Test methods, apparatus, systems and electronic devices for latency testing of naked-eye 3D systems
By acquiring time-series data of the actual human eye and tracking coordinates, generating trajectory curves, and calculating the overall latency, the latency problem of naked-eye 3D systems is solved, and efficient and accurate system latency testing is achieved.
Patent Information
- Application Number
- CN202611290205.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-24
- Publication Date
- 2026-11-17
AI Technical Summary
The naked-eye 3D system has a time lag of tens to hundreds of milliseconds, which causes interlaced image misalignment and 3D ghosting problems. Existing testing methods are highly subjective, inefficient, and require a large amount of work for individual testing of each product.
By acquiring time-series data of the actual coordinates of the human eye and the tracking coordinates, a trajectory curve is generated, and overall latency data is calculated. Combined with an automated deployment scheme, subjectivity is reduced, and testing accuracy and efficiency are improved.
It achieves high-precision, low-subjectivity, and efficient naked-eye 3D system latency testing, and is suitable for batch testing of the same batch of products.
Smart Images

Figure CN123123218A_ABST
Abstract
Description
Technical Field
[0001] This disclosure belongs to the field of latency testing technology, specifically relating to a method, apparatus, system, and electronic device for testing latency of a naked-eye 3D system. Background Technology
[0002] Naked-eye 3D systems typically use left and right parallax maps to form a left-right interlaced image. Through a special cylindrical lens beam splitting principle, the left view is projected onto the left eye and the right view is projected onto the right eye. Eye tracking algorithms are often embedded to obtain the position of the human eye, and the left-right interlaced image is updated in real time according to the position of the human eye. The output left-right interlaced image is required to strictly correspond to the position of the left and right eyes.
[0003] However, the system takes tens to hundreds of milliseconds to run. There is a lag of tens to hundreds of milliseconds between the system output of the human eye position and the actual human eye position. Moreover, the delay time varies with different system configurations. This directly leads to problems such as misalignment of interlaced images and 3D ghosting when viewed by the human eye. Therefore, it is necessary to eliminate this time delay.
[0004] In related technologies, this time delay is generally tested by combining manual observation, which is highly subjective. Different people will obtain different results, leading to low accuracy and low efficiency in time delay testing. Furthermore, this testing method requires testing each product individually, which is labor-intensive and inefficient for mass-produced products in the same batch. Summary of the Invention
[0005] This disclosure provides a method for testing the latency of a naked-eye 3D system, including:
[0006] Acquire first time series data; wherein, the first time series data includes multiple sets of actual human eye coordinate data and corresponding first timestamps; the actual human eye coordinate data is the coordinate data of the human eye in the human eye model obtained during the movement of the human eye model; the first timestamp is the acquisition time of the corresponding actual human eye coordinate data;
[0007] Acquire second time series data; wherein, the second time series includes multiple sets of human eye tracking coordinate data and corresponding second timestamps; the human eye tracking coordinate data is the coordinate data of the human eye in the human eye model calculated according to the human eye tracking algorithm during the movement of the human eye model; the second timestamp is the time when the corresponding human eye tracking coordinate data and the corresponding interlaced image are displayed on the display device set relative to the human eye model, and the interlaced image is obtained and displayed on the display device based on the human eye tracking coordinate data and the multi-viewpoint image synthesis algorithm;
[0008] A first trajectory curve is generated based on the first time series data, and a second trajectory curve is generated based on the second time series data; wherein, the first trajectory curve is the relationship curve between the actual coordinates of the human eye and time; and the second trajectory curve is the relationship curve between the human eye tracking coordinates and time.
[0009] The overall latency data is obtained based on the time difference between the first trajectory curve and the second trajectory curve.
[0010] In some embodiments, each set of coordinate data in the plurality of sets of actual human eye coordinate data is obtained based on the initial position and motion parameters of the human eye model.
[0011] In some embodiments, at the initial position of the human eye model, the line connecting the two eyes in the human eye model is parallel to the plane where the display device is located, and the line connecting the reference point of the human eye model and the reference point of the display device is perpendicular to the plane where the display device is located.
[0012] In some embodiments, the motion parameters of the human eye model include the direction of motion, the time of motion, and the speed of motion; the direction of motion of the human eye model is parallel to the plane on which the display device is located, and the speed of motion of the human eye model is uniform.
[0013] In some embodiments, the human eye model is driven by a drive device.
[0014] In some embodiments, the second time series data is obtained from multiple sets of captured images, which are obtained by the camera device capturing multiple images of the screen displayed on the display device.
[0015] In some embodiments, the second time series data is obtained by performing artificial intelligence recognition on each of the multiple sets of captured images.
[0016] In some embodiments, the method for testing the latency of a glasses-free 3D system further includes:
[0017] Acquire test latency data; the test latency data includes the time consumed by the human eye tracking algorithm, the time consumed by the multi-viewpoint image synthesis algorithm, the screen refresh time of the display device, and the shooting time of the camera device.
[0018] The fixed latency data of the system is obtained based on the difference between the overall latency data and the test latency data.
[0019] In some embodiments, generating the first trajectory curve based on the first time series data includes:
[0020] The actual coordinates of the human eye are obtained in the first direction, which is used as the first coordinate data; the first direction is the direction of motion of the human eye model.
[0021] A first location time scatter plot is obtained based on multiple first coordinate data and corresponding first timestamps;
[0022] The first trajectory curve is obtained by fitting the position scatter points in the first position time scatter distribution map.
[0023] In some embodiments, generating the second trajectory curve based on the second time series data includes:
[0024] The coordinate data of human eye tracking in the first direction is used as the second coordinate data; the first direction is the direction of movement of the human eye model; a scatter plot of time distribution of the second position is obtained based on multiple second coordinate data and corresponding second timestamps.
[0025] The second trajectory curve is obtained by fitting the position scatter points in the second position time scatter plot.
[0026] In some embodiments, obtaining the overall time delay data based on the time difference between the first trajectory curve and the second trajectory curve includes:
[0027] The time difference between the first trajectory curve and the second trajectory curve under multiple preset coordinate data is obtained to obtain multiple time differences;
[0028] The average value of the multiple time differences is obtained, and the overall delay data is the average value of the multiple time differences.
[0029] This disclosure provides a testing device for latency of a naked-eye 3D system, comprising:
[0030] The first data acquisition module is configured to: acquire first time series data; wherein, the first time series data includes multiple sets of actual human eye coordinate data and corresponding first timestamps; the actual human eye coordinate data is the coordinate data of the human eye in the human eye model acquired during the movement of the human eye model; the first timestamp is the acquisition time of the corresponding actual human eye coordinate data;
[0031] The second data acquisition module is configured to acquire second time series data; wherein the second time series includes multiple sets of human eye tracking coordinate data and corresponding second timestamps; the human eye tracking coordinate data is the coordinate data of the human eye in the human eye model calculated according to the human eye tracking algorithm during the movement of the human eye model; the second timestamp is the time when the corresponding human eye tracking coordinate data and the corresponding interlaced image are displayed on a display device positioned opposite to the human eye model, wherein the interlaced image is obtained and displayed on the display device based on the human eye tracking coordinate data and a multi-viewpoint image synthesis algorithm; wherein the first trajectory curve is the relationship curve between the actual coordinates of the human eye and time; the second trajectory curve is the relationship curve between the human eye tracking coordinates and time;
[0032] The curve acquisition module is configured to: generate a first trajectory curve based on the first time series data, and generate a second trajectory curve based on the second time series data;
[0033] The system latency acquisition module is configured to acquire overall latency data based on the time difference between the first trajectory curve and the second trajectory curve.
[0034] This disclosure provides a testing system for latency of a naked-eye 3D system, including:
[0035] The drive mechanism is configured to move the human eye model.
[0036] The display device is configured to display continuously updated eye-tracking coordinate data and interlaced images; the eye-tracking coordinate data is the coordinate data of the human eye in the eye model calculated according to the eye-tracking algorithm during the movement of the human eye model driven by the driving device; the interlaced images are obtained based on the eye-tracking coordinate data and a multi-viewpoint image synthesis algorithm.
[0037] The camera device is configured to: capture images displayed on the display device; and
[0038] The device for testing the latency of the aforementioned naked-eye 3D system.
[0039] This disclosure provides an electronic device, including:
[0040] At least one processor; and
[0041] A memory communicatively connected to the at least one processor; wherein,
[0042] The memory stores one or more computer programs that can be executed by the at least one processor, and the one or more computer programs are executed by the at least one processor to enable the at least one processor to perform the above-described method for testing the latency of a naked-eye 3D system. Attached Figure Description
[0043] The accompanying drawings are provided to further understand the present disclosure and form part of the specification. They are used together with the following detailed description to explain the present disclosure, but do not constitute a limitation thereof.
[0044] Figure 1 A flowchart illustrating a method for testing latency in a glasses-free 3D system, as provided in this embodiment of the disclosure;
[0045] Figure 2 A flowchart of another method for testing latency of a naked-eye 3D system provided in this embodiment of the disclosure;
[0046] Figure 3 A flowchart illustrating another method for testing latency in a glasses-free 3D system provided in this embodiment of the present disclosure;
[0047] Figure 4A A schematic diagram of the first trajectory curve and the second trajectory curve in another method for testing the latency of a naked-eye 3D system provided in an embodiment of this disclosure;
[0048] Figure 4B A flowchart illustrating another method for testing latency in a glasses-free 3D system provided in this embodiment of the present disclosure;
[0049] Figure 5 A flowchart illustrating another method for testing latency in a glasses-free 3D system provided in this embodiment of the present disclosure;
[0050] Figure 6 A block diagram of a device for testing the latency of a naked-eye 3D system provided in an embodiment of this disclosure;
[0051] Figure 7 A block diagram of a test system for latency of a naked-eye 3D system provided in this embodiment of the present disclosure;
[0052] Figure 8 This is a block diagram of an electronic device provided in an embodiment of the present disclosure. Detailed Implementation
[0053] To enable those skilled in the art to better understand the technical solutions of this disclosure, the disclosure will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0054] Unless otherwise defined, the technical or scientific terms used in this disclosure shall have the ordinary meaning understood by one of ordinary skill in the art to which this disclosure pertains. The terms “a,” “an,” “an,” “the,” and similar words used in this disclosure do not indicate quantity limitation and may indicate singular or plural. The terms “comprising,” “including,” “having,” and any variations thereof used in this disclosure are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units not listed, or may include other steps or units inherent to such processes, methods, products, or devices. The terms “connected,” “linked,” “coupled,” and similar words used in this disclosure are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. “Multiple” in this disclosure refers to two or more. “And / or” describes the relationship between related objects, indicating that three relationships may exist; for example, “A and / or B” can indicate: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following objects are in an "or" relationship. The terms "first," "second," "third," etc., used in this disclosure are merely to distinguish similar objects and do not represent a specific ordering of objects. "Above," "below," "left," "right," etc., are only used to indicate relative positional relationships; when the absolute position of the described objects changes, the relative positional relationship may also change accordingly.
[0055] Those skilled in the art should understand that the response to an instruction (or response to an operation) described in the embodiments of this disclosure refers to the situation when an instruction (or operation) is received.
[0056] like Figure 1 As shown, the method for testing the latency of a naked-eye 3D system provided in some embodiments of this disclosure includes steps S11 to S14.
[0057] S11. Obtain the first time series data.
[0058] In S11, the first time series data includes multiple sets of actual eye coordinate data and corresponding first timestamps. The actual eye coordinate data are the coordinate data of the human eye in the human eye model obtained during the movement of the human eye model; the first timestamp is the acquisition time of the corresponding actual eye coordinate data.
[0059] For example, a human eye model refers to a model that can simulate the characteristics of a human eye, such as a model that includes two human eyes. Or, for example, it could be a face model that includes two eyes, and the face model could be a three-dimensional human head model or a two-dimensional face image.
[0060] For example, the multiple sets of actual human eye coordinate data are coordinate data in a first coordinate system. In the first coordinate system, the origin O is the reference point of the display device, the X-axis is the direction of the line connecting the two eyes in the human eye model, the Z-axis is the direction perpendicular to the plane where the display device is located, and the Y-axis is perpendicular to the OXZ plane. The reference point of the display device can be any point on the display device, such as the center of the screen. The plane where the display device is located is the plane where the display surface of the display device is located.
[0061] For example, the actual coordinate data of the human eye can refer to the position coordinate data of the center point of the pupil.
[0062] In some embodiments, each set of coordinate data in the multiple sets of actual human eye coordinate data is obtained based on the initial position and motion parameters of the human eye model.
[0063] Understandably, the initial position of the human eye model can be preset, that is, the initial position of the human eye model is known. Thus, the initial coordinate data of the two eyes in the human eye model can also be known. Therefore, the coordinate data of each set of actual human eye coordinate data can be calculated from the initial coordinate data of the human eye and the motion parameters of the human eye model.
[0064] In some embodiments, at the initial position of the human eye model, the line connecting the two eyes in the human eye model is set parallel to the plane where the display device is located, and the line connecting the reference point of the human eye model and the reference point of the display device is perpendicular to the plane where the display device is located.
[0065] Understandably, by setting up the human eye model in the manner described in the above embodiments, in the first coordinate system, it is only necessary to know the positional relationship between the two eyes in the human eye model and the reference point of the human eye model to know the coordinates of the initial position of the two eyes in the human eye model. This setting can reduce the amount of computation during the testing process.
[0066] For example, the reference point of the display device can be any point on the display device (e.g., the center point of the screen), and the reference point of the human eye model can be any point on the line segment connecting the two eyes (e.g., the center of the eyebrows, the left eye). For ease of calculation, the center point of the display device screen can be used as the reference point of the display device, and the midpoint of the line segment connecting the two eyes can be used as the reference point of the human eye model.
[0067] In some embodiments, the motion parameters of the human eye model include the direction of motion, the time of motion, and the speed of motion. The direction of motion of the human eye model is parallel to the plane on which the display device is located, and the speed of motion of the human eye model is uniform.
[0068] Understandably, for ease of calculation, the movement direction of the human eye model can be set to be parallel to the plane where the display device is located, such as the movement direction being parallel to the X-axis or Y-axis. The following embodiment uses the movement direction of the human eye model being parallel to the X-axis as an example for illustrative purposes. Since the movement direction does not cause changes in the Y-axis data and Z-axis data, only the X-axis data changes between the previously obtained actual eye coordinate data and the current actual eye coordinate data. Therefore, only the X-axis data needs to be obtained.
[0069] Understandably, when the human eye model moves at a constant speed, the displacement can be obtained from the movement time and speed. Adding or subtracting the displacement from the previously acquired X-axis data of the human eye's actual coordinates gives the current X-axis data of the human eye's actual coordinates. If the movement direction is towards the positive X-axis, the X-axis data from the previously acquired human eye's actual coordinates can be added to the displacement to obtain the current X-axis data of the human eye's actual coordinates. If the movement direction is towards the negative X-axis, the displacement can be subtracted from the X-axis data from the previously acquired human eye's actual coordinates to obtain the current X-axis data of the human eye's actual coordinates.
[0070] For example, the acquisition period for the actual coordinates of the human eye can be set, that is, the movement time of each time is fixed. For example, the acquisition period is 1 millisecond, that is, the human eye model moves at a constant speed for 1 millisecond each time. When the human eye model moves at a constant speed, the displacement of the human eye model is the same in each period. Therefore, the displacement can be added to or subtracted from the previously acquired actual coordinates of the human eye to obtain the actual coordinates of the human eye in the current period. Finally, the multiple sets of actual coordinates of the human eye obtained are actually the actual coordinates of the human eye in each acquisition period.
[0071] In some embodiments, the human eye model is driven by a drive device.
[0072] Understandably, the driving device can be preset to drive the human eye model to move in the preset direction and speed.
[0073] For example, the driving device can be a robotic arm, which can directly acquire the first time series data. For instance, the method of acquiring the actual coordinate data of the human eye once per millisecond by the robotic arm can be as follows: the Cartesian space motion trajectory (such as a straight line, arc or spline curve) of the human eye model starting from a preset initial position and the velocity / acceleration parameters are preset in the host computer or controller of the robotic arm. The controller discretizes the trajectory into angular displacement commands of each joint in real time through inverse kinematics and sends them to the driver for execution at a servo cycle of 1 kHz (i.e., every millisecond). At the same time, the host computer of the robotic arm synchronously reads the encoder feedback value through a real-time communication bus (such as EtherCAT), calculates the three-dimensional coordinates of the human eye model in the first coordinate system through forward kinematics, and records and stores the timestamp and coordinate values as millisecond-level time series data.
[0074] S12. Obtain the second time series data.
[0075] In S12, the second time series includes multiple sets of human eye tracking coordinate data and corresponding second timestamps; the human eye tracking coordinate data is the coordinate data of the human eye in the human eye model calculated according to the human eye tracking algorithm during the movement of the human eye model; the second timestamp is the time when the corresponding human eye tracking coordinate data and the corresponding interlaced image are displayed on the display device set relative to the human eye model, and the interlaced image is obtained and displayed on the display device based on the human eye tracking coordinate data and the multi-viewpoint image synthesis algorithm.
[0076] For example, the eye-tracking coordinate data and the actual eye coordinate data are data in the same coordinate system. That is, the eye-tracking coordinate data is coordinate data in the first coordinate system.
[0077] For example, an eye-tracking algorithm is embedded in the naked-eye 3D system. First, an image frame containing a face is captured by a camera and preprocessed, such as by illumination correction. Then, a target detection algorithm is used to locate the face and left and right eye regions in the image, and further extracts key physiological features such as the pupil center and corneal reflection point. Based on this, the gaze direction is calculated using a traditional geometric model or a deep learning network, and according to a mapping function established in advance through a calibration step, this direction is transformed from the camera coordinate system to the screen display coordinate system (e.g., a two-dimensional coordinate system formed by the origin, X-axis, and Y-axis in the first coordinate system). Finally, the two-dimensional viewpoint coordinates of the human eye in the screen display coordinate system are output. It can be assumed that the Z-axis value of the eye-tracking coordinate data in the first coordinate system is 0.
[0078] Understandably, when viewing a screen, the human eye needs to maintain a certain distance from the screen, so the Z-axis value in the actual eye coordinate data is not 0. If the movement direction of the human eye model is set to be parallel to the X-axis, the Z-axis value of the actual eye coordinate data is a constant value. No matter how the human eye model moves, the Z-axis value of the actual eye coordinate data remains unchanged. Therefore, the Z-axis value in the calculated coordinates of the human eye can be ignored, and subsequent calculations only need to consider the X-axis value of the actual eye coordinate data.
[0079] Understandably, the purpose of calculating eye-tracking coordinates in a glasses-free 3D system is to use this coordinate data as input parameters to generate a strictly corresponding interlaced image for both eyes in real time. The ultimate goal is to display this interlaced image on a display device, presenting a 3D effect to the user without their eyes. Therefore, latency needs to consider not only the time of the eye-tracking algorithm but also the time required to generate and display the interlaced image. Thus, the second timestamp is the time when the corresponding eye-tracking coordinate data and the corresponding interlaced image are displayed on the display device, which is positioned relative to the eye model. In other words, the second timestamp, the eye-tracking coordinates, and the interlaced image are in one-to-one correspondence, and the second timestamp is the time when the interlaced image is displayed on the display device.
[0080] In some embodiments, the second time series data is obtained from multiple sets of captured images, which are obtained by the camera device capturing multiple images of the screen displayed on the display device.
[0081] For example, the camera device can be a high-speed camera, and the shooting frequency can be set according to the specific scene and the camera's hardware parameters. For example, it can shoot once every 6 milliseconds. The image displayed on the display device includes a one-to-one corresponding second timestamp, human eye tracking coordinates, and the left and right eye interlaced image obtained based on the human eye tracking coordinates.
[0082] In some embodiments, the second time series data is obtained by performing artificial intelligence recognition on each captured image in a plurality of captured images.
[0083] Understandably, testing typically takes several minutes. With a high shooting frequency, a large number of images are captured within the testing time. Manually identifying and recording each image would be time-consuming and labor-intensive. Using artificial intelligence, not only can numbers be identified from massive amounts of images, but data can also be output according to a preset format. This approach improves the speed of acquiring second-time-series data, simplifies the testing process, and increases testing efficiency.
[0084] S13. Generate a first trajectory curve based on the first time series data, and generate a second trajectory curve based on the second time series data.
[0085] In S13, the first trajectory curve is the relationship between the actual coordinates of the human eye and time; the second trajectory curve is the relationship between the coordinates tracked by the human eye and time.
[0086] Understandably, the first time series data contains the relationship between the actual position of the human eye and time, and the second time series data contains the relationship between the tracking position of the human eye and time. By taking time as one variable and position as another variable, the first trajectory curve and the second trajectory curve can be obtained.
[0087] In some embodiments, such as Figure 2 As shown, the method for generating the first trajectory curve based on the first time series data in S13 may include steps S21 to S23.
[0088] S21, obtain the coordinate data of the actual human eye in the first direction, and use it as the first coordinate data.
[0089] In S21, the first direction is the direction of motion of the human eye model.
[0090] For example, taking the first direction as parallel to the X-axis, the X-axis coordinate data in the actual human eye coordinate data can be taken as the first coordinate data. When the human eye model moves along the first direction, the Y-axis data and Z-axis data of the human eye model remain almost unchanged, which is more convenient for subsequent calculations.
[0091] S22, obtain the first location time scatter distribution map based on multiple first coordinate data and corresponding first timestamps.
[0092] For example, by using time as the horizontal axis variable and the first coordinate data as the vertical axis variable to form a second coordinate system, and marking multiple first coordinate data and corresponding first timestamps in the second coordinate system, a scatter plot of the first location time can be obtained.
[0093] S23, Fit the position scatter points in the first position time scatter distribution map to obtain the first trajectory curve.
[0094] Understandably, a curve can be obtained by fitting the scatter plot, and the first trajectory curve obtained by fitting can simulate the real trajectory of human eye movement to the greatest extent. Thus, the first trajectory curve is the relationship curve between the X-axis value in the actual coordinate data of the human eye and time.
[0095] In some embodiments, such as Figure 3 As shown in S13, the method for generating the second trajectory curve based on the second time series data includes S31~S33.
[0096] S31, acquire the coordinate data of the human eye tracking coordinate data in the first direction, and use it as the second coordinate data.
[0097] In S31, the first direction is the direction of motion of the human eye model.
[0098] S32, obtain the second location time scatter distribution map based on multiple second coordinate data and the corresponding second timestamp.
[0099] S33, the second trajectory curve is obtained by fitting the position scatter points in the second position time scatter distribution map.
[0100] Referring to the relevant descriptions of steps S21 to S23, the second trajectory curve is the curve showing the relationship between the X-axis value and time in the human eye tracking coordinate data.
[0101] S14. Obtain the overall delay data based on the time difference between the first trajectory curve and the second trajectory curve.
[0102] Understandably, since human eye tracking coordinate data is derived from actual human eye coordinate data, the tracking coordinate data lags behind the actual human eye data. Therefore, if... Figure 4A As shown, the first coordinate is the X-axis coordinate, and the data in the first coordinate is the same value (e.g., Figure 4A In the case of x0), the second timestamp (tx02) on the second trajectory curve corresponding to the value is different from the first timestamp (tx01) on the first trajectory curve corresponding to the value. Obtaining the difference between these two timestamps (tx02 minus the value of tx01) is the overall latency data in the current test process.
[0103] In some embodiments, such as Figure 4B As shown, the implementation method of step S14 may include steps S41 to S42.
[0104] S41, obtain the time difference between the first trajectory curve and the second trajectory curve under multiple preset coordinate data, and obtain multiple time differences.
[0105] S42, obtain the average of multiple time differences, and the overall latency data is the average of multiple time differences.
[0106] Understandably, if only one time difference is calculated as the overall delay data, the accuracy of the data is not high. Therefore, multiple time differences can be calculated, and the average value can be taken as the overall delay data, which can improve the accuracy of the data.
[0107] For example, such as Figure 4A As shown, multiple vertical axis variables (i.e., X-axis coordinates) can be selected within the range of [-30mm, 30mm], and multiple time difference data under multiple vertical axis variables can be obtained accordingly. The overall time delay data can be obtained by averaging these multiple time difference data.
[0108] The method for testing the latency of a naked-eye 3D system provided in some embodiments of this disclosure can achieve latency testing through an automated deployment scheme (such as deploying a driving device to control the movement of the human eye model, or deploying a high-speed camera to capture the display screen of the display device). It does not require manual observation, reduces subjectivity, reduces the overall latency data test error, and improves test accuracy and efficiency.
[0109] like Figure 5 As shown, the method for testing the latency of a naked-eye 3D system provided in some embodiments of this disclosure further includes steps S51 to S52.
[0110] S51, obtain test latency data.
[0111] In S51, the test latency data includes the time consumed by the human eye tracking algorithm, the time consumed by the multi-viewpoint image synthesis algorithm, the screen refresh time of the display device, and the shooting time of the camera device.
[0112] Understandably, the overall latency data obtained from the tests conducted according to steps S11-S14 includes not only the fixed latency data of the naked-eye 3D system but also the test latency data, which is related to the algorithms and hardware used in the testing process. The test latency data can be calculated. Algorithm latency data includes the time consumed by the eye-tracking algorithm and the time consumed by the multi-view image synthesis algorithm, which can be obtained from the algorithm's own data. Hardware latency data used in the testing process includes the screen refresh time of the display device and the shooting time of the camera device. Adding the time consumed by the eye-tracking algorithm, the multi-view image synthesis algorithm, the screen refresh time of the display device, and the shooting time of the camera device yields the test latency data.
[0113] S52, based on the difference between the overall latency data and the test latency data, obtains the system's fixed latency data.
[0114] Understandably, subtracting the test latency data from the overall experimental data yields the system's fixed latency data.
[0115] Through steps S51 to S52, for products related to naked-eye 3D systems produced in the same batch, only a few products need to be tested, and the fixed latency data of the product can be determined by taking the average value. This fixed latency parameter can then be used as the fixed latency parameter for products related to naked-eye 3D systems produced in the same batch.
[0116] In addition, this disclosure also provides a testing device, electronic device, and computer-readable storage medium for testing the latency of a naked-eye 3D system. All of the above can be used to implement any of the testing methods for the latency of a naked-eye 3D system provided in this disclosure. The corresponding technical solutions and descriptions are described in the corresponding section of the method and will not be repeated here.
[0117] like Figure 6As shown, this embodiment of the present disclosure provides a testing device for the latency of a naked-eye 3D system. The naked-eye 3D system latency testing device 600 includes a first data acquisition module 601, a second data acquisition module 602, a curve acquisition module 603, and a system latency acquisition module 604.
[0118] The first data acquisition module 601 is configured to acquire first time series data; wherein, the first time series data includes multiple sets of actual human eye coordinate data and corresponding first timestamps; the actual human eye coordinate data is the coordinate data of the human eye in the human eye model acquired during the movement of the human eye model; the first timestamp is the acquisition time of the corresponding actual human eye coordinate data.
[0119] The second data acquisition module 602 is configured to: acquire second time series data; wherein, the second time series includes multiple sets of human eye tracking coordinate data and corresponding second timestamps; the human eye tracking coordinate data is the coordinate data of the human eye in the human eye model calculated according to the human eye tracking algorithm during the movement of the human eye model; the second timestamp is the time when the corresponding human eye tracking coordinate data and the corresponding interlaced image are displayed on the display device set relative to the human eye model, and the interlaced image is obtained and displayed on the display device according to the human eye tracking coordinate data and the multi-viewpoint image synthesis algorithm; wherein, the first trajectory curve is the relationship curve between the actual human eye coordinates and time; the second trajectory curve is the relationship curve between the human eye tracking coordinates and time.
[0120] The curve acquisition module 603 is configured to generate a first trajectory curve based on the first time series data and a second trajectory curve based on the second time series data.
[0121] The system latency acquisition module 604 is configured to acquire overall latency data based on the time difference between the first trajectory curve and the second trajectory curve.
[0122] like Figure 7 As shown, this disclosure provides a test system for the latency of a naked-eye 3D system. The latency test system 700 includes a driving device 701, a display device 702, a camera device 703, and a naked-eye 3D system latency test device 600.
[0123] The drive unit 701 is configured to move the human eye model.
[0124] The display device 702 is configured to display continuously updated eye-tracking coordinate data and interlaced images; the eye-tracking coordinate data is the coordinate data of the human eye in the eye model calculated by the eye-tracking algorithm during the movement of the human eye model driven by the driving device; the interlaced images are obtained based on the eye-tracking coordinate data and a multi-viewpoint image synthesis algorithm.
[0125] The camera device 703 is configured to capture images displayed on the display device.
[0126] like Figure 8 As shown, this disclosure provides an electronic device, which includes: at least one processor 801; at least one memory 802; and one or more I / O interfaces 803 connected between the processor 801 and the memory 802; wherein the memory 802 stores one or more computer programs that can be executed by the at least one processor 801, and the one or more computer programs are executed by the at least one processor 801 to enable the at least one processor 801 to perform the above-described method for testing the latency of a naked-eye 3D system.
[0127] This disclosure also provides a computer-readable storage medium storing a computer program thereon, wherein the computer program, when executed by a processor, implements the aforementioned method for testing the latency of a naked-eye 3D system. The computer-readable storage medium may be volatile or non-volatile.
[0128] It is understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of this disclosure, and this disclosure is not limited thereto. For those skilled in the art, various modifications and improvements can be made without departing from the spirit and substance of this disclosure, and these modifications and improvements are also considered to be within the scope of protection of this disclosure.
Claims
1. A method for testing the latency of a glasses-free 3D system, comprising: Acquire first time series data; wherein, the first time series data includes multiple sets of actual human eye coordinate data and corresponding first timestamps; the actual human eye coordinate data is the coordinate data of the human eye in the human eye model obtained during the movement of the human eye model; the first timestamp is the acquisition time of the corresponding actual human eye coordinate data; Acquire second time series data; wherein, the second time series includes multiple sets of human eye tracking coordinate data and corresponding second timestamps; the human eye tracking coordinate data is the coordinate data of the human eye in the human eye model calculated according to the human eye tracking algorithm during the movement of the human eye model; the second timestamp is the time when the corresponding human eye tracking coordinate data and the corresponding interlaced image are displayed on the display device set opposite to the human eye model, and the interlaced image is obtained according to the human eye tracking coordinate data and the multi-viewpoint image synthesis algorithm and displayed on the display device; A first trajectory curve is generated based on the first time series data, and a second trajectory curve is generated based on the second time series data; the first trajectory curve is the relationship curve between the actual coordinates of the human eye and time; the second trajectory curve is the relationship curve between the human eye tracking coordinates and time. The overall latency data is obtained based on the time difference between the first trajectory curve and the second trajectory curve.
2. The method for testing the latency of a naked-eye 3D system according to claim 1, wherein, The coordinate data in each of the multiple sets of actual human eye coordinate data is obtained based on the initial position and motion parameters of the human eye model.
3. The method for testing the latency of a naked-eye 3D system according to claim 2, wherein, At the initial position of the human eye model, the line connecting the two eyes in the human eye model is set parallel to the plane where the display device is located, and the line connecting the reference point of the human eye model and the reference point of the display device is perpendicular to the plane where the display device is located.
4. The method for testing the latency of a naked-eye 3D system according to claim 2, wherein, The motion parameters of the human eye model include motion direction, motion time, and motion speed; The direction of motion of the human eye model is parallel to the plane where the display device is located, and the speed of motion of the human eye model is uniform.
5. The method for testing the latency of a naked-eye 3D system according to claim 1, wherein, The human eye model is driven by a drive device.
6. The method for testing the latency of a naked-eye 3D system according to claim 1, wherein, The second time series data is obtained from multiple sets of captured images, which are obtained by the camera device taking multiple shots of the screen displayed on the display device.
7. The method for testing the latency of a naked-eye 3D system according to claim 6, wherein, The second time series data is obtained by performing artificial intelligence recognition on each of the multiple sets of captured images.
8. The method for testing the latency of a naked-eye 3D system according to claim 6 further includes: Obtain test latency data; The test latency data includes the time consumed by the human eye tracking algorithm, the time consumed by the multi-viewpoint image synthesis algorithm, the screen refresh time of the display device, and the shooting time of the camera device. The fixed latency data of the system is obtained based on the difference between the overall latency data and the test latency data.
9. The method for testing the latency of a naked-eye 3D system according to any one of claims 1 to 8, wherein, The step of generating the first trajectory curve based on the first time series data includes: The actual coordinates of the human eye are obtained along the first direction, which is used as the first coordinate data; the first direction is the direction of motion of the human eye model. A first location time scatter plot is obtained based on multiple first coordinate data and corresponding first timestamps; The first trajectory curve is obtained by fitting the position scatter points in the first position time scatter distribution map.
10. The method for testing the latency of a naked-eye 3D system according to any one of claims 1 to 8, wherein, The step of generating the second trajectory curve based on the second time series data includes: The coordinate data of the human eye tracking coordinates in the first direction is obtained and used as the second coordinate data; the first direction is the motion direction of the human eye model. A scatter plot of time distribution at the second location is obtained based on multiple second coordinate data and corresponding second timestamps; The second trajectory curve is obtained by fitting the position scatter points in the second position time scatter plot.
11. The method for testing the latency of a naked-eye 3D system according to any one of claims 1 to 8, wherein, The step of obtaining overall time delay data based on the time difference between the first trajectory curve and the second trajectory curve includes: The time difference between the first trajectory curve and the second trajectory curve under multiple preset coordinate data is obtained to obtain multiple time differences; The average value of the multiple time differences is obtained, and the overall delay data is the average value of the multiple time differences.
12. A testing device for latency of a naked-eye 3D system, comprising: The first data acquisition module is configured to: acquire first time series data; wherein, the first time series data includes multiple sets of actual human eye coordinate data and corresponding first timestamps; the actual human eye coordinate data is the coordinate data of the human eye in the human eye model acquired during the movement of the human eye model; the first timestamp is the acquisition time of the corresponding actual human eye coordinate data; The second data acquisition module is configured to: acquire second time series data; wherein, the second time series includes multiple sets of human eye tracking coordinate data and corresponding second timestamps; the human eye tracking coordinate data is the coordinate data of the human eye in the human eye model calculated according to the human eye tracking algorithm during the movement of the human eye model; the second timestamp is the time when the corresponding human eye tracking coordinate data and the corresponding interlaced image are displayed on a display device positioned opposite to the human eye model, the interlaced image is obtained according to the human eye tracking coordinate data and a multi-viewpoint image synthesis algorithm and displayed on the display device; wherein, the first trajectory curve is the relationship curve between the actual coordinates of the human eye and time; the second trajectory curve is the relationship curve between the human eye tracking coordinates and time; The curve acquisition module is configured to: generate a first trajectory curve based on the first time series data, and generate a second trajectory curve based on the second time series data; The system latency acquisition module is configured to acquire overall latency data based on the time difference between the first trajectory curve and the second trajectory curve.
13. A system for testing the latency of a glasses-free 3D system, comprising: The drive mechanism is configured to move the human eye model. The display device is configured to display continuously updated eye-tracking coordinate data and interlaced images; The eye-tracking coordinate data is the coordinate data of the human eye in the eye model calculated by the eye-tracking algorithm during the process of the driving device moving the human eye model. The interwoven image is obtained based on the human eye tracking coordinate data and the multi-viewpoint image synthesis algorithm; The camera device is configured to capture images displayed on the display device; and The device for testing the latency of a naked-eye 3D system as described in claim 12.
14. An electronic device comprising: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores one or more computer programs that can be executed by the at least one processor, the one or more computer programs being executed by the at least one processor to enable the at least one processor to perform a method for testing the latency of a naked-eye 3D system as described in any one of claims 1-11.
