This disclosure relates generally to video processing and, more particularly, to generation of optical flow maps based on foreground and background image segmentation.
Many video applications utilize optical flow maps to interpolate between stereo image pairs included in input stereo video data. In some examples, each image of a stereo image pair is associated with a different camera field-of-view. Furthermore, in some such examples, each image of a stereo image pair is also associated with a respective optical flow map including an array of elements corresponding respectively to the pixels of that image. In such examples, each element of the optical flow map includes a displacement vector to map a respective pixel of that image to a corresponding pixel in the other image of the stereo image pair. Thus, video applications, such as an immersive video application, can utilize optical flow maps to interpolate between a first image in a stereo image pair and a second image in the stereo image pair to generate an interpolated image representative of a virtual camera field-of-view between a first camera field-of-view associated with the first image and a second camera field-of-view associated with the second image.
In general, the same reference numbers will be used throughout the drawing(s) and accompanying written description to refer to the same or like parts. The figures are not to scale.
As used herein, connection references (e.g., attached, coupled, connected, and joined) may include intermediate members between the elements referenced by the connection reference and/or relative movement between those elements unless otherwise indicated. As such, connection references do not necessarily infer that two elements are directly connected and/or in fixed relation to each other. As used herein, stating that any part is in “contact” with another part is defined to mean that there is no intermediate part between the two parts.
Unless specifically stated otherwise, descriptors such as “first,” “second,” “third,” etc., are used herein without imputing or otherwise indicating any meaning of priority, physical order, arrangement in a list, and/or ordering in any way, but are merely used as labels and/or arbitrary names to distinguish elements for ease of understanding the disclosed examples. In some examples, the descriptor “first” may be used to refer to an element in the detailed description, while the same element may be referred to in a claim with a different descriptor such as “second” or “third.” In such instances, it should be understood that such descriptors are used merely for identifying those elements distinctly that might, for example, otherwise share a same name.
As used herein, “approximately” and “about” modify their subjects/values to recognize the potential presence of variations that occur in real world applications. For example, “approximately” and “about” may modify dimensions that may not be exact due to manufacturing tolerances and/or other real world imperfections as will be understood by persons of ordinary skill in the art. For example, “approximately” and “about” may indicate such dimensions may be within a tolerance range of +/−10% unless otherwise specified in the below description. As used herein “substantially real time” refers to occurrence in a near instantaneous manner recognizing there may be real world delays for computing time, transmission, etc. Thus, unless otherwise specified, “substantially real time” refers to real time+/−1 second.
As used herein, the phrase “in communication,” including variations thereof, encompasses direct communication and/or indirect communication through one or more intermediary components, and does not require direct physical (e.g., wired) communication and/or constant communication, but rather additionally includes selective communication at periodic intervals, scheduled intervals, aperiodic intervals, and/or one-time events.
As used herein, “programmable circuitry” is defined to include (i) one or more special purpose electrical circuits (e.g., an application specific circuit (ASIC)) structured to perform specific operation(s) and including one or more semiconductor-based logic devices (e.g., electrical hardware implemented by one or more transistors), and/or (ii) one or more general purpose semiconductor-based electrical circuits programmable with instructions to perform specific functions(s) and/or operation(s) and including one or more semiconductor-based logic devices (e.g., electrical hardware implemented by one or more transistors). Examples of programmable circuitry include programmable microprocessors such as Central Processor Units (CPUs) that may execute first instructions to perform one or more operations and/or functions, Field Programmable Gate Arrays (FPGAs) that may be programmed with second instructions to cause configuration and/or structuring of the FPGAs to instantiate one or more operations and/or functions corresponding to the first instructions, Graphics Processor Units (GPUs) that may execute first instructions to perform one or more operations and/or functions, Digital Signal Processors (DSPs) that may execute first instructions to perform one or more operations and/or functions, XPUs, Network Processing Units (NPUs) one or more microcontrollers that may execute first instructions to perform one or more operations and/or functions and/or integrated circuits such as Application Specific Integrated Circuits (ASICs). For example, an XPU may be implemented by a heterogeneous computing system including multiple types of programmable circuitry (e.g., one or more FPGAs, one or more CPUs, one or more GPUs, one or more NPUs, one or more DSPs, etc., and/or any combination(s) thereof), and orchestration technology (e.g., application programming interface(s) (API(s)) that may assign computing task(s) to whichever one(s) of the multiple types of programmable circuitry is/are suited and available to perform the computing task(s).
As used herein integrated circuit/circuitry is defined as one or more semiconductor packages containing one or more circuit elements such as transistors, capacitors, inductors, resistors, current paths, diodes, etc. For example, an integrated circuit may be implemented as one or more of an ASIC, an FPGA, a chip, a microchip, programmable circuitry, a semiconductor substrate coupling multiple circuit elements, a system on chip (SoC), etc.
Example methods, apparatus, systems and articles of manufacture (e.g., physical storage media) to generate optical flow maps based on foreground and background image segmentation are disclosed herein. Optical flow maps are used in many video application to interpolate between a stereo pair of images corresponding to a frame of input stereo video data. Stereo video data includes a sequence of image frames, with each image frame including a pair of stereo images. In some examples, the pair of stereo images includes a first image corresponding to a first camera field-of-view and a second image corresponding to a second camera field-of-view. An optical flow map can be used by a video application to interpolate between the two images of the stereo image pair to generate a third image corresponding to a virtual camera field-of-view between the first camera and the second camera.
Example optical flow map generation techniques disclosed herein refine the optical flow maps determined for a stereo pair of images. Accurate optical flow maps are important as they enable high quality view interpolation results to be achieved. For example, prior optical map generation techniques face difficulties dealing with depth discontinuities, which can result in artifacts in view interpolation results.
Unlike such prior techniques, example optical flow map generation techniques disclosed herein refine, or enhance, the optical flow maps between a stereo pair of images using foreground and background segmentation results. Such refinements are better able to deal with depth discontinuities between the foreground and background segments of the stereo pair of images, thereby reducing or eliminating artifacts in the view interpolation results. In some disclosed examples, a deep learning-based approach is used to estimate alpha mattes for the captured input images in an input stereo image pair using a single background stereo image pair for the scene of interest as a reference. The generated alpha mattes include elements that represent the contributions of the individual pixels of the input images to the foreground versus background image segments. Then, using the generated alpha mattes as a guide, disclosed example optical flow map generation techniques combine, or merge, initial optical flow maps generated for the captured input images in the input stereo image pair with reference optical flow maps generated for the background stereo image pair to generate output optical flow maps for the captured input images in the input stereo image pair. The resulting output optical flow maps enable generation of higher quality view interpolation results than the initial optical flow maps generated for the input images in the input stereo image pair. Furthermore, disclosed example optical flow map generation techniques are compatible with any existing or future technique for generating the initial optical flow maps and reference optical flow maps, and/or for generating the alpha mattes. Also, it is worth noting that generation of the reference optical flow maps for the background frame can be done just once as a training or initialization procedure, thereby contributing minimal processing overhead.
These and other example methods, apparatus, systems and articles of manufacture (e.g., physical storage media) to implement generate optical flow maps based on foreground and background image segmentation are disclosed in further detail below.
Turning to the figures,
The video processor 105 of the illustrated example can be implemented by any number and/or types of video processors, graphic processing units (GPUs), video applications executed by GPUs and/or other programmable circuitry, etc. For example, the video processor 105 may implement/execute one or more computer vision applications involved in analyzing multi-view images, performing three-dimensional (3D) scene perception, implementing object tracking algorithms for surveillance systems, etc. Such video processing applications can benefit from accurate generation of optical flow maps by the optical map refinement circuitry 110 in accordance with the teachings of this disclosure. As another example, the video processor 105 may implement/execute one or more video applications to generate immersive virtual reality (VR) or augmented reality (AR) content using 360-degree Camera arrays including the first camera 115 is and the second camera 120. In such examples, the optical flow maps generated by the optical map refinement circuitry 110 can be used to interpolate between the available camera field-of-views to enhance the user's immersive experience.
In the illustrated example of
After completion of the training phase 130, the optical map refinement circuitry 110 enters the operational phase 135 and causes the video processor 105 to capture input stereo video which includes stereo pairs of input images corresponding to the first field-of-view associated with the first camera 115 and the second field-of-view associated with the second camera 120. Each stereo pair of input images is representative of the scene 125 after the addition of the foreground content (e.g., objects) of interest, such as the example foreground object 140 (e.g., a person) illustrated in
In the illustrated example of
A block diagram of an example implementation of the optical map refinement circuitry 110 of
The example optical map refinement circuitry 110 of
The background optical flow estimation circuitry 205 of the illustrated example then generates example reference optical flow maps 235 (also referred to as background optical flow maps 235) for the scene based on the stereo pair of background images 230. For example, assuming the stereo pair of background images 230 includes the first background image corresponding to the first camera field-of-view and the second background image corresponding to the second camera field-of-view, the background optical flow estimation circuitry 205 generates a first reference optical flow map 235 associated with the first background image and a second reference optical flow map 235 associated with the second background image. In such an example, the first reference optical flow map 235 includes an array of elements corresponding respectively to the pixels of the first background image such that the array of elements includes respective displacement vectors to map ones of the pixels of the first background image (which is associated with the first field-of-view) to corresponding ones of the pixels of the second background image (which is associated with the second field-of-view) of the stereo pair of background images 230. Likewise, the second reference optical flow map 235 includes an array of elements corresponding respectively to the pixels of the second background image such that the array of elements includes respective displacement vectors to map ones of the pixels of the second background image (which is associated with the second field-of-view) to corresponding ones of the pixels of the first background image (which is associated with the first field-of-view) of the stereo pair of background images 230. In some examples, the displacement vectors are data tuples containing magnitude and direction values to map a pixel from one of the background images to the other one of the background images in the stereo background image pair 230. The background optical flow estimation circuitry 205 can use any number and/or types of algorithms to generate optical flow maps from a stereo pair of background images. For example, the background optical flow estimation circuitry 205 can implement optical flow map generation using the PixFlow algorithm included in the Surround 360 System by Facebook®. However, the background optical flow estimation circuitry 205 is not limited to that algorithm but, rather, can implement any existing or future optical flow estimation algorithm to generate the reference optical flow maps 235 from the stereo pair of background images 230.
In the illustrated example, the captured optical flow estimation circuitry 210 causes the optical map refinement circuitry 110 to operate in the operational phase 135 described above. As described above, during the operational phase 135, the captured optical flow estimation circuitry 210 accesses example stereo pairs of input images 240 (e.g., a sequence of stereo pairs of input images 240 corresponding to frames of an input stereo video) that are representative of a first camera field-of-view and a second camera field-of-view of a scene (e.g., the scene 125) after the addition of the foreground content (e.g., objects) of interest. For example, a given stereo pair of input images 240 corresponding to a given frame time of an input stereo video can include a first input image containing a first array of pixels captured from the first camera field-of-view and a second input image containing a second array of pixels captured from the second camera field-of-view. The pixels can have values from an RGB color space, a YUV) color space, etc., and can conform with any appropriate image resolution.
The captured optical flow estimation circuitry 210 of the illustrated example then generates example input optical flow maps 245 (also referred to as initial optical flow maps 245) for the scene based on the given stereo pair of input images 240. For example, assuming the stereo pair of input images 240 includes the first input image corresponding to the first camera field-of-view and the second input image corresponding to the second camera field-of-view, the captured optical flow estimation circuitry 210 generates a first input optical flow map 245 associated with the first input image and a second input optical flow map 245 associated with the second input image. In such an example, the first input optical flow map 245 includes an array of elements corresponding respectively to the pixels of the first input image such that the array of elements includes respective displacement vectors to map ones of the pixels of the first input image (which is associated with the first field-of-view) to corresponding ones of the pixels of the second input image (which is associated with the second field-of-view) of the stereo pair of input images 240. Likewise, the second input optical flow map 245 includes an array of elements corresponding respectively to the pixels of the second input image such that the array of elements includes respective displacement vectors to map ones of the pixels of the second input image (which is associated with the second field-of-view) to corresponding ones of the pixels of the first input image (which is associated with the first field-of-view) of the stereo pair of input images 240. In some examples, the displacement vectors are data tuples containing magnitude and direction values to map a pixel from one of the input images to the other one of the input images in the stereo input image pair 240. The captured optical flow estimation circuitry 210 can use any number and/or types of algorithms to generate optical flow maps from a stereo pair of input images. For example, the captured optical flow estimation circuitry 210 can implement optical flow map generation using the PixFlow algorithm included in the Surround 360 System by Facebook®. However, the captured optical flow estimation circuitry 210 is not limited to that algorithm but, rather, can implement any existing or future optical flow estimation algorithm to generate the input optical flow maps 245 from the stereo pair of input images 240. Also, the captured optical flow estimation circuitry 210 can implement the same optical flow estimation algorithm or a different optical flow estimation algorithm than the background optical flow estimation circuitry 205.
In the illustrated example, the alpha matte estimation circuitry 215 is active in the operational phase 135 to generate example alpha mattes 250 for the accessed stereo pairs of input images 240 in the sequence of stereo input image pairs corresponding to frames of an input stereo video. The alpha mattes 250 represent the segmentation of the stereo pairs of input images 240 into foreground and background regions. For example, for a given accessed stereo pair of input images 240 that includes a first input image corresponding to the first camera field-of-view and a second input image corresponding to the second camera field-of-view, the alpha matte estimation circuitry 215 generates a first alpha matte associated with the first input image corresponding to the first camera field-of-view and a second alpha matte associated with the second input image corresponding to the second camera field-of-view. In some examples, the first alpha matte includes an array of elements corresponding respectively to the pixels of the first input image such that the array of elements includes respective alpha values, or weights, that represent how much the different pixels of the first input image (which is associated with the first field-of-view) contribute to the foreground or the background of the first input image. Likewise, the second alpha matte includes an array of elements corresponding respectively to the pixels of the second input image such that the array of elements includes respective alpha values, or weights, that represent how much the different pixels of the second input image (which is associated with the second field-of-view) contribute to the foreground or the background of the second input image. In some examples, the alpha values, or weights, of a given alpha matte are values in the range of 0 to 1, with a value of 1 corresponding to the foreground region of the image and a value of 0 corresponding to the background region of the image. In such examples, larger alpha values, or weights, represent that a corresponding pixel contributes more heavily to, or is more likely to correspond to, the foreground region of the image, whereas smaller alpha values, or weights, represent that a corresponding pixel contributes more heavily to, or is more likely to correspond to, the background region of the image.
In the illustrated example of
In the illustrated example, the optical flow merge circuitry 220 is active in the operational phase 135 to combine, or merge, the input (or initial) optical flow maps 245 for a given stereo input image pair 240 with the background optical flow maps 235 for the stereo background image pair 230 to generate example output optical flow maps 255 for the given stereo input image pair 240. In the illustrated example, the optical flow merge circuitry 220 performs such optical flow map combining using the alpha mattes 250 for the given stereo input image pair 240 as a guide. In some examples, the optical flow merge circuitry 220 combines the input (or initial) optical flow maps 245 with the background optical flow maps 235 elementwise using the alpha mattes 250 as a guide. For example, for a given stereo input image pair 240, the optical flow merge circuitry 220 can combine elements of the first background optical flow map 235 associated with the first camera field-of-view with the corresponding elements of the first input optical flow map 245 corresponding to the first input image of the stereo input image pair 240 (which corresponds to the first field-of-view) based on the corresponding elements of the first alpha matte 250 corresponding to the first input image of the stereo input image pair 240 (which corresponds to the first field-of-view). Likewise, for the given stereo input image pair 240, the optical flow merge circuitry 220 can combine elements of the second background optical flow map 235 associated with the second camera field-of-view with the corresponding elements of the second input optical flow map 245 corresponding to the second input image of the stereo input image pair 240 (which corresponds to the second field-of-view) based on the corresponding elements of the second alpha matte 250 corresponding to the second input image of the stereo input image pair 240 (which corresponds to the second field-of-view).
In some examples, the optical flow merge circuitry 220 performs optical flow map combining based on Equation 1, which is:
In Equation 1, OFrefine refers to the output (or refined) optical flow maps 255, OFbgr refers to the reference optical flow maps 235 associated with the background frame, OFfgr refers to the optical flow maps 245 of associated with the given input image frame, and alpha refers to the alpha mattes 250 for the given input image frame. Thus, to generate an output optical flow map 255 for a given image of a given stereo input image frame 240 based on Equation 1, the optical flow merge circuitry 220 multiplies values of the elements of the reference optical flow map 235 for that image field-of-view (e.g., OFbgr in Equation 1) by the values of the corresponding elements of the alpha matte 250 for that image field-of-view (e.g., w=alpha in Equation 1) subtracted from one (e.g., (1−w) in Equation 1) to determine respective first quantities (e.g., (1−w)×OFbgr in Equation 1). The optical flow merge circuitry 220 also multiplies values of the corresponding elements of the input optical flow map 245 for that image field-of-view (e.g., OFfgr in Equation 1) with the values of the corresponding elements of the alpha matte 250 for that image field-of-view (e.g., w=alpha in Equation 1) to determine respective second quantities (e.g., w×OFfgr in Equation 1). The optical flow merge circuitry 220 further adds the first quantities (e.g., (1−w)×OFbgr in Equation 1) and the second quantities (e.g., w×OFfgr in Equation 1) to determine the values of the corresponding elements included in the output optical flow map 255 for that image field-of-view (e.g., OFrefine=(1−w)×OFbgr+w×OFfgr of Equation 1).
In the illustrated example, the optical flow merge circuitry 220 outputs the generated, output optical flow maps 255 to a video processor, a video application, etc., such as the video processor 105 of
The example optical map refinement circuitry 110 of
In some examples, the optical map refinement circuitry 110 includes means for generating reference optical flow maps. For example, the means for generating reference optical flow maps may be implemented by the background optical flow estimation circuitry 205. In some examples, the background optical flow estimation circuitry 205 may be instantiated by programmable circuitry such as the example programmable circuitry 512 of
In some examples, the optical map refinement circuitry 110 includes means for generating input, or initial, optical flow maps. For example, the means for generating input optical flow maps may be implemented by the captured optical flow estimation circuitry 210. In some examples, the captured optical flow estimation circuitry 210 may be instantiated by programmable circuitry such as the example programmable circuitry 512 of
In some examples, the optical map refinement circuitry 110 includes means for generating alpha mattes. For example, the means for generating alpha mattes may be implemented by the alpha matte estimation circuitry 215. In some examples, the alpha matte estimation circuitry 215 may be instantiated by programmable circuitry such as the example programmable circuitry 512 of
In some examples, the optical map refinement circuitry 110 includes means for merging optical flow maps. For example, the means for merging optical flow maps may be implemented by the optical flow merge circuitry 220. In some examples, the optical flow merge circuitry 220 may be instantiated by programmable circuitry such as the example programmable circuitry 512 of
While an example manner of implementing the optical map refinement circuitry 110 is illustrated in
Flowchart(s) representative of example machine readable instructions, which may be executed by programmable circuitry to implement and/or instantiate the optical map refinement circuitry 110 of
The program may be embodied in instructions (e.g., software and/or firmware) stored on one or more non-transitory computer readable and/or machine readable storage medium such as cache memory, a magnetic-storage device or disk (e.g., a floppy disk, a Hard Disk Drive (HDD), etc.), an optical-storage device or disk (e.g., a Blu-ray disk, a Compact Disk (CD), a Digital Versatile Disk (DVD), etc.), a Redundant Array of Independent Disks (RAID), a register, ROM, a solid-state drive (SSD), SSD memory, non-volatile memory (e.g., electrically erasable programmable read-only memory (EEPROM), flash memory, etc.), volatile memory (e.g., Random Access Memory (RAM) of any type, etc.), and/or any other storage device or storage disk. The instructions of the non-transitory computer readable and/or machine readable medium may program and/or be executed by programmable circuitry located in one or more hardware devices, but the entire program and/or parts thereof could alternatively be executed and/or instantiated by one or more hardware devices other than the programmable circuitry and/or embodied in dedicated hardware. The machine readable instructions may be distributed across multiple hardware devices and/or executed by two or more hardware devices (e.g., a server and a client hardware device). For example, the client hardware device may be implemented by an endpoint client hardware device (e.g., a hardware device associated with a human and/or machine user) or an intermediate client hardware device gateway (e.g., a radio access network (RAN)) that may facilitate communication between a server and an endpoint client hardware device. Similarly, the non-transitory computer readable storage medium may include one or more mediums. Further, although the example program is described with reference to the flowchart(s) illustrated in
The machine readable instructions described herein may be stored in one or more of a compressed format, an encrypted format, a fragmented format, a compiled format, an executable format, a packaged format, etc. Machine readable instructions as described herein may be stored as data (e.g., computer-readable data, machine-readable data, one or more bits (e.g., one or more computer-readable bits, one or more machine-readable bits, etc.), a bitstream (e.g., a computer-readable bitstream, a machine-readable bitstream, etc.), etc.) or a data structure (e.g., as portion(s) of instructions, code, representations of code, etc.) that may be utilized to create, manufacture, and/or produce machine executable instructions. For example, the machine readable instructions may be fragmented and stored on one or more storage devices, disks and/or computing devices (e.g., servers) located at the same or different locations of a network or collection of networks (e.g., in the cloud, in edge devices, etc.). The machine readable instructions may require one or more of installation, modification, adaptation, updating, combining, supplementing, configuring, decryption, decompression, unpacking, distribution, reassignment, compilation, etc., in order to make them directly readable, interpretable, and/or executable by a computing device and/or other machine. For example, the machine readable instructions may be stored in multiple parts, which are individually compressed, encrypted, and/or stored on separate computing devices, wherein the parts when decrypted, decompressed, and/or combined form a set of computer-executable and/or machine executable instructions that implement one or more functions and/or operations that may together form a program such as that described herein.
In another example, the machine readable instructions may be stored in a state in which they may be read by programmable circuitry, but require addition of a library (e.g., a dynamic link library (DLL)), a software development kit (SDK), an application programming interface (API), etc., in order to execute the machine-readable instructions on a particular computing device or other device. In another example, the machine readable instructions may need to be configured (e.g., settings stored, data input, network addresses recorded, etc.) before the machine readable instructions and/or the corresponding program(s) can be executed in whole or in part. Thus, machine readable, computer readable and/or machine readable media, as used herein, may include instructions and/or program(s) regardless of the particular format or state of the machine readable instructions and/or program(s).
The machine readable instructions described herein can be represented by any past, present, or future instruction language, scripting language, programming language, etc. For example, the machine readable instructions may be represented using any of the following languages: C, C++, Java, C#, Perl, Python, JavaScript, HyperText Markup Language (HTML), Structured Query Language (SQL), Swift, etc.
As mentioned above, the example operations of
“Including” and “comprising” (and all forms and tenses thereof) are used herein to be open ended terms. Thus, whenever a claim employs any form of “include” or “comprise” (e.g., comprises, includes, comprising, including, having, etc.) as a preamble or within a claim recitation of any kind, it is to be understood that additional elements, terms, etc., may be present without falling outside the scope of the corresponding claim or recitation. As used herein, when the phrase “at least” is used as the transition term in, for example, a preamble of a claim, it is open-ended in the same manner as the term “comprising” and “including” are open ended. The term “and/or” when used, for example, in a form such as A, B, and/or C refers to any combination or subset of A, B, C such as (1) A alone, (2) B alone, (3) C alone, (4) A with B. (5) A with C. (6) B with C. or (7) A with B and with C. As used herein in the context of describing structures, components, items, objects and/or things, the phrase “at least one of A and B” is intended to refer to implementations including any of (1) at least one A, (2) at least one B, or (3) at least one A and at least one B. Similarly, as used herein in the context of describing structures, components, items, objects and/or things, the phrase “at least one of A or B” is intended to refer to implementations including any of (1) at least one A. (2) at least one B, or (3) at least one A and at least one B. As used herein in the context of describing the performance or execution of processes, instructions, actions, activities and/or steps, the phrase “at least one of A and B” is intended to refer to implementations including any of (1) at least one A, (2) at least one B, or (3) at least one A and at least one B. Similarly, as used herein in the context of describing the performance or execution of processes, instructions, actions, activities and/or steps, the phrase “at least one of A or B” is intended to refer to implementations including any of (1) at least one A, (2) at least one B, or (3) at least one A and at least one B.
As used herein, singular references (e.g., “a”, “an”, “first”, “second”, etc.) do not exclude a plurality. The term “a” or “an” object, as used herein, refers to one or more of that object. The terms “a” (or “an”), “one or more”, and “at least one” are used interchangeably herein. Furthermore, although individually listed, a plurality of means, elements, or actions may be implemented by, e.g., the same entity or object. Additionally, although individual features may be included in different examples or claims, these may possibly be combined, and the inclusion in different examples or claims does not imply that a combination of features is not feasible and/or advantageous.
At block 315, the optical map refinement circuitry 110 enters the operational phase 135 and begins processing each stereo pair of input images 240 corresponding to frames of the input stereo video. At block 320, the captured optical flow estimation circuitry 210 of the optical map refinement circuitry 110 accesses, as described above, a given stereo pair of input images 240 corresponding to a given frame time of the input stereo video. At block 325, the captured optical flow estimation circuitry 210 generates, as described above, the input (or initial) optical flow maps 245 for the stereo input image pair 240 accessed at block 320. At block 330, the alpha matte estimation circuitry 215 of the optical map refinement circuitry 110 generates, as described above, the alpha mattes 250 for the stereo input image pair 240 accessed at block 320 using the stereo background image pair 230 accessed at block 305.
At block 335, the optical flow merge circuitry 220 of the optical map refinement circuitry 110 combines, or merges, the input (or initial) optical flow maps 245 for the given stereo input image pair 240 with the background optical flow maps 235 for the stereo background image pair 230 based on the alpha mattes 250 to generate the output (or refined) optical flow maps 255 for the given stereo input image pair 240, as described above. Example machine readable instructions and/or example operations that may be executed and/or instantiated by processor circuitry to implement the processing at block 335 are illustrated in
At block 340, the optical flow merge circuitry 220 outputs, as described above, the generated output (or refined) optical flow maps 255 for the given stereo input image pair 240 to a video processor and/or video application, such as the video processor 105. At block 345, the optical map refinement circuitry 110 continues processing successive stereo input image pairs 240 of the input stereo video according to blocks 320-340, as described above. When optical flow map generation for the stereo input image pairs 240 of the input stereo video is finished, the machine readable instructions and/or the operations 300 of
At block 430, the optical flow merge circuitry 220 continues iterating over the flow map index until the elements of the output optical flow map 255 have been generated according to blocks 415-425. At block 435, the optical flow merge circuitry 220 continues iterating over the camera fields-of-view (or the stereo image dimensions) until output optical flow maps 255 have been generated for the different camera fields-of-view (or the stereo image dimensions). The machine readable instructions and/or the operations 335 of
The programmable circuitry platform 500 of the illustrated example includes programmable circuitry 512. The programmable circuitry 512 of the illustrated example is hardware. For example, the programmable circuitry 512 can be implemented by one or more integrated circuits, logic circuits, FPGAs, microprocessors, CPUs, GPUs, DSPs, and/or microcontrollers from any desired family or manufacturer. The programmable circuitry 512 may be implemented by one or more semiconductor based (e.g., silicon based) devices. In this example, the programmable circuitry 512 implements the example background optical flow estimation circuitry 205, the example captured optical flow estimation circuitry 210, the example alpha matte estimation circuitry 215 and/or the example optical flow merge circuitry 220.
The programmable circuitry 512 of the illustrated example includes a local memory 513 (e.g., a cache, registers, etc.). The programmable circuitry 512 of the illustrated example is in communication with main memory 514, 516, which includes a volatile memory 514 and a non-volatile memory 516, by a bus 518. The volatile memory 514 may be implemented by Synchronous Dynamic Random Access Memory (SDRAM), Dynamic Random Access Memory (DRAM), RAMBUS® Dynamic Random Access Memory (RDRAM®), and/or any other type of RAM device. The non-volatile memory 516 may be implemented by flash memory and/or any other desired type of memory device. Access to the main memory 514, 516 of the illustrated example is controlled by a memory controller 517. In some examples, the memory controller 517 may be implemented by one or more integrated circuits, logic circuits, microcontrollers from any desired family or manufacturer, or any other type of circuitry to manage the flow of data going to and from the main memory 514, 516.
The programmable circuitry platform 500 of the illustrated example also includes interface circuitry 520. The interface circuitry 520 may be implemented by hardware in accordance with any type of interface standard, such as an Ethernet interface, a universal serial bus (USB) interface, a Bluetooth® interface, a near field communication (NFC) interface, a Peripheral Component Interconnect (PCI) interface, and/or a Peripheral Component Interconnect Express (PCIe) interface.
In the illustrated example, one or more input devices 522 are connected to the interface circuitry 520. The input device(s) 522 permit(s) a user (e.g., a human user, a machine user, etc.) to enter data and/or commands into the programmable circuitry 512. The input device(s) 522 can be implemented by, for example, an audio sensor, a microphone, a camera (still or video), a keyboard, a button, a mouse, a touchscreen, a trackpad, a trackball, an isopoint device, and/or a voice recognition system.
One or more output devices 524 are also connected to the interface circuitry 520 of the illustrated example. The output device(s) 524 can be implemented, for example, by display devices (e.g., a light emitting diode (LED), an organic light emitting diode (OLED), a liquid crystal display (LCD), a cathode ray tube (CRT) display, an in-place switching (IPS) display, a touchscreen, etc.), a tactile output device, a printer, and/or speaker. The interface circuitry 520 of the illustrated example, thus, typically includes a graphics driver card, a graphics driver chip, and/or graphics processor circuitry such as a GPU.
The interface circuitry 520 of the illustrated example also includes a communication device such as a transmitter, a receiver, a transceiver, a modem, a residential gateway, a wireless access point, and/or a network interface to facilitate exchange of data with external machines (e.g., computing devices of any kind) by a network 526. The communication can be by, for example, an Ethernet connection, a digital subscriber line (DSL) connection, a telephone line connection, a coaxial cable system, a satellite system, a beyond-line-of-site wireless system, a line-of-site wireless system, a cellular telephone system, an optical connection, etc.
The programmable circuitry platform 500 of the illustrated example also includes one or more mass storage discs or devices 528 to store firmware, software, and/or data. Examples of such mass storage discs or devices 528 include magnetic storage devices (e.g., floppy disk, drives, HDDs, etc.), optical storage devices (e.g., Blu-ray disks, CDs, DVDs, etc.), RAID systems, and/or solid-state storage discs or devices such as flash memory devices and/or SSDs.
The machine readable instructions 532, which may be implemented by the machine readable instructions of
The cores 602 may communicate by a first example bus 604. In some examples, the first bus 604 may be implemented by a communication bus to effectuate communication associated with one(s) of the cores 602. For example, the first bus 604 may be implemented by at least one of an Inter-Integrated Circuit (I2C) bus, a Serial Peripheral Interface (SPI) bus, a PCI bus, or a PCIe bus. Additionally or alternatively, the first bus 604 may be implemented by any other type of computing or electrical bus. The cores 602 may obtain data, instructions, and/or signals from one or more external devices by example interface circuitry 606. The cores 602 may output data, instructions, and/or signals to the one or more external devices by the interface circuitry 606. Although the cores 602 of this example include example local memory 620 (e.g., Level 1 (L1) cache that may be split into an L1 data cache and an L1 instruction cache), the microprocessor 600 also includes example shared memory 610 that may be shared by the cores (e.g., Level 2 (L2 cache)) for high-speed access to data and/or instructions. Data and/or instructions may be transferred (e.g., shared) by writing to and/or reading from the shared memory 610. The local memory 620 of each of the cores 602 and the shared memory 610 may be part of a hierarchy of storage devices including multiple levels of cache memory and the main memory (e.g., the main memory 514, 516 of
Each core 602 may be referred to as a CPU, DSP, GPU, etc., or any other type of hardware circuitry. Each core 602 includes control unit circuitry 614, arithmetic and logic (AL) circuitry (sometimes referred to as an ALU) 616, a plurality of registers 618, the local memory 620, and a second example bus 622. Other structures may be present. For example, each core 602 may include vector unit circuitry, single instruction multiple data (SIMD) unit circuitry, load/store unit (LSU) circuitry, branch/jump unit circuitry, floating-point unit (FPU) circuitry, etc. The control unit circuitry 614 includes semiconductor-based circuits structured to control (e.g., coordinate) data movement within the corresponding core 602. The AL circuitry 616 includes semiconductor-based circuits structured to perform one or more mathematic and/or logic operations on the data within the corresponding core 602. The AL circuitry 616 of some examples performs integer based operations. In other examples, the AL circuitry 616 also performs floating-point operations. In yet other examples, the AL circuitry 616 may include first AL circuitry that performs integer-based operations and second AL circuitry that performs floating-point operations. In some examples, the AL circuitry 616 may be referred to as an Arithmetic Logic Unit (ALU).
The registers 618 are semiconductor-based structures to store data and/or instructions such as results of one or more of the operations performed by the AL circuitry 616 of the corresponding core 602. For example, the registers 618 may include vector register(s), SIMD register(s), general-purpose register(s), flag register(s), segment register(s), machine-specific register(s), instruction pointer register(s), control register(s), debug register(s), memory management register(s), machine check register(s), etc. The registers 618 may be arranged in a bank as shown in
Each core 602 and/or, more generally, the microprocessor 600 may include additional and/or alternate structures to those shown and described above. For example, one or more clock circuits, one or more power supplies, one or more power gates, one or more cache home agents (CHAs), one or more converged/common mesh stops (CMSs), one or more shifters (e.g., barrel shifter(s)) and/or other circuitry may be present. The microprocessor 600 is a semiconductor device fabricated to include many transistors interconnected to implement the structures described above in one or more integrated circuits (ICs) contained in one or more packages.
The microprocessor 600 may include and/or cooperate with one or more accelerators (e.g., acceleration circuitry, hardware accelerators, etc.). In some examples, accelerators are implemented by logic circuitry to perform certain tasks more quickly and/or efficiently than can be done by a general-purpose processor. Examples of accelerators include ASICs and FPGAs such as those discussed herein. A GPU, DSP and/or other programmable device can also be an accelerator. Accelerators may be on-board the microprocessor 600, in the same chip package as the microprocessor 600 and/or in one or more separate packages from the microprocessor 600.
More specifically, in contrast to the microprocessor 600 of
In the example of
In some examples, the binary file is compiled, generated, transformed, and/or otherwise output from a uniform software platform utilized to program FPGAs. For example, the uniform software platform may translate first instructions (e.g., code or a program) that correspond to one or more operations/functions in a high-level language (e.g., C, C++, Python, etc.) into second instructions that correspond to the one or more operations/functions in an HDL. In some such examples, the binary file is compiled, generated, and/or otherwise output from the uniform software platform based on the second instructions. In some examples, the FPGA circuitry 700 of
The FPGA circuitry 700 of
The FPGA circuitry 700 also includes an array of example logic gate circuitry 708, a plurality of example configurable interconnections 710, and example storage circuitry 712. The logic gate circuitry 708 and the configurable interconnections 710 are configurable to instantiate one or more operations/functions that may correspond to at least some of the machine readable instructions of
The configurable interconnections 710 of the illustrated example are conductive pathways, traces, vias, or the like that may include electrically controllable switches (e.g., transistors) whose state can be changed by programming (e.g., using an HDL instruction language) to activate or deactivate one or more connections between one or more of the logic gate circuitry 708 to program desired logic circuits.
The storage circuitry 712 of the illustrated example is structured to store result(s) of the one or more of the operations performed by corresponding logic gates. The storage circuitry 712 may be implemented by registers or the like. In the illustrated example, the storage circuitry 712 is distributed amongst the logic gate circuitry 708 to facilitate access and increase execution speed.
The example FPGA circuitry 700 of
Although
It should be understood that some or all of the circuitry of
In some examples, some or all of the circuitry of
In some examples, the programmable circuitry 512 of
A block diagram illustrating an example software distribution platform 805 to distribute software such as the example machine readable instructions 532 of
From the foregoing, it will be appreciated that example systems, methods, apparatus, and articles of manufacture have been disclosed that generate optical flow maps based on foreground and background image segmentation. Disclosed examples provide an optical flow estimation framework that generates refined optical flow maps based on background matting, but that are not dependent on a particular technique for background matting or optical flow estimation. The resulting, refined optical flow maps output from the disclosed examples can produce view interpolation results with fewer artifacts than prior optical flow estimation techniques. Moreover, the additional computation cost for the optical flow map refinement pipeline is minimal while the accuracy of the resulting optical flow map can be a substantial improvement over prior optical flow estimation techniques. Disclosed systems, methods, apparatus, and articles of manufacture are accordingly directed to one or more improvement(s) in the operation of a machine such as a computer or other electronic and/or mechanical device.
Further examples and combinations thereof include the following:
Example 1 includes an apparatus to generate optical flow maps, the apparatus comprising interface circuitry, machine readable instructions, and programmable circuitry to execute the machine readable instructions to at least generate a reference optical flow map based on a stereo pair of background images corresponding to a first camera field-of-view and a second camera field-of-view, generate a first optical flow map based on a stereo pair of input images corresponding to the first camera field-of-view and the second camera field-of-view, combine the reference optical flow map and the first optical flow map based on an alpha matte to generate a second optical flow map associated with the stereo pair of input images, the alpha matte representative of segmentation of at least one of the stereo pair of input images into foreground and background regions, and output the second optical flow map to at least one of a video processor or a video application.
Example 2 includes the apparatus of example 1, wherein the stereo pair of input images is a first stereo pair of a sequence of stereo pairs of input images, the alpha matte is a first alpha matte, and the programmable circuitry is to generate a third optical flow map based on a second stereo pair of input images in the sequence of stereo pairs of input images, combine the reference optical flow map and the third optical flow map based on a second alpha matte to generate a fourth optical flow map associated with the second stereo pair of input images, the second alpha matte representative of segmentation of at least one of the second stereo pair of input images into foreground and background regions, and output the fourth optical flow map to the at least one of the video processor or the video application.
Example 3 includes the apparatus of example 1 or example 2, wherein the reference optical flow map includes elements corresponding respectively to pixels of a first one of the stereo pair of background images, ones of the elements of the reference optical flow map including respective displacement vectors to map ones of the pixels of the first one of the stereo pair of background images to corresponding pixels of the second one of the stereo pair of background images, and the first optical flow map includes elements corresponding respectively to pixels of a first one of the stereo pair of input images, ones of the elements of the first optical flow map including respective displacement vectors to map ones of the pixels of the first one of the stereo pair of input images to corresponding pixels of the second one of the stereo pair of input images.
Example 4 includes the apparatus of example 3, wherein the reference optical flow map is a first reference optical flow map, the alpha matte is a first alpha matte representative of segmentation of the first one of the stereo pair of input images into foreground and background regions, and the programmable circuitry is to generate a second reference optical flow map based on the stereo pair of background images, the second reference optical flow map including elements corresponding respectively to the pixels of the second one of the stereo pair of background images, ones of the elements of the second reference optical flow map including respective displacement vectors to map ones of the pixels of the second one of the stereo pair of background images to the corresponding pixels of the first one of the stereo pair of background images, generate a third optical flow map based on the stereo pair of input images, the third optical flow map including elements corresponding respectively to the pixels of the second one of the stereo pair of input images, ones of the elements of the third optical flow map including respective displacement vectors to map ones of the pixels of the second one of the stereo pair of input images to the corresponding pixels of the first one of the stereo pair of input images, combine the second reference optical flow map and the third optical flow map based on a second alpha matte to generate a fourth optical flow map associated with the stereo pair of input images, the second alpha matte representative of segmentation of the second one of the stereo pair of input images into foreground and background regions, and output the fourth optical flow map to the at least one of the video processor or the video application.
Example 5 includes the apparatus of example 3, wherein the alpha matte includes elements corresponding respectively to the pixels of the first one of the stereo pair of input images, the elements of the alpha matte representative of segmentation of the corresponding pixels of the first one of the stereo pair of input images into the foreground and background regions, and to combine the reference optical flow map and the first optical flow map, the programmable circuitry is to combine one of the elements of the reference optical flow map with a corresponding one of the elements of the first optical flow map based on a corresponding one of the elements of the alpha matte.
Example 6 includes the apparatus of example 5, wherein the corresponding one of the elements of the alpha matte has a value in a range from zero to one, and to combine the one of the elements of the reference optical flow map with the corresponding one of the elements of the first optical flow map, the programmable circuitry is to multiply a value of the one of the elements of the reference optical flow map by the value of the corresponding one of the elements of the alpha matte subtracted from one to determine a first quantity, multiply a value of the corresponding one of the elements of the first optical flow map with the value of the corresponding one of the elements of the alpha matte to determine a second quantity, and add the first quantity and the second quantity to determine a value of a corresponding one of a plurality of elements included in the second optical flow map.
Example 7 includes the apparatus of example 1, wherein to combine the reference optical flow map and the first optical flow map, the programmable circuitry is to combine respective elements of the reference optical flow map with corresponding elements of the first optical flow map based on corresponding elements of the alpha matte to determine corresponding elements of the second optical flow map.
Example 8 includes at least one non-transitory computer readable medium comprising computer readable instructions to cause programmable circuitry to at least generate a reference optical flow map based on a stereo pair of background images corresponding to a first camera field-of-view and a second camera field-of-view, generate a first optical flow map based on a stereo pair of input images corresponding to the first camera field-of-view and the second camera field-of-view, merge the reference optical flow map and the first optical flow map based on an alpha matte to generate a second optical flow map associated with the stereo pair of input images, the alpha matte representative of segmentation of at least one of the stereo pair of input images into foreground and background regions, and output the second optical flow map to at least one of a video processor or a video application.
Example 9 includes the at least one non-transitory computer readable medium of example 8, wherein the stereo pair of input images is a first stereo pair of a sequence of stereo pairs of input images, the alpha matte is a first alpha matte, and the instructions are to cause the programmable circuitry to generate a third optical flow map based on a second stereo pair of input images in the sequence of stereo pairs of input images, merge the reference optical flow map and the third optical flow map based on a second alpha matte to generate a fourth optical flow map associated with the second stereo pair of input images, the second alpha matte representative of segmentation of at least one of the second stereo pair of input images into foreground and background regions, and output the fourth optical flow map to the at least one of the video processor or the video application.
Example 10 includes the at least one non-transitory computer readable medium of example 8 or example 9, wherein the reference optical flow map includes elements corresponding respectively to pixels of a first one of the stereo pair of background images, ones of the elements of the reference optical flow map including respective displacement vectors to map ones of the pixels of the first one of the stereo pair of background images to corresponding pixels of the second one of the stereo pair of background images, and the first optical flow map includes elements corresponding respectively to pixels of a first one of the stereo pair of input images, ones of the elements of the first optical flow map including respective displacement vectors to map ones of the pixels of the first one of the stereo pair of input images to corresponding pixels of the second one of the stereo pair of input images.
Example 11 includes the at least one non-transitory computer readable medium of example 10, wherein the reference optical flow map is a first reference optical flow map, the alpha matte is a first alpha matte representative of segmentation of the first one of the stereo pair of input images into foreground and background regions, and the instructions are to cause the programmable circuitry to generate a second reference optical flow map based on the stereo pair of background images, the second reference optical flow map including elements corresponding respectively to the pixels of the second one of the stereo pair of background images, ones of the elements of the second reference optical flow map including respective displacement vectors to map ones of the pixels of the second one of the stereo pair of background images to the corresponding pixels of the first one of the stereo pair of background images, generate a third optical flow map based on the stereo pair of input images, the third optical flow map including elements corresponding respectively to the pixels of the second one of the stereo pair of input images, ones of the elements of the third optical flow map including respective displacement vectors to map ones of the pixels of the second one of the stereo pair of input images to the corresponding pixels of the first one of the stereo pair of input images, merge the second reference optical flow map and the third optical flow map based on a second alpha matte to generate a fourth optical flow map associated with the stereo pair of input images, the second alpha matte representative of segmentation of the second one of the stereo pair of input images into foreground and background regions, and output the fourth optical flow map to the at least one of the video processor or the video application.
Example 12 includes the at least one non-transitory computer readable medium of example 10, wherein the alpha matte includes elements corresponding respectively to the pixels of the first one of the stereo pair of input images, the elements of the alpha matte representative of segmentation of the corresponding pixels of the first one of the stereo pair of input images into the foreground and background regions, and to merge the reference optical flow map and the first optical flow map, the instructions are to cause the programmable circuitry to merge one of the elements of the reference optical flow map with a corresponding one of the elements of the first optical flow map based on a corresponding one of the elements of the alpha matte.
Example 13 includes the at least one non-transitory computer readable medium of example 12, wherein the corresponding one of the elements of the alpha matte has a value in a range from zero to one, and to merge the one of the elements of the reference optical flow map with the corresponding one of the elements of the first optical flow map, the instructions are to cause the programmable circuitry to multiply a value of the one of the elements of the reference optical flow map by the value of the corresponding one of the elements of the alpha matte subtracted from one to determine a first quantity, multiply a value of the corresponding one of the elements of the first optical flow map with the value of the corresponding one of the elements of the alpha matte to determine a second quantity, and add the first quantity and the second quantity to determine a value of a corresponding one of a plurality of elements included in the second optical flow map.
Example 14 includes the at least one non-transitory computer readable medium of example 8, wherein to merge the reference optical flow map and the first optical flow map, the instructions are to cause the programmable circuitry to merge respective elements of the reference optical flow map with corresponding elements of the first optical flow map based on corresponding elements of the alpha matte to determine corresponding elements of the second optical flow map.
Example 15 includes a method to generate optical flow maps, the method comprising generating, by executing an instruction with programmable circuitry, a reference optical flow map based on a stereo pair of background images corresponding to a first camera field-of-view and a second camera field-of-view, generating, by executing an instruction with the programmable circuitry, a first optical flow map based on a stereo pair of input images corresponding to the first camera field-of-view and the second camera field-of-view, combining, by executing an instruction with the programmable circuitry, the reference optical flow map and the first optical flow map based on an alpha matte to generate a second optical flow map associated with the stereo pair of input images, the alpha matte representative of segmentation of at least one of the stereo pair of input images into foreground and background regions, and outputting the second optical flow map to at least one of a video processor or a video application.
Example 16 includes the method of example 15, wherein the reference optical flow map includes elements corresponding respectively to pixels of a first one of the stereo pair of background images, ones of the elements of the reference optical flow map including respective displacement vectors to map ones of the pixels of the first one of the stereo pair of background images to corresponding pixels of the second one of the stereo pair of background images, and the first optical flow map includes elements corresponding respectively to pixels of a first one of the stereo pair of input images, ones of the elements of the first optical flow map including respective displacement vectors to map ones of the pixels of the first one of the stereo pair of input images to corresponding pixels of the second one of the stereo pair of input images.
Example 17 includes the method of example 16, wherein the reference optical flow map is a first reference optical flow map, the alpha matte is a first alpha matte representative of segmentation of the first one of the stereo pair of input images into foreground and background regions, and the further including generating a second reference optical flow map based on the stereo pair of background images, the second reference optical flow map including elements corresponding respectively to the pixels of the second one of the stereo pair of background images, ones of the elements of the second reference optical flow map including respective displacement vectors to map ones of the pixels of the second one of the stereo pair of background images to the corresponding pixels of the first one of the stereo pair of background images, generating a third optical flow map based on the stereo pair of input images, the third optical flow map including elements corresponding respectively to the pixels of the second one of the stereo pair of input images, ones of the elements of the third optical flow map including respective displacement vectors to map ones of the pixels of the second one of the stereo pair of input images to the corresponding pixels of the first one of the stereo pair of input images, combining the second reference optical flow map and the third optical flow map based on a second alpha matte to generate a fourth optical flow map associated with the stereo pair of input images, the second alpha matte representative of segmentation of the second one of the stereo pair of input images into foreground and background regions, and outputting the fourth optical flow map to the at least one of the video processor or the video application.
Example 18 includes the method of example 16, wherein the alpha matte includes elements corresponding respectively to the pixels of the first one of the stereo pair of input images, the elements of the alpha matte representative of segmentation of the corresponding pixels of the first one of the stereo pair of input images into the foreground and background regions, and the combining of the reference optical flow map and the first optical flow map includes combining one of the elements of the reference optical flow map with a corresponding one of the elements of the first optical flow map based on a corresponding one of the elements of the alpha matte.
Example 19 includes the method of example 18, wherein the corresponding one of the elements of the alpha matte has a value in a range from zero to one, and the combining of the one of the elements of the reference optical flow map with the corresponding one of the elements of the first optical flow map includes multiply a value of the one of the elements of the reference optical flow map by the value of the corresponding one of the elements of the alpha matte subtracted from one to determine a first quantity, multiply a value of the corresponding one of the elements of the first optical flow map with the value of the corresponding one of the elements of the alpha matte to determine a second quantity, and add the first quantity and the second quantity to determine a value of a corresponding one of a plurality of elements included in the second optical flow map.
Example 20 includes the method of example 15, wherein the combining of the reference optical flow map and the first optical flow map including combining respective elements of the reference optical flow map with corresponding elements of the first optical flow map based on corresponding elements of the alpha matte to determine corresponding elements of the second optical flow map.
The following claims are hereby incorporated into this Detailed Description by this reference. Although certain example systems, apparatus, articles of manufacture, and methods have been disclosed herein, the scope of coverage of this patent is not limited thereto. On the contrary, this patent covers all systems, apparatus, articles of manufacture, and methods fairly falling within the scope of the claims of this patent.