Synchronization method of active shutter type 3D glasses and active shutter type 3D glasses

Abstract

The invention discloses a synchronization method of active shutter type 3D glasses and the active shutter type 3D glasses. The method comprises the following steps: acquiring an image sequence shot by an image sensor in the process of alternately opening a left lens and a right lens, wherein the image sequence comprises a left eye frame and a right eye frame; the homologous similarity and the heterogenous similarity are respectively determined, the homologous similarity and the heterogenous similarity are respectively cosine similarities between the current frame and a target homologous frame and between the current frame and a target heterogenous frame, and the target homologous frame is a historical image shot by an image sensor when the same lens corresponding to the current frame is in an open state; the target heterogenous frame is a historical image shot by the image sensor when different lenses corresponding to the current frame are in an open state; if the homologous similarity is smaller than the heterologous similarity, determining a phase compensation amount according to the homologous similarity and the heterologous similarity; and respectively compensating the switching phases of the left lens and the right lens according to the phase compensation amount. According to the method, accurate synchronization of the 3D display and the 3D eyes can be realized, so that the user experience is improved.

Description

Technical Field

[0001] This application relates to the field of active shutter 3D glasses technology, and more particularly to a synchronization method for active shutter 3D glasses and active shutter 3D glasses. Background Technology

[0002] Active shutter 3D glasses technology is one of the mainstream solutions for achieving high-definition, full-resolution 3D displays. When active shutter 3D glasses are working, the 3D display device alternately plays image sequences corresponding to the left and right eyes at a doubled frequency. Simultaneously, the active shutter 3D glasses worn by the user receive synchronization signals from the display device via wireless communication (such as infrared or Bluetooth), and based on these signals, precisely control the alternating opening and closing of the liquid crystal shutters on their left and right lenses. When the screen displays the left-eye image, the left lens of the glasses opens and the right lens closes, ensuring that only the left eye can see the image; and vice versa. Through this rapid alternation and synchronization, the left and right eyes receive images with parallax, which are ultimately fused in the visual center of the brain to form a stereoscopic image.

[0003] In related technologies, the wireless synchronization signal used in active shutter 3D glasses technology has random delays and is subject to environmental interference, which makes it easy for the instantaneous timing of the shutter switching of the glasses to be misaligned with the screen display frame, resulting in users experiencing severe image blurring and dizziness. Summary of the Invention

[0004] This application provides a synchronization method for active shutter 3D glasses and active shutter 3D glasses to solve the technical problem that the wireless synchronization signal used in active shutter 3D glasses technology has random delay and is subject to environmental interference, which makes the instantaneous timing of the shutter switching of the glasses easily misalign with the screen display frame, resulting in severe image blurring and dizziness for the user.

[0005] To address the aforementioned technical problems, in a first aspect, this application provides a synchronization method for active shutter 3D glasses, the 3D glasses including an image sensor, a left lens, and a right lens, the image sensor being used to capture images on a 3D display, the method comprising: The image sequence captured by the image sensor during the alternating opening of the left lens and the right lens is obtained. The image sequence includes a left eye frame and a right eye frame. The left eye frame is the image captured by the image sensor when the left lens is in the open state, and the right eye frame is the image captured by the image sensor when the right lens is in the open state. The same-source similarity and different-source similarity are determined separately. The same-source similarity is the cosine similarity between the current frame and the target same-source frame. The different-source similarity is the cosine similarity between the current frame and the target different-source frame. The target same-source frame is a historical image captured by the image sensor when the same lens corresponding to the current frame is in the open state. The target different-source frame is a historical image captured by the image sensor when a different lens corresponding to the current frame is in the open state. If the homologous similarity is less than the heterologous similarity, then the phase compensation amount is determined based on the homologous similarity and the heterologous similarity. The switching phases of the left and right lenses are compensated according to the phase compensation amount.

[0006] In the aforementioned synchronization method for active shutter 3D glasses, if the synchronization between the 3D display and the 3D glasses is normal, consecutive left-eye frames and consecutive right-eye frames should have spatiotemporal continuity, resulting in high similarity. However, adjacent frames from different sources (such as left-right) should have lower similarity due to parallax. If the synchronization between the 3D display and the 3D glasses is normal, the similarity between the same source and the different sources is greater than the similarity between different sources. When an abnormal relationship is detected where the similarity between the same source and the different sources is less, the system can autonomously and accurately diagnose the out-of-synchronization state and immediately trigger an internal compensation mechanism. This enables the 3D glasses to have visual self-checking and autonomous tuning capabilities, fundamentally solving the problems of viewing dizziness and blurring caused by signal interruption, delay, or clock drift. Users can obtain a continuous, stable, and comfortable 3D viewing experience without any operation. It can be seen that this synchronization method for active shutter 3D glasses completely eliminates the absolute dependence on manual user intervention and fragile wireless synchronization signals, achieving fully automatic and intelligent maintenance and recovery of the synchronization state.

[0007] In one embodiment, the 3D glasses further include a local clock, and the method further includes: Obtain the refresh period and current timestamp of the local clock; The target historical time point is determined based on the refresh cycle and the current timestamp. In the image sequence, the frame whose timestamp is closest to the target historical time point and is of the same category as the current frame is determined as the target homogeneous frame; The frame immediately preceding the current frame is identified as the target heterogeneous frame.

[0008] In one embodiment, the method further includes: Based on the image sequence, a global motion vector is determined from the target source frame to the current frame, and the global motion vector is used to characterize the motion of the image sensor; Affine transformation is performed on the target source frame based on the global motion vector, so that the target source frame after affine transformation is background-aligned with the current frame. The similarity is determined based on the target source frame after affine transformation and the current frame.

[0009] In one embodiment, before the steps of determining homologous similarity and heterologous similarity respectively, the method further includes: The current frame, the target same-source frame, and the target different-source frame are subjected to low-resolution processing to generate a first processed frame, a second processed frame, and a third processed frame, respectively. A first similarity and a second similarity are determined respectively, wherein the first similarity is the similarity between the first processing frame and the second processing frame, and the second similarity is the similarity between the first processing frame and the third processing frame; If the first similarity is less than the second similarity, then the step of determining the homologous similarity and the heterologous similarity is performed.

[0010] In one embodiment, before performing low-resolution processing on the current frame, the target same-origin frame, and the target different-origin frame, the method further includes: Determine the image feature information entropy of the current frame, the target homogeneous frame, and the target heterogeneous frame respectively; If the image feature information entropy of the current frame, the target same-source frame, and the target different-source frame are all greater than the preset entropy value, then the step of performing low-resolution processing on the current frame, the target same-source frame, and the target different-source frame is executed.

[0011] In one embodiment, determining the phase compensation amount based on the homo-source similarity and the hetero-source similarity includes: If both the homologous similarity and the heterologous similarity are greater than a preset similarity threshold, then a first proportional coefficient and a first integral coefficient are obtained, and the phase compensation amount is determined based on the first proportional coefficient, the first integral coefficient, and the similarity difference. If the homologous similarity is not greater than the preset similarity threshold, or the heterologous similarity is not greater than the preset similarity threshold, then the second proportional coefficient and the second integral coefficient are obtained, and the phase compensation amount is determined based on the second proportional coefficient, the second integral coefficient and the similarity difference. Wherein, the similarity difference is equal to the difference between the homologous similarity and the heterologous similarity.

[0012] In one implementation, the phase compensation amount is equal to the sum of the first compensation amount and the second compensation amount; The first compensation amount is equal to the product of the first proportional coefficient and the similarity difference, and the second compensation amount is equal to the product of the first integral coefficient and the integral of the difference; or, the first compensation amount is equal to the product of the second proportional coefficient and the similarity difference, and the second compensation amount is equal to the product of the second integral coefficient and the integral of the difference. The difference integral is calculated by using the finite-time sliding window integration method on the similarity difference.

[0013] In one embodiment, after compensating the switching phases of the left lens and the right lens according to the phase compensation amount, the method further includes: If the phase compensation amount is greater than the preset compensation amount threshold, then the similarity difference value within the sliding window is cleared to zero. If the phase compensation amount is not greater than the preset compensation amount threshold, then the similarity difference within the sliding window is attenuated based on the preset attenuation coefficient.

[0014] In one embodiment, the method further includes: Obtain multiple consecutive sample values ​​of the phase compensation amount within a preset statistical period; Determine the standard deviation of the plurality of consecutive sample values; The preset compensation threshold is determined based on the standard deviation, wherein the preset compensation threshold is equal to the product of the standard deviation and a preset coefficient.

[0015] Secondly, this application provides an active shutter 3D glasses, including a left lens, a right lens, an image sensor, a processor, and a memory. The memory is used to store a computer program, which, when executed by the processor, implements the above-described synchronization method for the active shutter 3D glasses. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the synchronization method of active shutter 3D glasses according to an embodiment of this application; Figure 2 This is a schematic diagram of the structure of an active shutter 3D glasses according to an embodiment of this application. Detailed Implementation

[0017] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0018] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating a synchronization method for active shutter 3D glasses provided in an embodiment of this application. The active shutter 3D glasses include an image sensor, a left lens, and a right lens. The image sensor is used to capture images on a 3D display. Figure 1 As shown, the synchronization method of the active shutter 3D glasses in this embodiment includes steps S101 to S104, which are detailed below: Step S101: Obtain the image sequence captured by the image sensor during the alternating opening of the left lens and the right lens. The image sequence includes a left eye frame and a right eye frame. The left eye frame is the image captured by the image sensor when the left lens is in the open state, and the right eye frame is the image captured by the image sensor when the right lens is in the open state.

[0019] In this step, the left and right lenses are alternately activated, allowing the user to experience stereoscopic images while wearing 3D glasses. During the alternating activation of the left and right lenses, the image sensor continuously captures images on the 3D display. Preferably, the image sensor captures one left-eye frame during the time the left lens is activated, and one right-eye frame during the time the right lens is activated.

[0020] Step S102: Determine the same-source similarity and different-source similarity respectively. The same-source similarity is the cosine similarity between the current frame and the target same-source frame. The different-source similarity is the cosine similarity between the current frame and the target different-source frame. The target same-source frame is a historical image captured by the image sensor when the same lens corresponding to the current frame is in the open state. The target different-source frame is a historical image captured by the image sensor when a different lens corresponding to the current frame is in the open state.

[0021] In this step, the target homologous frames and target heterologous frames are first identified, and then the homologous similarity and heterologous similarity are determined respectively. Further, when determining homologous and heterologous similarities, image preprocessing is performed on the current frame, the target homologous frames, and the target heterologous frames respectively, such as grayscale conversion and noise reduction. Then, high-texture regions are defined in the current frame, the target homologous frames, and the target heterologous frames respectively. A normalized cross-correlation algorithm or structural similarity index is used to determine the cosine similarity between the current frame and the target homologous frames as homologous similarity, and the cosine similarity between the current frame and the target heterologous frames as heterologous similarity. By performing grayscale conversion and noise reduction on the current frame, the target same-source frame, and the target different-source frame, the computational efficiency and robustness can be effectively improved. By defining high-texture regions in the current frame, the target same-source frame, and the target different-source frame respectively, and determining the same-source similarity and different-source similarity based on the defined high-texture regions, the interference of smooth regions (such as sky and walls) in the image on the similarity calculation is eliminated, ensuring that the calculation results can truly reflect the structural differences between images, rather than noise or lighting changes, and significantly improving the reliability of the criterion.

[0022] In a preferred embodiment, the 3D glasses also include a local clock. When determining the target heterogeneous frame, the refresh cycle and current timestamp of the local clock are first obtained, and then the target historical time point is determined according to the refresh cycle and current timestamp. Then, in the image sequence, the frame whose timestamp is closest to the target historical time point and is of the same category as the current frame is determined as the target homogeneous frame, and the frame immediately preceding the current frame is determined as the target heterogeneous frame.

[0023] In this embodiment, the target heterogeneous frame is the frame immediately preceding the current frame, meaning there is a one-refresh-cycle interval between the target heterogeneous frame and the current frame. The target homogeneous frame, however, is not the frame immediately preceding the target heterogeneous frame, but can be determined by tracing back K refresh cycles (K being an integer greater than 1) from the current frame. It should be noted that in dynamic scenes, the frame immediately preceding the target heterogeneous frame (approximately 33ms interval, 60Hz per eye) can vary significantly due to rapid object movement. By tracing back K refresh cycles from the current frame (e.g., K=3, approximately 100ms interval), the scene can revert to a more similar state (e.g., periodic motion or relative stillness), thus maintaining a high homogeneous similarity when the 3D glasses and 3D display are properly synchronized. This leverages the relationship between time interval and scene changes to greatly enhance the probability of the core criterion (homogeneous similarity not less than heterogeneous similarity) holding true in dynamic scenes. This embodiment uses the refresh cycle T of the 3D glasses' local clock for determining the target homogeneous frame, forming an internal loop to achieve a high degree of system intelligence and integration.

[0024] In a preferred embodiment, when determining the similarity of origin, firstly, a global motion vector from the target origin frame to the current frame is determined based on the image sequence. The global motion vector is used to characterize the motion of the image sensor. Then, an affine transformation is performed on the target origin frame based on the global motion vector, so that the target origin frame after the affine transformation is aligned with the background of the current frame. Finally, the similarity of origin is determined based on the target origin frame after the affine transformation and the current frame.

[0025] In this embodiment, before determining the similarity of origin, motion estimation and compensation can be performed by performing an affine transformation on the target origin frame based on the global motion vector. Before calculating the similarity of origin, the target origin frame is aligned to the viewpoint of the current frame, actively eliminating scene motion, the main interference factor that causes the decrease in similarity of origin. After performing an affine transformation on the target origin frame based on the global motion vector, the difference between the target origin frame and the current frame mainly comes from: uncompensated local motion (object's own motion) and fundamentally different image content caused by loss of synchronization (e.g., the left eye seeing the right eye image). This makes the decrease in similarity of origin more accurately attributed to loss of synchronization itself, rather than scene motion, significantly improving the specificity and accuracy of loss of synchronization detection. By applying motion compensation technology from computer vision to the field of 3D synchronous detection, the accuracy and reliability of 3D synchronous detection are effectively improved.

[0026] In a preferred embodiment, before determining the same-source similarity and different-source similarity respectively, the current frame, the target same-source frame, and the target different-source frame are first processed at low resolution to generate a first processed frame, a second processed frame, and a third processed frame respectively; then, a first similarity and a second similarity are determined respectively, where the first similarity is the similarity between the first processed frame and the second processed frame, and the second similarity is the similarity between the first processed frame and the third processed frame; if the first similarity is less than the second similarity, then the steps of determining the same-source similarity and different-source similarity are performed.

[0027] In this embodiment, before determining the similarity between sources and different sources, a first similarity and a second similarity are first determined. Only when the first similarity is less than the second similarity is the determination of the similarity between sources and different sources executed. This achieves a two-level calculation process of coarse detection and fine judgment, allowing the system to confirm the state with only lightweight calculations at low resolution in most normally synchronized frames (if the first similarity is not less than the second similarity, the determination of the similarity between sources and different sources is not executed), significantly reducing average power consumption and computational latency. Only when a suspected loss of synchronization is initially detected (first similarity less than second similarity) is the high-precision, high-consumption precision calculation module activated. This strategy of allocating computational resources on demand aligns with the stringent low-power requirements of wearable devices, enabling complex visual synchronization algorithms to run stably and sustainably on resource-constrained embedded platforms, demonstrating high engineering practical value.

[0028] In a preferred embodiment, before performing low-resolution processing on the current frame, the target same-source frame, and the target different-source frame, the image feature information entropy of the current frame, the target same-source frame, and the target different-source frame can be determined respectively; if the image feature information entropy of the current frame, the target same-source frame, and the target different-source frame are all greater than a preset entropy value, then the step of performing low-resolution processing on the current frame, the target same-source frame, and the target different-source frame is executed.

[0029] In this embodiment, image feature information entropy can reflect the image quality of the current frame, the target same-source frame, and the target different-source frame. If the image feature information entropy is greater than a preset entropy value, it can be determined that the image content of the current frame, the target same-source frame, or the target different-source frame is extremely monotonous (such as pure black, pure white, or monochrome screen), and low-resolution processing will be meaningless in this case. By calculating the image feature information entropy (which can be simplified to the entropy of the grayscale histogram or the sum of gradient magnitudes) to quantify the image information of the current frame, the target same-source frame, and the target different-source frame, such scenarios can be intelligently identified, making the determination of same-source similarity and different-source similarity more reliable.

[0030] Step S103: If the homologous similarity is less than the heterologous similarity, then determine the phase compensation amount based on the homologous similarity and the heterologous similarity.

[0031] In this step, if the similarity between the same source and the different source is less than the similarity between different source, it is determined that there is an abnormality in the synchronization process between the 3D glasses and the 3D display, and it is necessary to further determine the phase compensation amount based on the similarity between the same source and the different source.

[0032] In a preferred embodiment, when determining the phase compensation amount based on homologous similarity and heterologous similarity, if both homologous similarity and heterologous similarity are greater than a preset similarity threshold, a first proportional coefficient and a first integral coefficient are obtained, and the phase compensation amount is determined based on the first proportional coefficient, the first integral coefficient, and the similarity difference; if homologous similarity is not greater than the preset similarity threshold, or heterologous similarity is not greater than the preset similarity threshold, a second proportional coefficient and a second integral coefficient are obtained, and the phase compensation amount is determined based on the second proportional coefficient, the second integral coefficient, and the similarity difference; the similarity difference is equal to the difference between homologous similarity and heterologous similarity.

[0033] In this embodiment, a confidence-driven adaptive adjustment mechanism is introduced. Instead of using fixed adjustment parameters, the control parameters are dynamically selected based on the confidence level of the similarity itself (measured by the relationship between homologous similarity, heterologous similarity, and a preset similarity threshold). If both homologous and heterologous similarities are greater than the preset similarity threshold, it indicates that the image quality is good, the features are obvious, and the judgment is reliable. Therefore, more aggressive and sensitive adjustment parameters are used to achieve rapid convergence. If homologous similarity is not greater than the preset similarity threshold, or heterologous similarity is not greater than the preset similarity threshold, it indicates that the image is blurry and the content is monotonous. The judgment may be unreliable. Therefore, more conservative and robust parameters are used to avoid oscillations caused by misadjustment.

[0034] In a preferred embodiment, the phase compensation amount is equal to the sum of the first compensation amount and the second compensation amount; the first compensation amount is equal to the product of the first proportional coefficient and the similarity difference, and the second compensation amount is equal to the product of the first integral coefficient and the difference integral, or the first compensation amount is equal to the product of the second proportional coefficient and the similarity difference, and the second compensation amount is equal to the product of the second integral coefficient and the difference integral; the difference integral is calculated by using the finite-time sliding window integration method on the similarity difference.

[0035] In this embodiment, the difference integral can be determined using the finite-time sliding window integration method, that is, only the similarity differences generated within the most recent N calculation cycles are accumulated to obtain the difference integral, where N is a positive integer. If both the homologous similarity and heterologous similarity are greater than a preset similarity threshold, the first compensation amount is equal to the product of the first proportional coefficient and the similarity difference, and the second compensation amount is equal to the product of the first integral coefficient and the difference integral; if the homologous similarity is not greater than the preset similarity threshold, or the heterologous similarity is not greater than the preset similarity threshold, the first compensation amount is equal to the product of the second proportional coefficient and the similarity difference, and the second compensation amount is equal to the product of the second integral coefficient and the difference integral. By dynamically selecting control parameters based on the confidence level of the similarity itself (measured by the relationship between homologous similarity, heterologous similarity, and the preset similarity threshold), the system's adaptability and overall stability under different screen content and viewing environments are significantly improved.

[0036] In a preferred embodiment, after compensating the switching phase of the left and right lenses respectively according to the phase compensation amount, the relationship between the phase compensation amount and a preset compensation amount threshold can be determined. If it is determined that the phase compensation amount is greater than the preset compensation amount threshold, the similarity difference within the sliding window is cleared to zero; if the phase compensation amount is not greater than the preset compensation amount threshold, the similarity difference within the sliding window is attenuated based on a preset attenuation coefficient.

[0037] In this embodiment, if the current phase compensation amount is greater than a preset compensation threshold, it indicates that the current correction magnitude is significant, and the system determines that the historical error has been effectively addressed. Therefore, all historical error values ​​within the sliding window are cleared to zero. Conversely, if the current phase compensation amount is not greater than the preset compensation threshold, it indicates that the current adjustment is only minor. The system then multiplies each historical error value within the sliding window by a decay coefficient less than 1 (e.g., 0.5) to weaken it rather than completely eliminate it. This avoids unnecessary superposition compensation due to the continued existence of historical errors after the error has been largely compensated, which could lead to overshoot or system instability. It allows the integrator to more quickly reflect the new synchronization state after the compensation action, focusing on newly emerging deviations rather than being dragged down by old data. The intelligent selection of clearing or decaying based on the magnitude of the phase compensation amount reflects the system's evaluation of its own correction effect and differentiated learning strategy, enhancing its intelligence.

[0038] In a preferred embodiment, when determining the preset compensation threshold, firstly, multiple consecutive sample values ​​of phase compensation within a preset statistical period are obtained, then the standard deviation of the multiple consecutive sample values ​​is determined, and finally the preset compensation threshold is determined based on the standard deviation. The preset compensation threshold is equal to the product of the standard deviation and the preset coefficient.

[0039] In this embodiment, the preset compensation threshold is not a fixed value, but is adaptively adjusted according to the system's operating state. For example, when the system continuously satisfies the same-source similarity > different-source similarity for N frames (e.g., N=50), and the calculated phase compensation amount is consistently less than a certain minimum value, the system is determined to have entered a stable synchronization state. In the stable synchronization state, the phase compensation amount is collected over multiple (e.g., 100) calculation cycles. The phase compensation amount is mainly caused by computational noise and random interference, and its mean should be close to zero. The standard deviation of these phase compensation amount sample values ​​is calculated. The preset coefficient can be set to, for example, 2.5. The selection of the preset coefficient ensures that random fluctuations and true out-of-synchronization signals are distinguished with a high probability (approximately 99%). The above adaptive process can be re-executed periodically (e.g., every 5 minutes) or when a significant change in the environment is detected (e.g., a sudden change in ambient brightness) to ensure that the threshold always adapts to the current viewing conditions. In another embodiment, the preset compensation threshold can be set according to the minimum time resolution of the timer inside the 3D glasses. For example, if the minimum counting unit of the timer controlling lens switching is 1 microsecond, the preset compensation threshold can be set to the counting value corresponding to 2-5 microseconds. It can be seen that the preset compensation threshold is not fixed, but dynamically adjusted according to the actual noise level of the system. This allows the system to automatically optimize its behavior under different environments (such as different movie content and different ambient light), providing a quantitative and statistical basis for the choice between zeroing and attenuation, avoiding subjective settings, and effectively preventing inappropriate zeroing operations triggered by accidental large values ​​of phase compensation caused by random noise, thereby improving the stability of the system.

[0040] Step S104: Compensate the switching phase of the left lens and the right lens respectively according to the phase compensation amount.

[0041] In this step, by compensating the switching phase of the left and right lenses according to the phase compensation amount, the switching timing of the left and right lenses can be precisely adjusted, thereby switching the synchronization process between the 3D glasses and the 3D display from an abnormal state to a normal state. This fundamentally solves the problem of viewing dizziness and blurring caused by signal interruption, delay or clock drift. Users can obtain a continuous, stable and comfortable 3D viewing experience without any operation.

[0042] In the aforementioned synchronization method for active shutter 3D glasses, if the synchronization between the 3D display and the 3D glasses is normal, consecutive left-eye frames and consecutive right-eye frames should have spatiotemporal continuity, resulting in high similarity. However, adjacent frames from different sources (such as left-right) should have lower similarity due to parallax. If the synchronization between the 3D display and the 3D glasses is normal, the similarity between the same source and the different sources is greater than the similarity between different sources. When an abnormal relationship is detected where the similarity between the same source and the different sources is less, the system can autonomously and accurately diagnose the out-of-synchronization state and immediately trigger an internal compensation mechanism. This enables the 3D glasses to have visual self-checking and autonomous tuning capabilities, fundamentally solving the problems of viewing dizziness and blurring caused by signal interruption, delay, or clock drift. Users can obtain a continuous, stable, and comfortable 3D viewing experience without any operation. It can be seen that this synchronization method for active shutter 3D glasses completely eliminates the absolute dependence on manual user intervention and fragile wireless synchronization signals, achieving fully automatic and intelligent maintenance and recovery of the synchronization state.

[0043] Figure 2 This is a schematic diagram of the structure of an active shutter 3D glasses provided in an embodiment of this application. Figure 2 As shown, the active shutter 3D glasses 20 of this embodiment includes: a left lens (not shown), a right lens (not shown), an image sensor (not shown), and at least one processor 21. Figure 2 (Only one is shown in the diagram), memory 22, and computer program 23 stored in memory 22 and executable on the at least one processor 21, wherein the processor 21 executes the computer program 23 to implement the steps in any of the above method embodiments.

[0044] The active shutter 3D glasses may include, but are not limited to, a processor 21 and a memory 22. Those skilled in the art will understand that... Figure 2 This is merely an example of an active shutter 3D glasses 20 and does not constitute a limitation on the active shutter 3D glasses 20. It may include more or fewer components than shown in the figure, or combine certain components, or different components, such as input / output devices, network access devices, etc.

[0045] The processor 21 may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.

[0046] In some embodiments, the memory 22 may be an internal storage unit of the active shutter 3D glasses 20. The memory 21 is used to store the operating system, applications, bootloader, data, and other programs, such as the program code of the computer program. The memory 31 can also be used to temporarily store data that has been output or will be output.

[0047] In addition, embodiments of this application also provide a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps in any of the above method embodiments.

[0048] This application provides a computer program product that, when run on a computer device, enables the computer device to execute the steps described in the various method embodiments above.

[0049] In the several embodiments provided in this application, it will be understood that each block in the flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the figures. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved.

[0050] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0051] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of this application. It should be understood that the above descriptions are merely specific embodiments of this application and are not intended to limit the scope of protection of this application. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application for those skilled in the art.

Claims

1. A synchronization method for active shutter 3D glasses, characterized in that, The 3D glasses include an image sensor, a left lens, and a right lens. The image sensor is used to capture images on the 3D display. The method includes: The image sequence captured by the image sensor during the alternating opening of the left lens and the right lens is obtained. The image sequence includes a left eye frame and a right eye frame. The left eye frame is the image captured by the image sensor when the left lens is in the open state, and the right eye frame is the image captured by the image sensor when the right lens is in the open state. The same-source similarity and different-source similarity are determined separately. The same-source similarity is the cosine similarity between the current frame and the target same-source frame. The different-source similarity is the cosine similarity between the current frame and the target different-source frame. The target same-source frame is a historical image captured by the image sensor when the same lens corresponding to the current frame is in the open state. The target different-source frame is a historical image captured by the image sensor when a different lens corresponding to the current frame is in the open state. If the homologous similarity is less than the heterologous similarity, then the phase compensation amount is determined based on the homologous similarity and the heterologous similarity. The switching phases of the left and right lenses are compensated according to the phase compensation amount.

2. The synchronization method for active shutter 3D glasses as described in claim 1, characterized in that, The 3D glasses also include a local clock, and the method further includes: Obtain the refresh period and current timestamp of the local clock; The target historical time point is determined based on the refresh cycle and the current timestamp. In the image sequence, the frame whose timestamp is closest to the target historical time point and is of the same category as the current frame is determined as the target homogeneous frame; The frame immediately preceding the current frame is identified as the target heterogeneous frame.

3. The synchronization method for active shutter 3D glasses as described in claim 2, characterized in that, The method further includes: Based on the image sequence, a global motion vector is determined from the target source frame to the current frame, and the global motion vector is used to characterize the motion of the image sensor; Affine transformation is performed on the target source frame based on the global motion vector, so that the target source frame after affine transformation is background-aligned with the current frame. The similarity is determined based on the target source frame after affine transformation and the current frame.

4. The synchronization method for active shutter 3D glasses as described in claim 1, characterized in that, Before the steps of determining homology similarity and heterology similarity respectively, the method further includes: The current frame, the target same-source frame, and the target different-source frame are subjected to low-resolution processing to generate a first processed frame, a second processed frame, and a third processed frame, respectively. A first similarity and a second similarity are determined respectively, wherein the first similarity is the similarity between the first processing frame and the second processing frame, and the second similarity is the similarity between the first processing frame and the third processing frame; If the first similarity is less than the second similarity, then the step of determining the homologous similarity and the heterologous similarity is performed.

5. The synchronization method for active shutter 3D glasses as described in claim 4, characterized in that, Before performing low-resolution processing on the current frame, the target same-origin frame, and the target different-origin frame, the method further includes: Determine the image feature information entropy of the current frame, the target homogeneous frame, and the target heterogeneous frame respectively; If the image feature information entropy of the current frame, the target same-source frame, and the target different-source frame are all greater than the preset entropy value, then the step of performing low-resolution processing on the current frame, the target same-source frame, and the target different-source frame is executed.

6. The synchronization method for active shutter 3D glasses as described in claim 1, characterized in that, The step of determining the phase compensation amount based on the homogeneity similarity and the heterogeneity similarity includes: If both the homologous similarity and the heterologous similarity are greater than a preset similarity threshold, then a first proportional coefficient and a first integral coefficient are obtained, and the phase compensation amount is determined based on the first proportional coefficient, the first integral coefficient, and the similarity difference. If the homologous similarity is not greater than the preset similarity threshold, or the heterologous similarity is not greater than the preset similarity threshold, then the second proportional coefficient and the second integral coefficient are obtained, and the phase compensation amount is determined based on the second proportional coefficient, the second integral coefficient and the similarity difference. Wherein, the similarity difference is equal to the difference between the homologous similarity and the heterologous similarity.

7. The synchronization method for active shutter 3D glasses as described in claim 6, characterized in that, The phase compensation amount is equal to the sum of the first compensation amount and the second compensation amount; The first compensation amount is equal to the product of the first proportional coefficient and the similarity difference, and the second compensation amount is equal to the product of the first integral coefficient and the integral of the difference; or, the first compensation amount is equal to the product of the second proportional coefficient and the similarity difference, and the second compensation amount is equal to the product of the second integral coefficient and the integral of the difference. The difference integral is calculated by using the finite-time sliding window integration method on the similarity difference.

8. The synchronization method for active shutter 3D glasses as described in claim 7, characterized in that, After compensating the switching phases of the left and right lenses according to the phase compensation amount, the method further includes: If the phase compensation amount is greater than the preset compensation amount threshold, then the similarity difference value within the sliding window is cleared to zero. If the phase compensation amount is not greater than the preset compensation amount threshold, then the similarity difference within the sliding window is attenuated based on the preset attenuation coefficient.

9. The synchronization method for active shutter 3D glasses as described in claim 8, characterized in that, The method further includes: Obtain multiple consecutive sample values ​​of the phase compensation amount within a preset statistical period; Determine the standard deviation of the plurality of consecutive sample values; The preset compensation threshold is determined based on the standard deviation, wherein the preset compensation threshold is equal to the product of the standard deviation and a preset coefficient.

10. An active shutter 3D glasses, characterized in that, It includes a left lens, a right lens, an image sensor, a processor, and a memory, wherein the memory is used to store a computer program, which, when executed by the processor, implements the synchronization method of the active shutter 3D glasses as described in any one of claims 1 to 9.

Images