Method and apparatus for efficient motion estimation

Abstract

The present disclosure relates to methods and apparatus for graphics processing. An apparatus may determine a motion estimation priority for one or more regions in a frame. In some aspects, each of the one or more regions may include a pixel group. Additionally, the apparatus may adjust the quality of a motion estimation process for each pixel group based on the motion estimation priority of the region including the pixel group. The apparatus may also perform a motion estimation process for each pixel group in at least one pixel group based on the adjusted quality of the motion estimation process to generate a motion vector for the at least one pixel group.

Description

[0001] Cross-references

[0002] This application claims priority to U.S. Provisional Application No. 62 / 888,974, filed on August 19, 2019, entitled “METHODS AND APPARATUS FOR EFFICIENT MOTION ESTIMATION,” and U.S. Non-Provisional Patent Application No. 16 / 987,983, filed on August 7, 2020, entitled “METHODS AND APPARATUS FOR EFFICIENT MOTION ESTIMATION,” which have been assigned to the assignee of this application, and the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present disclosure relates generally to processing systems and, more particularly, to one or more techniques for graphics processing. Background Art

[0004] Computing devices often utilize a graphics processing unit (GPU) to render the rendering of graphical data for display. Such computing devices may include, for example, computer workstations, mobile phones such as so-called smartphones, embedded systems, personal computers, tablet computers, and video game consoles. The GPU executes a graphics processing pipeline comprising one or more processing stages that operate together to execute graphics processing commands and output frames. A central processing unit (CPU) can control the operation of the GPU by issuing one or more graphics processing commands to the GPU. Modern CPUs are typically capable of executing multiple applications simultaneously, each of which may need to utilize the GPU during execution. Devices that provide content for visual display on a display typically include a GPU.

[0005] Typically, a device's GPU is configured to perform processes in a graphics processing pipeline. However, with the advent of wireless communications and smaller, handheld devices, an increased need for improved graphics processing has developed. Summary of the Invention

[0006] The following presents a simplified summary of one or more aspects in order to provide a basic understanding of such aspects. This summary is not an extensive overview of all contemplated aspects and is not intended to identify key elements of all aspects, nor is it intended to describe the scope of any or all aspects. This summary presents some concepts of one or more aspects in a simplified form as a prelude to the more detailed description that will be presented later.

[0007] In one aspect of the present disclosure, a method, a computer-readable medium, and an apparatus are provided. The apparatus may be a graphics processing unit (GPU). In some aspects, the apparatus may be configured to determine a motion estimation priority for one or more regions in a frame, wherein each of the one or more regions includes at least one pixel group. The apparatus may be configured to adjust the quality of a motion estimation process for each pixel group based on the motion estimation priority of the region including the pixel group. The apparatus may be configured to perform a motion estimation process for each pixel group in at least one pixel group based on the adjusted quality of the motion estimation process to generate a motion vector for the at least one pixel group.

[0008] In some implementations, the one or more regions include a center region and an edge region.

[0009] In some implementations, the one or more regions include at least one intermediate region between the central region and the edge region.

[0010] In some implementations, the apparatus is configured to track the position of the eye. One or more regions in the frame are defined relative to the position of the eye.

[0011] In some implementations, the apparatus is configured to determine to set a constant value for at least one pixel group in a region having a lowest motion estimation priority.

[0012] In some implementations, the apparatus is configured to set a size of at least one pixel group based on a motion estimation priority.

[0013] In some implementations, the apparatus is configured to adjust the size of a search window to be covered during the motion estimation process based on the motion estimation priority.

[0014] In some implementations, the apparatus is configured to determine the inclusion or omission of a multi-path algorithm with motion vector refinement in a motion estimation process based on a motion estimation priority.

[0015] In some implementations, the apparatus is configured to determine a size of a block for a sum of absolute error (SAD) operation or a sum of squared error (SSD) operation based on a motion estimation priority.

[0016] In some implementations, the apparatus is configured to adjust the amount of filtering performed during motion search filtering or on the final motion vector based on the motion estimation priority.

[0017] In some implementations, the apparatus is configured to extrapolate a new frame from a frame based on a motion vector for each pixel group.

[0018] The details of one or more examples of the disclosure are set forth in the accompanying drawings and the description below. Other features, objects, and advantages of the disclosure will be apparent from the description and drawings, and from the claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 is a block diagram illustrating an example content generation system in accordance with one or more techniques of this disclosure.

[0020] Figure 2 An example GPU is shown in accordance with one or more techniques of this disclosure.

[0021] Figure 3 Example regions of a frame are shown in accordance with one or more techniques of this disclosure.

[0022] Figure 4A and Figure 4B Motion vectors in an example frame are shown in accordance with one or more techniques of this disclosure.

[0023] Figure 5A is a photograph of an example of an image generated based on motion vectors with uniform motion estimation quality.

[0024] Figure 5B is a photograph of an example of an image generated based on motion vectors with adjusted motion estimation quality for different regions in accordance with one or more techniques of this disclosure.

[0025] Figure 5C An example heat map is shown, which shows the Figure 5A According to one or more technologies of the present disclosure Figure 5B The difference between.

[0026] Figure 6 An example flow diagram illustrating an example method in accordance with one or more techniques of this disclosure. DETAILED DESCRIPTION

[0027] When deploying frame extrapolation solutions in virtual reality (VR), augmented reality (AR), extended reality (XR), or gaming applications, performing high-quality motion estimation can be a significant cost. In these applications, areas that are farther from the focal point of an image can be considered less important than areas of the image that are closer to the focal point of the image. In some aspects, the quality of the motion estimation can be scaled based on the distance from the focal point of the image. In order to reduce the potential cost of motion estimation, aspects of the present disclosure can perform motion estimation based on the distance from the focal point of the image. For example, aspects of the present disclosure can determine a motion estimation priority for one or more regions in a frame, where each region in the one or more regions can include a corresponding motion vector. Aspects of the present disclosure can also adjust the quality of each corresponding motion vector based on the motion estimation priority. Aspects of the present disclosure can also perform motion estimation for each region in the one or more regions based on the quality of the corresponding motion vector.

[0028] The various aspects of the system, device, computer program product and method are described more fully below with reference to the accompanying drawings. However, the present disclosure can be embodied in many different forms and should not be interpreted as being limited to any specific structure and function displayed throughout the present disclosure. More precisely, these aspects are provided so that the present disclosure will be detailed and complete, and will fully convey the scope of the present disclosure to those skilled in the art. Based on the teachings herein, it should be understood by those skilled in the art that the scope of the present disclosure is intended to cover any aspect of the system, device, computer program product and method disclosed herein, whether implemented independently of other aspects of the present disclosure or combined with other aspects of the present disclosure. For example, any number of aspects set forth herein can be used to implement a device or practice method. In addition, the scope of the present disclosure is intended to cover such a device or method, which is practiced using other structures, functions, or structures and functions other than or different from the various aspects of the present disclosure set forth herein. Any aspect disclosed herein can be embodied by one or more elements in the claims.

[0029] Although various aspects are described herein, many variations and permutations of these aspects also fall within the scope of the present disclosure. Although some potential benefits and advantages of various aspects of the present disclosure are mentioned, the scope of the present disclosure is not intended to be limited to a particular benefit, use, or object. More specifically, various aspects of the present disclosure are intended to be broadly applicable to different wireless technologies, system configurations, networks, and transmission protocols, some of which are illustrated by way of example in the accompanying drawings and in the following description. The detailed description and drawings are merely illustrative of the present disclosure and are not limiting, and the scope of the present disclosure is defined by the appended claims and their equivalents.

[0030] Several aspects are illustrated with reference to various devices and methods. These devices and methods are described in the following detailed description and illustrated in the accompanying drawings by various blocks, components, circuits, processes, algorithms, etc. (collectively referred to as "elements"). These elements can be implemented using electronic hardware, computer software, or any combination thereof. Whether such elements are implemented as hardware or software depends on the specific application and the design constraints imposed on the overall system.

[0031] By way of example, an element or any part of an element or any combination of elements can be implemented as a "processing system" including one or more processors (the processors can also be referred to as processing units). Examples of processors include microprocessors, microcontrollers, graphics processing units (GPUs), general-purpose GPUs (GPGPUs), central processing units (CPUs), application processors, digital signal processors (DSPs), reduced instruction set computing (RISC) processors, systems on chip (SOCs), baseband processors, application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), programmable logic devices (PLDs), state machines, gating logic, discrete hardware circuits, and other appropriate hardware configured to perform the various functions described throughout this disclosure. One or more processors in a processing system can execute software. Whether referred to as software, firmware, middleware, microcode, hardware description language, or other terms, software can be broadly interpreted as meaning instructions, instruction sets, codes, code segments, program codes, programs, subroutines, software components, applications, software applications, software packages, routines, subroutines, objects, executable files, execution threads, processes, functions, etc. The term application can refer to software. As described herein, one or more technologies may refer to an application, i.e., software, configured to perform one or more functions. In such examples, the application may be stored on a memory, such as an on-chip memory of a processor, a system memory, or any other memory. The hardware described herein, such as a processor, may be configured to execute the application. For example, an application may be described as including code that, when executed by hardware, causes the hardware to perform one or more technologies described herein. As an example, the hardware may access code from the memory and execute the code accessed from the memory to perform one or more technologies described herein. In some examples, components are identified in this disclosure. In such examples, a component may be hardware, software, or a combination thereof. A component may be a separate component or a subcomponent of a single component.

[0032] Accordingly, in one or more examples described herein, the functions described can be implemented in hardware, software, or a combination thereof. If implemented in software, the functions can be stored on a computer-readable medium or encoded as one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media. Storage media can be any available medium that can be accessed by a computer. By way of example and not limitation, such computer-readable media can include random access memory (RAM), read-only memory (ROM), electrically erasable programmable ROM (EEPROM), optical disk storage, magnetic disk storage, other magnetic storage devices, a combination of the above-mentioned types of computer-readable media, or any other medium that can be used to store computer-executable code in the form of instructions or data structures that can be accessed by a computer.

[0033] In general, this disclosure describes techniques for having a graphics processing pipeline in a single device or multiple devices, improving the rendering of graphics content, and / or reducing the load on a processing unit, i.e., any processing unit, such as a GPU, configured to perform one or more of the techniques described herein. For example, this disclosure describes techniques for graphics processing in any device that utilizes graphics processing. Other example benefits are described throughout this disclosure.

[0034] As used herein, instances of the term "content" may refer to "graphics content," "images," and vice versa. This is true regardless of whether these terms are used as adjectives, nouns, or other parts of speech. In some examples, as used herein, the term "graphics content" may refer to content generated by one or more processes of a graphics processing pipeline. In some examples, as used herein, the term "graphics content" may refer to content generated by a processing unit configured to perform graphics processing. In some examples, as used herein, the term "graphics content" may refer to content generated by a graphics processing unit.

[0035] As used herein, an instance of the term "content" may refer to graphics content or display content. In some examples, as used herein, the term "graphics content" may refer to content generated by a processing unit configured to perform graphics processing. For example, the term "graphics content" may refer to content generated by one or more processes of a graphics processing pipeline. In some examples, as used herein, the term "graphics content" may refer to content generated by a graphics processing unit. In some examples, as used herein, the term "display content" may refer to content generated by a processing unit configured to perform display processing. In some examples, as used herein, the term "display content" may refer to content generated by a display processing unit. Graphics content may be processed to become display content. For example, a graphics processing unit may output graphics content, such as frames, to a buffer (the buffer may be referred to as a frame buffer). A display processing unit may read graphics content, such as one or more frames, from the buffer and perform one or more display processing techniques on it to generate display content. For example, a display processing unit may be configured to perform synthesis on one or more rendered layers to generate frames. As another example, the display processing unit can be configured to composite, blend, or otherwise combine two or more layers into a single frame. The display processing unit can be configured to perform scaling on the frame, for example, scaling up or scaling down. In some examples, a frame can refer to a layer. In other examples, a frame can refer to two or more layers that have been blended together to form a frame, i.e., a frame includes two or more layers, and a frame including two or more layers can then be blended.

[0036] Figure 1 is a block diagram illustrating an example content generation system 100 configured to implement one or more techniques of this disclosure. Content generation system 100 includes a device 104. Device 104 may include one or more components or circuits for performing the various functions described herein. In some examples, one or more components of device 104 may be components of a system-on-chip (SoC). Device 104 may include one or more components configured to perform one or more techniques of this disclosure. In the example shown, device 104 may include a processing unit 120 and system memory 124. In some aspects, device 104 may include several optional components, such as a communication interface 126, a transceiver 132, a receiver 128, a transmitter 130, a display processor 127, and one or more displays 131. References to displays 131 may refer to one or more displays 131. For example, display 131 may include a single display or multiple displays. Display 131 may include a first display and a second display. The first display may be a left-eye display, and the second display may be a right-eye display. In some examples, the first and second displays may receive different frames for presentation thereon. In other examples, the first display and the second display may receive the same frame for presentation thereon. In further examples, the results of the graphics processing may not be displayed on the device, for example, the first display and the second display may not receive any frames for presentation thereon. Instead, the frames or graphics processing results may be transferred to the other device. In some aspects, this may be referred to as split-rendering.

[0037] Processing unit 120 may include internal memory 121. Processing unit 120 may be configured to perform graphics processing, such as in graphics processing pipeline 107. In some examples, device 104 may include a display processor, such as display processor 127, to perform one or more display processing techniques on one or more frames generated by processing unit 120 prior to presentation on one or more displays 131. Display processor 127 may be configured to perform display processing. For example, display processor 127 may be configured to perform one or more display processing techniques on one or more frames generated by processing unit 120. One or more displays 131 may be configured to display or otherwise present the frames processed by display processor 127. In some examples, one or more displays 131 may include one or more of a liquid crystal display (LCD), a plasma display, an organic light emitting diode (OLED) display, a projection display device, an augmented reality display device, a virtual reality display device, a head-mounted display, or any other type of display device.

[0038] Memory external to processing unit 120, such as system memory 124, may be accessible to processing unit 120. For example, processing unit 120 may be configured to read from and / or write to external memory, such as system memory 124. Processing unit 120 may be communicatively coupled to system memory 124 via a bus. In some examples, processing units 120 may be communicatively coupled to each other via a bus or a different connection.

[0039] Internal memory 121 or system memory 124 may include one or more volatile or non-volatile memories or storage devices. In some examples, internal memory 121 or system memory 124 may include RAM, SRAM, DRAM, erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, magnetic data media, optical storage media, or any other type of memory.

[0040] According to some examples, internal memory 121 or system memory 124 may be a non-transitory storage medium. The term "non-transitory" may indicate that the storage medium is not embodied in a carrier wave or propagated signal. However, the term "non-transitory" should not be interpreted as meaning that internal memory 121 or system memory 124 is non-removable or that its contents are static. As one example, system memory 124 may be removable from device 104 or moved to another device. As another example, system memory 124 may not be removable from device 104.

[0041] The processing unit 120 may be a central processing unit (CPU), a graphics processing unit (GPU), a general-purpose GPU (GPGPU), or any other processing unit that can be configured to perform graphics processing. In some examples, the processing unit 120 may be integrated into the motherboard of the device 104. In some examples, the processing unit 120 may exist on a graphics card installed in a port in the motherboard of the device 104, or may be otherwise incorporated into a peripheral device configured to interoperate with the device 104. The processing unit 120 may include one or more processors, such as one or more microprocessors, GPUs, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), arithmetic logic units (ALUs), digital signal processors (DSPs), discrete logic, software, hardware, firmware, other equivalent integrated or discrete logic circuits, or any combination thereof. If the technology is partially implemented in software, the processing unit 120 may store instructions for the software in an appropriate, non-transitory computer-readable medium, such as internal memory 121, and may use one or more processors to execute the instructions in hardware to perform the technology of the present disclosure. Any of the foregoing, including hardware, software, a combination of hardware and software, etc., may be considered to be one or more processors.

[0042] In some aspects, the content generation system 100 may include an optional communication interface 126. The communication interface 126 may include a receiver 128 and a transmitter 130. The receiver 128 may be configured to perform any of the receiving functions described herein with respect to the device 104. Additionally, the receiver 128 may be configured to receive information from another device, such as eye or head position information, rendering commands, or position information. The transmitter 130 may be configured to perform any of the transmitting functions described herein with respect to the device 104. For example, the transmitter 130 may be configured to transmit information to another device, which may include a request for content. The receiver 128 and the transmitter 130 may be combined into a transceiver 132. In such an example, the transceiver 132 may be configured to perform any of the receiving functions and / or transmitting functions described herein with respect to the device 104.

[0043] Reference again Figure 1In some aspects, the graphics processing pipeline 107 may include a determination component 198 configured to determine a motion estimation priority for one or more regions in a frame. In some aspects, each of the one or more regions may include at least one pixel group. Additionally, the determination component 198 may be configured to adjust a quality of a motion estimation process for each pixel group based on the motion estimation priority of the region including the pixel group. The determination component 198 may also be configured to perform a motion estimation process for each pixel group in the at least one pixel group based on the adjusted quality of the motion estimation process to generate a motion vector for the at least one pixel group. In some implementations, the determination component 198 may also be configured to extrapolate a new frame from the frame based on the motion vector for each pixel group.

[0044] As described herein, a device, such as device 104, can refer to any device, apparatus, or system configured to perform one or more of the techniques described herein. For example, a device can be a server, a base station, a user equipment, a client device, a station, an access point, a computer (e.g., a personal computer, a desktop computer, a laptop computer, a tablet computer, a computer workstation, a mainframe computer), an end product, an apparatus, a phone, a smartphone, a server, a video game platform or console, a handheld device (e.g., a portable video game device or a personal digital assistant (PDA)), a wearable computing device (e.g., a smartwatch, an augmented reality device, or a virtual reality device), a non-wearable device, a display or display device, a television, a television set-top box, an intermediate network device, a digital media player, a video streaming device, a content streaming device, an in-vehicle computer, any mobile device, any device configured to generate graphical content, or any device configured to perform one or more of the techniques described herein.

[0045] Figure 2 An example GPU 200 is shown in accordance with one or more techniques of this disclosure. Figure 2 As shown in FIG, the GPU 200 includes a command processor (CP) 210, a draw call group 212, a vertex fetcher (VFD) 220, a vertex shader (VS) 222, a vertex cache (VPC) 224, a triangle setup engine (TSE) 226, a rasterizer (RAS) 228, a Z-pass engine (VPC) 230, a pixel interpolator (PI) 232, a fragment shader (FS) 234, a rendering backend (RB) 236, an L2 cache (UCHE) 238, and a system memory 240. Although Figure 2GPU 200 is shown to include processing units 220-238, but GPU 200 may include several additional processing units. Furthermore, processing units 220-238 are merely examples, and any combination or order of processing units may be used by a GPU according to the present disclosure. GPU 200 also includes a command buffer 250, a context register group 260, and a context state 261.

[0046] The GPU can process multiple types of data on the GPU pipeline. For example, in some aspects, the GPU can process two types of data or data packets, such as context register packets and draw call data. Figure 2 As shown in FIG, the GPU may utilize a CP, such as CP 210, or a hardware accelerator to parse a command buffer into context register packets (e.g., context register packet 260) and / or draw call data packets (e.g., draw call packet 212). CP 210 may then send context register packet 260 or draw call packet 212 to a processing unit or block in the GPU via separate channels. Furthermore, command buffer 250 may alternate between different states of context registers and draw calls. For example, the command buffer may be structured as follows: context registers for context N, draw calls for context N, context registers for context N+1, and draw calls for context N+1.

[0047] Motion estimation is the process of analyzing multiple two-dimensional (2D) images and generating motion vectors that describe the movement of an area from one image to another. For example, motion estimation is a method for estimating the motion experienced during consecutive frames, such as during virtual reality (VR), augmented reality (AR), extended reality (XR), and / or gaming applications. Motion estimation can generate motion vectors that can describe how an object or group of pixels moves within certain parts of an image.

[0048] Motion vectors can have a variety of uses, including video compression, post-processing effects such as motion blur, and frame extrapolation or interpolation. To reduce the rendering workload on the GPU, a VR or AR system can utilize motion estimation to extrapolate frames from previously rendered content. By doing so, this can allow the GPU to render frames at a reduced rate, where the extrapolated frames are displayed to the user instead of the rendered content. Motion estimation can be useful because there is a strong drive to reduce the rendering workload, for example, at the GPU, in VR or AR systems.

[0049] Frame extrapolation is a method of increasing the frame rate of VR, AR, XR, and / or gaming applications. For example, frame extrapolation can estimate the motion between two previously rendered frames and extrapolate a new frame based on the detected motion. In some aspects, when extrapolating frames, instead of rendering at a specific frame rate, aspects of the present disclosure can render at a lower frame rate, perform motion estimation on the rendered frames, and then use the motion estimation to extrapolate the content. This content can then be interspersed with other content to produce a higher frame rate.

[0050] Some aspects of VR, AR, XR and / or gaming applications may have stringent constraints on both power and operational efficiency. Accordingly, these applications may have constraints on both increasing performance and reducing utilized power consumption. To alleviate these constraints, motion estimation may be performed on previously rendered content used to extrapolate frames. This, in turn, may allow frame rendering to be processed at a reduced frame rate. For example, to help reduce performance and power, every other frame may be rendered rather than every frame. Motion estimation may then be performed between frames. In some aspects, motion estimation may be performed between rendered frames, and this may yield information about how objects and pixel groups of the image have moved. Using motion estimation and previously rendered frames, frames may be extrapolated rather than rendered.

[0051] While running at a reduced frame rate can yield power consumption and performance gains, the cost of performing high-quality motion estimation can also be a significant overhead, for example when deploying a frame extrapolation solution. Therefore, motion estimation can also utilize resources and / or require high costs. Consequently, the cost of motion estimation can reduce the power and performance gains of the reduced frame rate when performing frame extrapolation. Accordingly, while the frame rate can be improved, there may also be an increase in the amount of resources required to perform motion estimation.

[0052] Some aspects of performing motion estimation may not require uniform motion estimation quality across the image. There may be several contributing factors that motivate the need for non-uniform motion estimation quality. For example, a lens on an AR or VR device may have reduced optical clarity around the edges. So, at the center of the lens, the picture can be sharp, but the image near the edge of the lens may be blurry.

[0053] Furthermore, barrel distortion can cause over-rendering at the edges of an image, so images near the edge of the display may appear over-rendered. For example, a standard projection of a map may have polar regions that appear overly large. Furthermore, the human eye perceives high resolution in a very narrow region near the center of the eye, known as the fovea. Therefore, the region near the fovea may contain the highest acuity, while images outside the fovea may appear more natural.

[0054] In some aspects, the motion estimation workload can be scaled along with the above-described parameters, where the highest quality motion estimation can occur near the center of the image, intermediate quality motion estimation can occur in the middle of the image, and in some aspects, no motion estimation can be performed near the edges of the image. Therefore, aspects of the present disclosure can use the above-described parameters when performing motion estimation, such as by adjusting motion estimation to focus computational resources on areas close to focus and / or reduce computation for areas outside of focus.

[0055] In some aspects, in the presence of eye tracking, the quality of motion estimation can also be scaled based on the distance from the focus, for example because lower quality or inadequacy of motion estimation may not be perceptible near edge regions. Therefore, aspects of the present disclosure can introduce varying levels of computation applied across different regions of the image based on the focus of the eye.

[0056] Some aspects of the present disclosure may include several parameters that may be included in adjusting the quality of motion estimation. For example, aspects of the present disclosure may determine whether to perform motion estimation or not for a certain area. Thus, aspects of the present disclosure may not perform motion estimation in edge areas and perform motion estimation in center areas. In areas where motion estimation is not performed, the motion vector for the group of pixels in the area may be set to a constant, such as zero. In some implementations, the constant for the area may be based on the average motion vector of the area where motion estimation is performed.

[0057] In addition, the density of the motion vectors generated can be a factor in adjusting the quality of the motion estimation. For example, a vector can correspond to a certain number of pixels, such as an 8 by 8 block of pixels, or a vector can correspond to a single pixel. In some aspects, the motion vectors generated to describe the motion of pixels in the image can depend on how the motion estimation is performed. As such, there can be different levels of granularity in the motion estimation. So the granularity of the motion estimation can vary across different areas of the image. For example, a higher priority area can use a smaller group of pixels (e.g., a 4 by 4 block of pixels), while a lower priority area can use a larger group of pixels (e.g., a 16 by 16 block of pixels).

[0058] In some aspects, a search window can be utilized during motion estimation. For example, a search window of a particular size can exist around each point from which motion estimation is performed. So if a point is in a portion of an image, such as the upper left corner of the image, a search window can be utilized around that point so that candidate positions for motion estimation are within that particular area of the image. For example, aspects of the present disclosure can have search windows of varying sizes at the edges of an image or at the center of the image. For example, the center of an image can have a larger search window than at the edges of the image. So the search window size for each potential vector can be adjusted across different areas of the image.

[0059] In some instances, a search window can control the maximum magnitude of motion vectors that can be generated. For example, an algorithm can perform block matching so that blocks from a first image are compared with corresponding blocks in a subsequent image. Blocks can be matched through various calculations that can provide information about how an area has moved from frame to frame or image to image. The search window can determine how far in a subsequent image from its location in the first image to search for a matching block. Therefore, in some instances, to reduce workload, a smaller search window can be utilized for lower priority areas, as larger search windows can be more resource intensive.

[0060] In some aspects, the inclusion or omission of a multi-path algorithm with motion vector refinement can be utilized during motion estimation. Some motion estimation approaches can have multiple paths or multiple stages in the algorithm. There can be a hierarchical structure where motion estimation starts at a low resolution and then is performed at a high resolution during subsequent steps. Accordingly, motion estimation can be improved and resolution can be increased. So there can be multiple stages where a course search is performed and then a refinement stage and / or vector filtering. In some implementations, a coarse search can be performed for lower priority areas and a refinement stage and / or vector filtering stage can be performed only for higher priority areas.

[0061] In some instances, a more complex algorithm with multiple paths may be more exhaustive and more expensive than a simpler algorithm that may be less accurate. Aspects of the present disclosure may also vary the algorithm used for motion estimation depending on the region, such as a more exhaustive algorithm near focus and a simpler algorithm near edges.

[0062] In addition, the block size of motion estimation can utilize the sum of absolute differences (SAD) operation or the sum of squared differences (SSD) operation as a block matching operation. The algorithm used for motion estimation can perform a block matching operation. The SAD and / or SSD operation can determine how well these blocks match each other. Various aspects of the present disclosure can also change the size of the blocks used based on the region of the image. Therefore, larger blocks can be used for central areas with higher priority, and smaller blocks can be used for edge areas with lower priority.

[0063] In addition, filtering can be performed during the motion search and / or on a certain path of motion vectors (e.g., the final motion vector). In some aspects, there may be some type of smoothing or post-processing of the vector data during the filtering path. For example, there may be noise in the motion estimate that may need to be filtered. Filtering may also consume processing resources. As such, aspects of the present disclosure may perform more thorough filtering in areas of greater importance (e.g., higher priority areas near the center of the image) and less thorough filtering in areas of lesser importance (e.g., lower priority areas near the edge of the image).

[0064] Aspects of the present disclosure may include several different implementations of performing the above-described processes. For example, quality assessment may be implemented in conjunction with a viewer and frame extrapolation joint test application. In some aspects, multiple level parameter selections may be performed, such as full quality motion estimation or no motion estimation. In some instances, full motion estimation may be performed, for example, in a circular region having a radius of 75% of the distance from the center of the lens to the edge of the eye buffer. In these instances, quality may not be significantly affected, allowing animation to remain smooth in the center lens area where clarity is high. Additionally, these above-described motion estimation methods may result in a reduction in motion estimation workload, for example, a 56% reduction in motion estimation workload. Furthermore, adding eye tracking may provide further workload reduction, as the high-quality region may follow the user's concave region, which may allow for more aggressive motion estimation parameters.

[0065] Figure 33 is a schematic diagram 300 illustrating regions of a frame 302. Regions can be defined based on the center of the frame. Where eye tracking is used, regions can be defined based on the focus of the eye. For example, regions can generally be defined as circular or annular concentric areas. A central region 310 can be located at the center of the frame or the focus of the eye. Central region 310 can be assigned the highest motion estimation priority. Edge region 340 can be located near the edge of the frame. For example, points within edge region 340 can be located at a distance from the center or focus of the frame that is greater than a threshold distance. In some implementations, the threshold distance can be 75% of the distance from the center of the frame to the edge of the eye buffer. Edge region 340 can be assigned the lowest motion estimation priority. In some implementations, a region can include one or more intermediate regions located between central region 310 and edge region 340. For example, as shown, a first intermediate region 320 can surround central region 310, and a second intermediate region 330 can surround first intermediate region 320. In some implementations, each intermediate region can be defined by a threshold distance from the center of the frame or the focus of the eye. For areas farther from the center of the frame or the focus of the eye, the intermediate areas may have decreasing motion estimation priorities. For example, as shown using four areas, the order of motion estimation priorities may be: center area 310, first intermediate area 320, second intermediate area 330, edge area 340. Motion vector quality or motion estimation quality may be assigned to each area based on the order of motion estimation priority.

[0066] Figure 4A and Figure 4B Motion vectors in example frames 400 and 410 , respectively, are shown in accordance with one or more techniques of this disclosure. Figure 4A Motion vectors are shown in frame 400 without utilizing the above-described motion estimation process of the present disclosure. Figure 4B The motion vectors in frame 410 are shown using the above-described motion estimation process of the present disclosure. Figure 4A As shown in , motion vectors are estimated for groups of pixels across the frame 400. Figure 4BAs shown in FIG, motion vectors are estimated only for pixel groups within a central region 412. Motion vectors for pixel groups within an edge region 414 can be set to a constant value (e.g., 0 for no motion). Consequently, the estimated motion vectors are more concentrated near regions of greater importance, such as near the center of frame 410. That is, motion vectors can be estimated for every pixel group within central region 412 near the center of frame 410. Accordingly, when performing frame extrapolation, the motion of the pixel groups in central region 412 can be more accurate. To the user, the movement of objects represented in frame 410 can appear as smooth as the movement of objects in frame 400. However, compared to the estimation of motion vectors for frame 400, the estimation of motion vectors for frame 410 can consume significantly fewer resources. For example, in an implementation where region 412 is defined as 75% of the distance from the center of the lens to the edge of the eye buffer and motion estimation is not performed for edge region 414, the overall motion estimation workload can be reduced by 56% compared to frame 400. Accordingly, aspects of the present disclosure provide for more concentrated estimation of motion vectors and associated use of computational resources near regions of greater importance (e.g., near the center of an image). Further, aspects of the present disclosure can reduce the number of motion estimates performed in regions of lesser importance (e.g., regions near the edges of an image), thereby reducing motion estimation workload and improving frame rate and / or power consumption.

[0067] Figure 5A 、 Figure 5B and Figure 5C The results of an example implementation using different motion estimation techniques for different regions are shown. For example, Figure 5A is based on Figure 4A The motion vectors are used to generate a photograph of an example of image 500, and Figure 5B is based on Figure 4B The motion vector is used to generate the photograph of the image 510 . Figure 5C It is shown in Figure 5A and Figure 5B A heat map 520 of the differences between regions. Figure 5A and Figure 5B The difference between will be difficult for a user to discern, especially when viewed as one frame in a series of frames. Figure 5C As shown in , the difference is only in edge areas where the user is less likely to look. Accordingly, the disclosed techniques using reduced motion estimation for edge areas can provide a significant reduction in the use of processing resources without significantly reducing the perceived quality.

[0068] Figure 6An example flow chart illustrating an example method 600 according to one or more techniques of this disclosure. Method 600 may be performed by a GPU or device for graphics processing. Optional blocks are illustrated using dashed lines.

[0069] At block 610, method 600 may optionally include tracking the position of the eye. For example, the apparatus may include an eye tracker having one or more sensors for the user's eye to determine the position of the eye. The eye tracker may determine the focus of the eye on the display. The eye tracker may determine a corresponding point within the frame. Processing unit 120 and / or determination component 198 may receive an indication of the position of the eye relative to the frame.

[0070] At block 620, method 600 may include determining a motion estimation priority for one or more regions in a frame, wherein each of the one or more regions includes at least one pixel group. For example, an apparatus including processing unit 120 and / or determining component 198 may determine a motion estimation priority for one or more regions in a frame, such as in conjunction with Figure 4A and Figure 4B As described in the examples in . In some aspects, each of the one or more regions can include a group of pixels. For example, the one or more regions can include a center region and an edge region. In some implementations, the one or more regions include at least one intermediate region between the center region and the edge region. In some implementations, the one or more regions in the frame are defined relative to the position of the eye.

[0071] At block 630, method 600 may include adjusting the quality of the motion estimation process for each pixel group based on the motion estimation priority of the region including the pixel group. For example, an apparatus including processing unit 120 and / or determining component 198 may adjust the quality of the motion estimation process for each pixel group based on the motion estimation priority, as described in conjunction with Figure 4A and Figure 4B As described in the examples.

[0072] For example, in some implementations, at sub-block 632, block 630 may optionally include determining to set a constant value for at least one pixel group in the region having the lowest motion estimation priority. In some implementations, at sub-block 634, block 630 may optionally include setting the size of at least one pixel group based on the motion estimation priority. In some implementations, at sub-block 636, block 630 may optionally include adjusting the size of a search window to be covered during the motion estimation process based on the motion estimation priority. In some implementations, at sub-block 638, block 630 may optionally include determining to include or omit a multi-path algorithm with motion vector refinement in the motion estimation process based on the motion estimation priority. In some implementations, at sub-block 640, block 630 may optionally include determining a block size for a SAD operation or an SSD operation based on the motion estimation priority. In some implementations, at sub-block 642, block 630 may optionally include adjusting the amount of filtering performed during motion search filtering or on the final motion vector based on the motion estimation priority. In one aspect, block 630 may include sub-blocks 632, 634, 636, 638, 640, or two or more of 642 in combination.

[0073] At block 650, method 600 may include performing a motion estimation process for each of the at least one pixel group based on the adjusted quality of the motion estimation process to generate a motion vector for the at least one pixel group. For example, an apparatus including processing unit 120 and / or determining component 198 may further perform a motion estimation process for each of the one or more regions based on the adjusted quality of the motion estimation process, as described in conjunction with Figure 4A and Figure 4B As described in the examples.

[0074] At block 660, method 600 may optionally include extrapolating a new frame from the frame based on the motion vector for each pixel group. For example, processing unit 120 and / or determination component 198 may extrapolate a new frame (e.g., frame 510) from the frame based on the motion vector for each pixel group.

[0075] In one configuration, a method or apparatus for graphics processing is provided. The apparatus may be a GPU or some other processor that can perform graphics processing. In one aspect, the apparatus may be a processing unit 120 within the device 104, or may be some other hardware within the device 104 or another device. The apparatus may include a unit for tracking the position of the eye. The apparatus may include a unit for determining a motion estimation priority for one or more regions in a frame. The apparatus may also include a unit for adjusting the quality of the motion estimation process for each pixel group based on the motion estimation priority of the region that includes the pixel group. The apparatus may also include a unit for performing a motion estimation process for each pixel group in at least one pixel group based on the adjusted quality of the motion estimation process to generate a motion vector for the at least one pixel group. The apparatus may also include a unit for extrapolating a new frame from a frame based on the motion vector for each pixel group.

[0076] The subject matter described herein can be implemented to achieve one or more benefits or advantages. For example, the graphics processing techniques described can be implemented at a low cost compared to other graphics processing techniques. In addition, the graphics processing techniques herein can improve or accelerate data processing or execution of a GPU. Further, the graphics processing techniques herein can improve resource or data utilization and / or resource efficiency of a GPU. In addition, aspects of the present disclosure can improve motion estimation of a GPU, which can in turn improve the accuracy and efficiency of the GPU. Aspects of the present disclosure can also improve the accuracy of motion estimation.

[0077] According to the present disclosure, where the context does not indicate otherwise, the term "or" may be interrupted to "and / or." In addition, although phrases such as "one or more" or "at least one" may be used for some features disclosed herein but not for other features, where the context does not indicate otherwise, the features for which such language is not used may be interpreted as having such implicit meanings.

[0078] In one or more examples, the functionality described herein may be implemented in hardware, software, firmware, or any combination thereof. For example, although the term "processing unit" has been used throughout this disclosure, such a processing unit may be implemented in hardware, software, firmware, or any combination thereof. If any functionality, processing unit, technique, or other module described herein is implemented in software, then the functionality, processing unit, technique, or other module described herein may be stored on a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. Computer-readable media may include computer data storage media or communication media, including any media that facilitates the transfer of a computer program from one place to another. In this manner, a computer-readable medium may generally correspond to (1) a tangible computer-readable storage medium that is non-transitory, or (2) a communication medium such as a signal or carrier wave. A data storage medium may be any available medium that can be accessed by one or more computers or one or more processors to retrieve instructions, code, and / or data structures for the implementation of the techniques described in this disclosure. By way of example and not limitation, such computer-readable media may include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage, or other magnetic storage devices. As used herein, disk and optical disc include compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk, and Blu-ray disc, where disks typically reproduce data magnetically, while optical discs reproduce data optically using lasers. Combinations of the above are also intended to be included within the scope of computer-readable media. A computer program product may include computer-readable media.

[0079] The code may be executed by one or more processors, such as one or more digital signal processors (DSPs), general-purpose microprocessors, application-specific integrated circuits (ASICs), arithmetic logic units (ALUs), field-programmable logic devices (FPGAs), or other equivalent integrated or discrete logic circuits. Accordingly, the term "processor" as used herein may refer to any of the foregoing structures or any other structure that may be suitable for the implementation of the techniques described herein. Additionally, the techniques may be fully implemented in one or more circuits or logic elements.

[0080] The techniques of the present disclosure can be implemented in a wide variety of devices or apparatuses, including wireless handsets, integrated circuits (ICs), or sets of ICs, such as chipsets. Various components, modules, or units are described in this disclosure to emphasize functional aspects of devices configured to perform the disclosed techniques, and do not necessarily need to be implemented by different hardware units. Specifically, as described above, the various units can be combined in any hardware unit, or provided by multiple interoperable hardware units (including one or more processors as described above) in combination with appropriate software and / or firmware.

[0081] Various examples have been described. These and other examples are within the scope of the following claims.

Claims

1. A method for graphics processing, comprising: determining a motion estimation priority for one or more regions in a frame, wherein each of the one or more regions includes at least one pixel group; adjusting a quality of a motion estimation process for each pixel group based on a motion estimation priority of a region including the pixel group, wherein adjusting the quality of the motion estimation process comprises: determining a constant value to be set for at least one pixel group in a region having a lowest motion estimation priority for reducing a number of motion estimations performed at the region having the lowest motion estimation priority; and A motion estimation process is performed for each of the at least one pixel group based on the adjusted quality of the motion estimation process to generate a motion vector for the at least one pixel group.

2. The method according to claim 1, wherein The one or more regions include a central region and an edge region.

3. The method according to claim 2, wherein: The one or more regions include at least one intermediate region between the central region and the edge region.

4. The method according to claim 1, further comprising: A position of an eye is tracked, wherein the one or more regions in the frame are defined relative to the position of the eye.

5. The method according to claim 1, wherein Adjusting the quality of the motion estimation process for each pixel group based on a motion estimation priority of a region including the pixel group includes setting a size of the at least one pixel group based on the motion estimation priority of the region.

6. The method according to claim 1, wherein Adjusting the quality of the motion estimation process for each pixel group based on a motion estimation priority of a region including the pixel group includes adjusting a size of a search window to be covered during the motion estimation process based on the motion estimation priority of the region.

7. The method according to claim 1, wherein Adjusting the quality of the motion estimation process for each pixel group based on a motion estimation priority of a region including the pixel group includes determining inclusion or omission of a multi-path algorithm with motion vector refinement in the motion estimation process based on the motion estimation priority of the region.

8. The method according to claim 1, wherein Adjusting the quality of the motion estimation process for each pixel group based on the motion estimation priority of the region including the pixel group includes: determining a block size for a sum of absolute errors (SAD) operation or a sum of squared errors (SSD) operation based on the motion estimation priority of the region.

9. The method according to claim 1, wherein Adjusting the quality of the motion estimation process for each pixel group based on the motion estimation priority of the region comprising the pixel group includes adjusting the amount of filtering performed during motion search filtering or on the final motion vector based on the motion estimation priority of the region.

10. The method according to claim 1, further comprising: A new frame is extrapolated from the frame based on the motion vector for each pixel group.

11. A device for graphics processing, comprising: Memory; as well as at least one processor coupled to the memory and configured to: determining a motion estimation priority for one or more regions in a frame, wherein each of the one or more regions includes at least one pixel group; adjusting a quality of a motion estimation process for each pixel group based on a motion estimation priority of a region including the pixel group, wherein the at least one processor is configured to: determine a constant value to be set for at least one pixel group in a region having a lowest motion estimation priority for reducing a number of motion estimations performed at the region having the lowest motion estimation priority; and A motion estimation process is performed for each of the at least one pixel group based on the adjusted quality of the motion estimation process to generate a motion vector for the at least one pixel group.

12. The device according to claim 11, wherein The one or more regions include a central region and an edge region.

13. The device according to claim 12, wherein The one or more regions include at least one intermediate region between the central region and the edge region.

14. The device according to claim 11, wherein The at least one processor is configured to track a position of an eye, wherein the one or more regions in the frame are defined relative to the position of the eye.

15. The device according to claim 11, wherein The at least one processor is configured to set a size of the at least one pixel group based on the motion estimation priority of the region.

16. The device according to claim 11, wherein The at least one processor is configured to adjust a size of a search window to be covered during the motion estimation process based on the motion estimation priority of the region.

17. The device according to claim 11, wherein The at least one processor is configured to determine inclusion or omission of a multi-path algorithm with motion vector refinement in the motion estimation process based on the motion estimation priority of the region.

18. The device according to claim 11, wherein The at least one processor is configured to determine a block size for a sum of absolute differences (SAD) operation or a sum of squared differences (SSD) operation based on the motion estimation priority of the region.

19. The device according to claim 11, wherein The at least one processor is configured to adjust an amount of filtering performed during motion search filtering or on a final motion vector based on the motion estimation priority of the region.

20. The device according to claim 11, wherein The at least one processor is configured to extrapolate a new frame from the frame based on the motion vector for each pixel group.

21. A device for graphics processing, comprising: means for determining a motion estimation priority for one or more regions in a frame, wherein each of the one or more regions includes at least one pixel group; means for adjusting a quality of a motion estimation process for each pixel group based on a motion estimation priority of a region including the pixel group, wherein the means for adjusting the quality of the motion estimation process is configured to: determine a constant value to be set for at least one pixel group in a region having a lowest motion estimation priority for reducing the number of motion estimations performed at the region having the lowest motion estimation priority; and Means for performing a motion estimation process for each of the at least one pixel group to generate a motion vector for the at least one pixel group based on the adjusted quality of the motion estimation process.

22. The apparatus according to claim 21, further comprising: Means for tracking a position of an eye, wherein the one or more regions in the frame are defined relative to the position of the eye.

23. The device according to claim 21, wherein The one or more regions include a central region and an edge region.

24. The apparatus according to claim 21, wherein The means for adjusting the quality of the motion estimation process for each pixel group based on a motion estimation priority of a region comprising the pixel group is configured to set a size of the at least one pixel group based on the motion estimation priority of the region.

25. The apparatus according to claim 21, wherein The unit for adjusting the quality of the motion estimation process for each pixel group based on the motion estimation priority of the region comprising the pixel group is configured to: adjust the size of a search window to be covered during the motion estimation process based on the motion estimation priority of the region.

26. The apparatus of claim 21, further comprising: Means for extrapolating a new frame from the frame based on the motion vector for each group of pixels.

27. A non-transitory computer-readable medium storing computer-executable code for graphics processing, comprising code for: Determining motion estimation priorities for one or more regions in a frame, wherein Each of the one or more regions includes at least one pixel group; adjusting a quality of a motion estimation process for each pixel group based on a motion estimation priority of a region including the pixel group, wherein the code for adjusting the quality of the motion estimation process comprises: determining a constant value to be set for at least one pixel group in a region having a lowest motion estimation priority for reducing a number of motion estimations performed at the region having the lowest motion estimation priority; and A motion estimation process is performed for each of the at least one pixel group based on the adjusted quality of the motion estimation process to generate a motion vector for the at least one pixel group.

Images