineness of the blurred part of the picture is reduced, thereby reducing the software's requirements for computer performance; finely modeling and rendering the focused object to improve the fidelity of the virtual scene. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 is a schematic diagram of sight tracking of the present invention;

[0041] Figure 2 It is a schematic diagram of the sight line extraction process of the present invention;

[0042] Figure 3 This is a schematic diagram of the current VR helmet rendering method;

[0043] Figure 4 It is a schematic diagram of the focus area model refinement and edge model LOD hierarchical subdivision of the present invention;

[0044] Figure 5 It is a schematic diagram of the principle of pixel level of the edge of the focus point of the present invention;

[0045] Figure 6 is a depth of field principle diagram of the present invention;

[0046] Figure 7 It is a schematic diagram of human eye focusing simulation of the present invention. DETAILED DESCRIPTION

[0047] In order to better understand the technical solution of the present invention, the content of the present invention includes but is not limited to the specific implementation methods described below, and similar technologies and methods should be considered to be within the scope of protection of the present invention. In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, the following will be described in detail with reference to the accompanying drawings and specific embodiments.

[0048] It should be clear that the embodiments described in the present invention are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] A method for processing the focus and deviation distance of human eyes of the present invention is used in virtual reality equipment, and the method comprises the steps of:

[0050] S1. Get the real-time human eye gaze point position;

[0051] S2. Processing the object at the position of the human eye's gaze point;

[0052] S3. During the processing, the focusing time of human eyes is simulated to adapt to the focusing time of different people.

[0053] Preferably, the S1 includes:

[0054] S11. Capture an image of a person's eyes and obtain the pupil center position through image processing;

[0055] S12. Taking the corneal reflection point as the base point of the relative position between the photographing device and the eyeball;

[0056] S13. The sight line vector coordinates can be obtained according to the pupil center, thereby determining the gaze point of the human eye.

[0057] Specifically, the implementation process of the present invention is as follows:

[0058] Step 1: Get the real-time gaze point position

[0059] like Figure 1As shown, the ideal sight tracking system in the virtual reality device needs to meet the following requirements: accurate, reliable, comfortable, no initial calibration required and real-time. Based on the above requirements, the present invention adopts the sight tracking method of pupil-corneal reflection technology, uses a camera to shoot an eye image, and obtains the pupil center position through processing. Then the corneal reflection point is used as the base point of the relative position of the camera and the eyeball, and the sight vector coordinates can be obtained according to the pupil center obtained by image processing. The corneal reflection point refers to the reflection generated on the iris of the eyeball when the eye is irradiated by the light of the picture being viewed on the corneal surface of the eye, presenting a bright point or area. By observing and analyzing the position and shape of the corneal reflection, the position, direction and rotation angle of the eyeball can be understood. The position and direction of the eyeball can be known by the position of the corneal reflection point; the sight direction of the eyeball can be known by the position of the pupil center and the direction of the eyeball; the sight vector can be obtained by the position coordinates of the camera, the pupil center and the sight direction, and the sight vector can be obtained. According to the relative spatial position of the sight vector and the picture viewed by the eye, the position of the human eye gaze point can be determined, and the intersection of the sight vector and the picture viewed by the eye is the human eye gaze point.

[0060] Preferably, an image of a person's eye is captured, and the pupil center position is obtained by image processing as follows: an active infrared light source is used to detect and track the eye, the pupil is located by using the reflection phenomenon of the pupil to near-infrared light, two infrared light sources are used to generate bright pupil and dark pupil images, the background effect is eliminated by dynamically setting the threshold of the differential image, and then the differential image is filtered to eliminate external light interference, so that the pupil in the differential image can be effectively tracked, the edge of the pupil area is detected, and the corneal reflection is searched based on the grayscale near the eye area, the center of the corneal reflection is located by the centroid, and the pupil edge is filtered to eliminate the influence of the corneal reflection on the pupil edge contour, and the pupil center is located by ellipse fitting to obtain the sub-pixel center coordinate, such as Figure 2 As shown in Figure 2, the extraction process of the line of sight feature vector is as follows:

[0061] A1. Input an adjacent sequence of images, where the images are taken by a camera, and the adjacent images taken by the camera are the adjacent sequence of images;

[0062] A2. Determine whether the images in the adjacent sequence are bright pupil and dark pupil images. If so, proceed to the next step. Otherwise, return to A1 and make a judgment based on the physical characteristics of the bright pupil and the dark pupil. The pixel grayscale values ​​of the bright pupil and the dark pupil are different. The pupil grayscale value of the bright pupil feature is low, for example, red eyeballs appear when taking pictures. The pupil of the dark pupil feature is black, and no red eyeball phenomenon appears;

[0063] A3. Rule-based filtering, that is, under given filtering conditions, filter out the image content that has nothing to do with the bright pupil and dark pupil features. The filtering conditions are formulated according to actual needs;

[0064] A4. Perform multiple continuous filtering analysis on the image features of the bright pupil and the dark pupil to accurately determine the center position of the pupil, that is, the pupil features, that is, the image features of the bright pupil and the dark pupil; use the filtering conditions in A3 to perform multiple continuous filtering, and perform an analysis after each filtering to more accurately obtain the center position of the pupil;

[0065] A5. Perform positioning check on the eye area where the center of the pupil is located. If it is the eye area, the positioning check passes, and A61 and A62 are performed at the same time. Otherwise, the positioning check is performed to determine whether the number of times exceeds the maximum number. If the number exceeds the maximum number, return to A1, otherwise adjust the filter threshold and then return to A3;

[0066] A61 includes:

[0067] A611. Pupil motion compensation, that is, interpolation compensation of pupil positions in two adjacent sequence images;

[0068] A612. Edge filtering, i.e. filtering the pupil in the camera image to remove the influence of corneal reflection on the pupil edge contour;

[0069] A613. Ellipse fitting to locate pupil: infinitely fit the pupil and cornea size thresholds in the shape of an ellipse to obtain the best threshold process, draw a fitting eyeball, pupil in space and determine the position of the cornea;

[0070] A62 includes:

[0071] A621. Searching for corneal reflections near the determined eye area, that is, using the changes in the corneal reflection points to determine the direction of eye movement. Once the direction of eye movement is determined, the change pattern of the gaze point can be determined, thereby obtaining the positions of the pupil and cornea;

[0072] A622. Find the centroid to locate the corneal reflection. According to the position of the pupil and cornea, the centroid or geometric center of the image is obtained, that is, the corneal reflection is located at the centroid.

[0073] A63. Perform feature vector verification, that is, compare the size results of the eyeball, pupil, and cornea positions obtained by A62 using the centroid algorithm with those obtained by A61 using the ellipse fitting method to see if they meet the preset relationship. If so, set the threshold, obtain the line of sight evaluation, and extract the line of sight feature vector. Otherwise, return to A1.

[0074] Among them, the extracted line of sight feature vector The expression is as follows:

[0075] (1)

[0076] in:

[0077] The pupil center Corneal reflex A vector of is the ratio of the major and minor axes of the pupil ellipse, is the long axis of the pupil, is the short axis of the pupil; θ is the angle between the long axis of the pupil ellipse and the vertical direction; is the position of the pupil center in the image; is the position of the corneal reflection in the fixation point image.

[0078] In order to improve the accuracy of the gaze point position, the present invention also combines pupil motion compensation calculation, centroid positioning corneal reflection center solution, edge filtering and head position compensation. For example, when pupil motion compensation is used, the pupil sequence image taken by the camera is interpolated; head position compensation is the compensation of the head direction. The direction of the eyeball remains unchanged, and a slight movement of the head will also change the gaze point position, thereby further improving the accuracy of the gaze point position.

[0079] Step 2: Simulate the human eye focusing on an object

[0080] In life, human eyes have a depth of field effect when observing things, such as Figure 6 As shown in the figure, only objects within a certain range before and after the focus point can form a clear image in the human eye, and objects that are too far or too close to the focus point of the human eye are blurred. At present, the content presented by VR helmets is the same level of real-time rendering of all objects in the visible area, such as Figure 3 As shown in the figure, due to hardware limitations, the display quality has to give way to real-time rendering efficiency, and the overall accuracy of the virtual model is reduced, resulting in the object being focused on by the eyes not being detailed enough, and other objects in the peripheral vision being too clear. When observing, the human eye cannot distinguish between the primary and secondary targets in the picture. If the attention is over-focused in a short period of time, dizziness will soon occur. The present invention adds a defocus blur filter to the focus position to simulate the depth of field effect of the human eye. Figure 5 As shown; at the same time, the LOD level detail control is performed on the focus area of ​​the gaze point, the front and rear depth of field area, and the residual light area model and texture, such as Figure 4 As shown, reducing computer pressure and resetting rendering distribution can make the gaze object more realistic and detailed. The gaze point is relative to the position of the picture seen by the eyes, and the focus point is relative to the content displayed on the picture. For example, the gaze point is a point on the display when the eyes are looking at it, and the focus point is the object on the display when the eyes are looking at it. The focus point will change when the object moves, but the gaze point does not change.

[0081] After real-time capture of the gaze point, the object currently being viewed by the eyes is calculated. According to the relative position between the eye point and the object, objects that are too far or too close are blurred with the focal object as the center. An example of precision rendering is as follows: If there are three identical houses in front of you, the positions of 50 meters, 200 meters, and 400 meters away from the eye point are respectively, and the eyes are looking at the house at 200 meters, then the point where the eyes focus is 200 meters. If the house viewed at 200 meters is presented in detail, then fine rendering is performed. The house at that location may have 10,000 feature faces. At this time, the feature faces of the houses at 50 meters and 400 meters may be 1,000 and 100 feature faces respectively. An example of blur processing is as follows: In order to make the texture of the house at 200 meters clear and to see the green plants on the windowsill, and to make the houses at 50 meters and 400 meters blurred, the green plants on the windowsill at 50 meters are set to a large area of ​​blurred green, and the green plants on the windowsill at 400 meters are set to a blurred green dot. The rendering threshold can be the change of the distance around the focus object, and the blur and model accuracy also change accordingly to ensure the highest accuracy and clarity of the focus object. Therefore, the object is processed separately from two levels: one is the fine level of the object model (the number of vertices and the fine level of texture), and the other is the clarity of the object imaging. Objects that are too far from the focus have high clarity and high accuracy; objects that are too close have the opposite. The following two formulas are used to calculate the spatial distance between the eye position and the focus object;

[0082] (2)

[0083] (3)

[0084] Assume the eye point position is (eye point space coordinates in the virtual image), the viewing direction is , that is, the direction vector from the eye point to the object, For the eyes To the object A vector of For Objects The spatial coordinates of the object the eye is looking at, It is half of the angle of the visual cone. The visual cone is the cone formed by the eyes seeing the edge of the entire picture (top, bottom, left, right). Here is the angle between the line of sight and this cone. The eyes point to the object The object is processed by multiple levels of detail LOD, the effect is as follows Figure 4 As shown in the following formula:

[0085] = (4)

[0086] The range of α is determined by the amount of model refinement, The eyes point to the object Distance, h 1 is the distance from the focal object;

[0087] in It is the depth of field distance before and after the focus position, and the picture clarity is processed as follows:

[0088] (5)

[0089] in Indicates distance from an object Clarity at different distances, is a positive number, indicating how quickly the clarity decreases with distance. is a constant representing the characteristics of the imaging system, such as the parameters of the imaging screen, the resolution of the computer output, etc. Figure 5 shown.

[0090] Step 3: Simulate the focusing time of the human eye. When the observation point moves, add variable focusing time to adapt to the focusing time of different people, so as to simulate the time difference required when moving from a fixed point to another observation target. Among them, the fixation point is the position on the screen where the eyes see it, and the observation point is the position of the human eye in the virtual space. The closer it is to the actual position of the human eye, the better the immersion.

[0091] Physiologists have found that the time it takes for a normal human eye to see an object clearly from a very far distance to a very close distance is no more than 0.5 seconds, and it varies from person to person and from time to time. As the human eye's observation point shifts, the image zooms with the different spatial positions of the object being observed, thus affecting the focusing time. The factors mainly include the speed of human eye and head rotation. , moving direction vector between observation points and zoom rate , the relationship between the above variables is as follows:

[0092] (6).

[0093] like Figure 7 As shown, the human eye focusing simulation effect obtained by the present invention shows that the method can focus on a point, such as a point on a clear cactus, specifically a white dot on the cactus. Specifically, the present invention can focus on a pixel on the gaze point image and blur the pixels around the pixel.

[0094] As an embodiment disclosed in the present invention, the present invention further discloses a system for processing the focus and deviation distance of human eyes, wherein the system implements the method described, and comprises the following modules:

[0095] An acquisition module is used to obtain the real-time gaze point position of human eyes;

[0096] A processing module, used for processing the object at the position of the human eye gaze point;

[0097] The simulation module is used to simulate the focusing time of human eyes during the processing to adapt to the focusing time of different people.

[0098] As an embodiment disclosed in the present invention, the present invention further discloses an electronic device, the electronic device comprising:

[0099] A memory storing executable instructions;

[0100] A processor is used to execute the executable instructions in the memory to implement the method described in the present invention.

[0101] As an embodiment disclosed in the present invention, the present invention further discloses a computer storage medium, on which a computer program is stored, and the computer program is executed by a processor to implement the method described in the present invention.

[0102] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.

[0103] The above description shows and describes several preferred embodiments of the present invention, but as mentioned above, it should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in various other combinations, modifications and environments, and can be modified within the scope of the application concept described herein through the above teachings or the technology or knowledge of the relevant field. The changes and modifications made by those skilled in the art shall not depart from the spirit and scope of the present invention, and shall be within the scope of protection of the claims attached to the present invention.

Claims

1. A method for processing the focus and deviation distance of human eyes, used in virtual reality equipment, characterized in that: The method comprises the steps of: S1. Get the real-time human eye gaze point position; S2. Processing the object at the human eye's gaze point, including precision rendering at the human eye's focus distance and blurring at the distance that deviates from the focus distance, specifically including: by capturing the gaze point in real time, calculating the object the eye is currently looking at, and blurring objects that are too far or too close with the focus object as the center according to the relative position of the eye point and the object: using the following two formulas to calculate the spatial distance between the eye position and the focus object: （2） （3） Assume the eye point position is , the observation direction is , that is, the eye position toward the object The direction vector of For the eyes To the object A vector of For Objects The spatial coordinates of the object the eye is looking at, The object is processed by multiple levels of detail LOD, which is half of the angle of the viewing cone, as follows: = （4） The range of α is determined by the amount of model refinement, The eyes point to the object , h1 is the distance from the object; It is the depth of field distance before and after the focus position, and the picture clarity is processed as follows: (5) in Indicates distance from an object Clarity at different distances, is a positive number, indicating how quickly the clarity decreases with distance. is a constant; S3. During the processing, the focusing time of human eyes is simulated to adapt to the focusing time of different people.

2. The method for processing the focusing and deviation distance of human eyes according to claim 1, characterized in that: The S1 includes: S11. Capture an image of a person's eyes and obtain the pupil center position through image processing; S12. Taking the corneal reflection point as the base point of the relative position between the photographing device and the eyeball; S13. Obtain the sight line vector coordinates according to the pupil center, thereby determining the gaze point of the human eye.

3. The method for processing the focusing and deviation distance of human eyes according to claim 2, characterized in that: The S11 specifically includes: A1. Input adjacent sequence of images; A2. Determine whether the adjacent sequence of images is a bright pupil and dark pupil image, if so, proceed to the next step, otherwise return to A1; A3. Under given filtering conditions, filter out the image content that has nothing to do with the bright pupil and dark pupil features; A4. Perform multiple continuous filtering analyses on the image features of the bright pupil and the dark pupil to accurately determine the pupil center position; A5. Perform positioning check on the eye area where the center of the pupil is located. If it is the eye area, the positioning check passes, and A61 and A62 are performed at the same time. Otherwise, the positioning check is performed to determine whether the number of times exceeds the maximum number. If the number exceeds the maximum number, return to A1, otherwise adjust the filter threshold and then return to A3; A61 uses ellipse fitting to locate the pupil, pupil and cornea in adjacent sequence images; A62 uses a centroid algorithm to calculate the eye area to obtain the position of the pupil, pupil, and cornea; A63. Compare the sizes of the eyeball, pupil, and cornea positions obtained by A62 and A61 to see if they satisfy the preset relationship. If so, set a threshold to obtain the line of sight evaluation and extract the line of sight feature vector. Otherwise, return to A1.

4. The method for processing the focusing and deviation distance of human eyes according to claim 3, characterized in that: Extracted line of sight feature vector as follows: ， in: is the vector from pupil center to corneal reflection; is the ratio of the major and minor axes of the pupil ellipse, is the long axis of the pupil, is the short axis of the pupil; is the position of the pupil center in the image; is the position of the corneal reflection in the image.

5. The method for processing the focusing and deviation distance of human eyes according to claim 2, characterized in that: Also includes: The accuracy of the human eye's gaze point is improved by pupil motion compensation, centroid positioning corneal reflection center solution, edge filtering and head position compensation.

6. The method for processing the focusing and deviation distance of human eyes according to claim 1, characterized in that: The focusing time is determined by the eyeball and head rotation speed, the moving direction vector between observation points and the zoom rate.

7. A system for processing the focus and deviation distance of human eyes, characterized in that: The system implements the method described in any one of claims 1 to 6, and includes the following modules: An acquisition module is used to obtain the real-time gaze point position of human eyes; A processing module, used for processing the object at the position of the human eye gaze point; The simulation module is used to simulate the focusing time of the human eye during the processing to adapt to the focusing time of different people.

8. An electronic device, characterized in that: The electronic device comprises: A memory storing executable instructions; A processor, wherein the processor runs the executable instructions in the memory to implement the method according to any one of claims 1 to 6.

9. A computer storage medium, characterized in that: The medium stores a computer program, and the computer program is executed by a processor to implement the method according to any one of claims 1 to 6.

Images