Patent Application
20160080670
1. Field
Certain aspects of the present disclosure generally relate to neural system engineering and, more particularly, to systems and methods for event-based down sampling.
2. Background
Machine vision enables machines to see and perceive. In conventional systems, a sensor, such as a video camera, is used for machine vision. Video cameras may be referred to as frame-based vision sensors because the video camera periodically samples visual data from a two-dimensional array of photosensitive elements. The human retina, in contrast, is an example of an event-based vision sensor. That is, individual neurons in the retina are sensitive to a portion of the visual field such that each neuron sends a signal toward the brain when there is a change to the portion of the visual field. Unlike conventional frame-based video cameras, there is no periodic sampling of all of the neurons in the retina. Rather, visual data is transmitted by the retina whenever there are transient visual events.
Similar to the retina, dynamic vision sensors (DVSs) include an array of photosensitive elements that detect and transmit visual events. An individual photosensitive element of a dynamic vision sensor transmits a signal when there is a change in luminance at a portion of a visual scene. As an example, an event-driven object detection system may use a dynamic vision sensor to detect moving objects, such as faces or cars, and classify the detected objects in real time based on prior training.
In some cases, event-based sampling is specified to improve temporal sensitivity. That is, a frame-based sensor may be limited to sampling visual information based on the speed at which the frame-based sensor can read a frame of image data. In contrast, a photosensitive element within a dynamic vision sensor may sample visual information based on the speed at which the photosensitive element can detect changes in a portion of the visual field. In addition to improved temporal sensitivity, a photosensitive element may consume less power, on average, in comparison with a frame-based sensor because a photosensitive element remains inactive when there are no changes to the visual scene.
Still, the improved temporal sensitivity and lower power consumption have yet to be fully realized in conventional event-based vision systems. Specifically, the number of known processing techniques for event-based sensor outputs is less than the number of machine vision techniques for frame-based vision sensors. Common techniques developed for frame-based sensors include efficient down-sampling, sub-sampling, interpolation, fast Fourier transforms, and neural network based object classification.
In some cases, to improve techniques that have been developed for frame-based machine vision systems, the output of a dynamic vision sensor may be used to periodically reconstruct image frames. Furthermore, conventional image processing techniques may be applied to the resulting frames. However, the conversion to image frames may reduce the performance of an event-based vision system. Thus, it is desirable to convert frame-based techniques to an event-based system without reducing the performance of an event-based sensor.
In one aspect of the present disclosure, a method of event-based down sampling is disclosed. The method includes receiving multiple sensor events corresponding to addresses and time stamps. The method also includes spatially down sampling the addresses based on the time stamps and the addresses.
Another aspect of the present disclosure is directed to an apparatus including means for receiving multiple sensor events corresponding to addresses and time stamps. The apparatus also includes means for spatially down sampling the addresses based on the time stamps and the addresses.
In another aspect of the present disclosure, a non-transitory computer-readable medium with program code recorded thereon is disclosed. The program code for event-based down sampling is executed by a processor and includes program code to receive multiple sensor events corresponding to addresses and time stamps. The program code also includes program code to spatially down sample the addresses based on the time stamps and the addresses.
Another aspect of the present disclosure is directed to an apparatus for event-based down sampling having a memory and one or more processors coupled to the memory. The processor(s) is configured to receive multiple sensor events corresponding to addresses and time stamps. The processor(s) is also configured to spatially down sample the addresses based on the time stamps and the addresses.
Additional features and advantages of the disclosure will be described below. It should be appreciated by those skilled in the art that this disclosure may be readily utilized as a basis for modifying or designing other structures for carrying out the same purposes of the present disclosure. It should also be realized by those skilled in the art that such equivalent constructions do not depart from the teachings of the disclosure as set forth in the appended claims. The novel features, which are believed to be characteristic of the disclosure, both as to its organization and method of operation, together with further objects and advantages, will be better understood from the following description when considered in connection with the accompanying figures. It is to be expressly understood, however, that each of the figures is provided for the purpose of illustration and description only and is not intended as a definition of the limits of the present disclosure.
The features, nature, and advantages of the present disclosure will become more apparent from the detailed description set forth below when taken in conjunction with the drawings in which like reference characters identify correspondingly throughout.
The detailed description set forth below, in connection with the appended drawings, is intended as a description of various configurations and is not intended to represent the only configurations in which the concepts described herein may be practiced. The detailed description includes specific details for the purpose of providing a thorough understanding of the various concepts. However, it will be apparent to those skilled in the art that these concepts may be practiced without these specific details. In some instances, well-known structures and components are shown in block diagram form in order to avoid obscuring such concepts.
Based on the teachings, one skilled in the art should appreciate that the scope of the disclosure is intended to cover any aspect of the disclosure, whether implemented independently of or combined with any other aspect of the disclosure. For example, an apparatus may be implemented or a method may be practiced using any number of the aspects set forth. In addition, the scope of the disclosure is intended to cover such an apparatus or method practiced using other structure, functionality, or structure and functionality in addition to or other than the various aspects of the disclosure set forth. It should be understood that any aspect of the disclosure disclosed may be embodied by one or more elements of a claim.
The word “exemplary” is used herein to mean “serving as an example, instance, or illustration.” Any aspect described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other aspects.
Although particular aspects are described herein, many variations and permutations of these aspects fall within the scope of the disclosure. Although some benefits and advantages of the preferred aspects are mentioned, the scope of the disclosure is not intended to be limited to particular benefits, uses or objectives. Rather, aspects of the disclosure are intended to be broadly applicable to different technologies, system configurations, networks and protocols, some of which are illustrated by way of example in the figures and in the following description of the preferred aspects. The detailed description and drawings are merely illustrative of the disclosure rather than limiting, the scope of the disclosure being defined by the appended claims and equivalents thereof.
As previously discussed, event-driven object detection system may use a dynamic vision sensor (DVS) to detect moving objects, such as faces or cars, and classify the detected objects in real time based on prior training. The computations in the system may be triggered by sensor events. The event-driven object detection system may be referred to as the detection system.
Aspects of the present disclosure are directed to applying image processing techniques to the output of an event-based sensor. The image processing techniques may perform functions that are substantially similar to techniques applied to frame-based vision systems. More specifically, aspects of the present disclosure are directed to event-based processing techniques that improve event-driven object detection. In one configuration, the dynamic vision sensor outputs are down sampled and convolved with a re-usable kernel.
According to an aspect of the present disclosure, a detection system processes visual input when an event is generated. That is, the detection system does not perform processing when events are not output from the dynamic vision sensor. For example, a dynamic vision sensor may be part of a surveillance system and may be pointed at a hallway and a door. If there is no change in the scene, the dynamic vision sensor will not send any outputs, and consequently, the detection system will not perform any computations. The dynamic vision sensor may produce outputs, and the event-based detection system may perform computations, when there is a change in the scene. For example, a dynamic vision sensor focused on a doorway may produce outputs when a person walks through the door.
The processing load of an event-based system scales linearly with the sensor event rate. In contrast, the processing load of a frame-based system scales linearly with the frame rate. When the event rate is less than the frame rate, as in the surveillance system described above, the detection performance of an event-based system may improve as a function of power consumption. However, in scenarios in which the event rate exceeds the frame rate, for example, when the field of view of a dynamic vision sensor includes the rotating blades of an unmanned aerial vehicle, the power consumption of an event-based system may exceed the power consumption of a frame-based system that ignores the individual rotations of the rotating blades. When an increased number of sensor events do not produce a desired increase in system performance (e.g., object detection accuracy) it may be desirable to down sample the sensor events. According to aspects of the present disclosure, the sensor events may be down sampled spatially and/or temporally.
The SOC 100 may also include additional processing blocks tailored to specific functions, such as a GPU 104, a DSP 106, a connectivity block 110, which may include fourth generation long term evolution (4G LTE) connectivity, unlicensed Wi-Fi connectivity, USB connectivity, Bluetooth connectivity, and the like, and a multimedia processor 112 that may, for example, detect and recognize gestures. In one implementation, the NPU is implemented in the CPU, DSP, and/or GPU. The SOC 100 may also include a sensor processor 114, image signal processors (ISPs), and/or navigation 120, which may include a global positioning system.
The SOC 100 may be based on an ARM instruction set. In an aspect of the present disclosure, the instructions loaded into the general-purpose processor 102 may comprise code for receiving multiple sensor events corresponding to addresses and time stamps. The instructions loaded into the general-purpose processor 102 may also comprise code for spatially down sampling the addresses based on the time stamps and the addresses.
According to aspects of the present disclosure, the detection system comprises various components for processing data. As an example, as shown in
In one configuration, the dynamic vision sensor is a sensor that detects events. As previously discussed, the events are generated from a change in intensity received at a photosensor element. For example, the dynamic vision sensor may be a DVS128 sensor from iniLabs. The sensor array may have a size of N×N (N=128) of which each pixel is a level-crossing sampler of log-luminance in time. The temporal resolution of the pixel is on the order of 10 micro seconds. The output of the dynamic vision sensor may be a polarized, coordinate addressed event train {(tk; pk; μk; vk)}, where tk and pk are time stamps and polarities of events and (μk; vk) are the pixel coordinates of event k at time tk. Here, tkε, pkε{−1,1} and μk, vkε{1, . . . , 128}.
The pixel response functions may be defined as:
x
μ,v(t)=Σkpkδμ,μ
where (μ, v) ε{1, . . . , 128}2 index pixels, δ is the Kroenecker delta, and δ( ) is the Dirac delta function. The matrix may also be written as:
X(t)=[xμ,v](t) (2)
An event-driven short-time spatial DFT (eSTsDFT) component may receive an event train {(tk; pk; μk; vk)} as an input and output a real-time complex N×N (N=128) matrix. The event train may be received in one or more data packets. The short-time spatial DFT matrix, {tilde over (X)}(t), may be computed as:
{tilde over (X)}(t)=∫SNX(τ)SNTw(τ−t)dτ, (3)
where the N-th order DFT matrix (SN) may be computed as:
where
is the n-th root of unity and
is the (n+1)-th column Of SN.
Furthermore, w(−t)=θ(t)exp(−w0t) is an exponential short-time window function. The event-driven short-time spatial DFT module may compute the value of {tilde over (x)}(t) at each sensor event {tk}. In one configuration, the sensor events are first down sampled accordingly and the event-driven short-time spatial DFT module computes the values of the short term spatial DFT, {tilde over (x)}(t), upon reciept of each down sampled event.
The event-driven feature extraction (eFE) component may further reduce the dimensionality of the processed event stream, from the N×N×2 dimensions of the event-driven short-time spatial DFT module (N=128 if not spatially down sampled) to an L=64-dimensional feature vector (e.g., from 128 128 complex numbers to 64 real numbers). Specifically, the L features may be binned instantaneous spectral power of {tilde over (x)}(t),y(t)=φ({tilde over (X)}*(t){tilde over (X)}(t)) where * is the conjugate transpose and φ( ) is a log-linear transformation function.
{tilde over (X)}*{tilde over (X)} may be written as a 128×128 dimensional vector x and the linear function φ( ) may be expressed as a matrix multiplication followed by a logarithm y=log(φx), where
is a binary matrix of size 64×(128×128), which may contain two components of size 32×(128×128), corresponding to 32 radial and 32 angular power bins. These matrices are constant valued and computed priorly. For example, the matrices may be generated from training a machine learning algorithm on collected training data. Alternatively, the matrices may be user specified.
The event-driven feature extraction (eFE) component may compute the value of y(t) in an event-based manner. For example, y(t) may be computed at the end of every sensor event packet or group of packets that include multiple events.
The classification component, such as the event-driven support vector classification (eSVM) component, computes a time-varying class label function z(t) based on the real-time feature vector y(t) extracted by the eFE component, e.g., z(t)=ψ(y(t)), by virtue of a support vector machine with Gaussian radial basis function (RBF) as kernels.
The event-driven support vector classification component may compute the value of z(t) at the end of every sensor event packet or group of packets that occurs at least Δtmin since the last classification. As described below, TABLE 1 specifies the mathematical description of the input/output objects to and from the components of the system. In the present application, the components may be referred to as modules. Although the update schedule in TABLE 1 indicates that the event-driven short-time spatial DFT module may update on every event, the present disclosure is not so limiting. According to aspects of the present disclosure, the event-driven short-time spatial DFT module may update after every down sampled event, or after receipt of multiple down sampled events.
TABLES 2 and 3 provide the constant and state variables used for the event-driven short-time spatial DFT. TABLE 4 is pseudo-code for a single iteration of the detection system during which an event packet of length K is processed from the dynamic vision sensor.
Conventional image processing techniques down sample an image and convolve the down sampled image with a kernel. The kernel may be referred to as a convolution matrix or a mask. Still, in conventional systems, down sampling techniques are not specified for event-based sensors. Therefore, it is desirable to specify event-based sampling techniques, such as down sampling and/or cascaded down sampling, for the output of an event-based sensor. Furthermore, it is desirable to convolve the down sampled output with a kernel. The kernel may be re-used to improve performance.
For convolution, a value of a pixel in an image is calculated by multiplying each kernel value by the corresponding image pixel value. The convolution process may reduce performance if a large image is to be processed. That is, because each pixel in the image is convolved, it is desirable to down sample an image prior to convolving the image with a kernel.
In one configuration, the event-based image is scaled by a power of two in each x and y direction. Of course, the scaling is in not limited to a power of two and events may be scaled by any desirable factor. In one configuration, to scale the image, an event from an activated sensor, such as a photosensor is selected, and the least significant address bit from the row and column address is dropped. The event may be specified by (tk; pk; μk; vk)) where tk and pk are time stamps and polarities of events and (μk; vk) are the pixel coordinates of event k at time tk.
Additionally, in one configuration, if a photosensitive element generates multiple events within a pre-determined time span, one or more of the events are selected for output. In this configuration, events of the resulting temporally down sampled event stream are assigned timestamps based on the time of the events. Thus, in one example, if there are two or more events from one photosensitive element within a pre-determined time span, the first event or the last event may be selected as the output. Furthermore, the timestamp assigned to the first event may be the original timestamp of the event. In another example, a new timestamp may be assigned to the output event corresponding to an average of the two or more timestamps that are received within the pre-determined time span.
The temporal down sampling may be applied to an event data packet that includes multiple events or multiple data packets. If multiple pixel events are generated in the same event packet, or within a group of packets within the pre-determined timespan, the timestamp assigned to the down sampled events may be the latest timestamp of the multiple pixel events, the earliest timestamp of the multiple pixel events, and/or an average timestamp of the multiple pixel events.
The time span of temporal down sampling may be referenced from the receipt of a first event packet. According to aspects of the present disclosure, the down sampling and subsequent processing steps may be triggered on the receipt of a data packet, or after a pre-determined time period after the receipt of a data packet.
Additionally, spatial and temporal down sampling may be simultaneously applied, such that output events have a lower spatial specificity in comparison to input events. Furthermore, the output events may be assigned modified timestamps based on the input events.
According to an aspect of the present disclosure, pixel values associated with a sensor or group of sensors are updated based on the down sampling. For example, two or more events, such as positive polarity events, may be received from one sensor and one or more of the events may be dropped based on temporal down sampling. Thus, in this example, updating may be performed by increasing a pixel value associated with the remaining events by a factor of two to offset the discarded luminance information.
In some aspects, the event-based down sampling may be applied to event-based-sensors in other modalities. For example, the event-based down sampling method may be applied to outputs of an artificial cochlea representing auditory stimuli, or may be applied to the outputs of touch sensitive elements of a trackpad.
If the multiple sensor events were generated within a predetermined time of each other, at block 708, the system spatially down samples the addresses based on the time stamps and addresses. Additionally, at block 710, the pixel values corresponding to the addresses are convolved.
In one configuration, a model, such as an event based model or a machine learning model, is configured for receiving multiple sensor events corresponding to addresses and time stamps and spatially down sampling the addresses based on the time stamps and the addresses. The model includes a receiving means and/or spatially down sampling means. In one aspect, the receiving means and/or spatially down sampling means may be the general-purpose processor 102, program memory associated with the general-purpose processor 102, memory block 118, local processing units 202, and or the routing connection processing units 216 configured to perform the functions recited. In another configuration, the aforementioned means may be any module or any apparatus configured to perform the functions recited by the aforementioned means.
According to certain aspects of the present disclosure, each local processing unit 202 may be configured to determine parameters of the model based upon desired one or more functional features of the model, and develop the one or more functional features towards the desired functional features as the determined parameters are further adapted, tuned and updated.
The various operations of methods described above may be performed by any suitable means capable of performing the corresponding functions. The means may include various hardware and/or software component(s) and/or module(s), including, but not limited to, a circuit, an application specific integrated circuit (ASIC), or processor. Generally, where there are operations illustrated in the figures, those operations may have corresponding counterpart means-plus-function components with similar numbering.
As used herein, the term “determining” encompasses a wide variety of actions. For example, “determining” may include calculating, computing, processing, deriving, investigating, looking up (e.g., looking up in a table, a database or another data structure), ascertaining and the like. Additionally, “determining” may include receiving (e.g., receiving information), accessing (e.g., accessing data in a memory) and the like. Furthermore, “determining” may include resolving, selecting, choosing, establishing and the like.
As used herein, a phrase referring to “at least one of” a list of items refers to any combination of those items, including single members. As an example, “at least one of: a, b, or c” is intended to cover: a, b, c, a-b, a-c, b-c, and a-b-c.
The various illustrative logical blocks, modules and circuits described in connection with the present disclosure may be implemented or performed with a general-purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array signal (FPGA) or other programmable logic device (PLD), discrete gate or transistor logic, discrete hardware components or any combination thereof designed to perform the functions described herein. A general-purpose processor may be a microprocessor, but in the alternative, the processor may be any commercially available processor, controller, microcontroller or state machine. A processor may also be implemented as a combination of computing devices, e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration.
The steps of a method or algorithm described in connection with the present disclosure may be embodied directly in hardware, in a software module executed by a processor, or in a combination of the two. A software module may reside in any form of storage medium that is known in the art. Some examples of storage media that may be used include random access memory (RAM), read only memory (ROM), flash memory, erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), registers, a hard disk, a removable disk, a CD-ROM and so forth. A software module may comprise a single instruction, or many instructions, and may be distributed over several different code segments, among different programs, and across multiple storage media. A storage medium may be coupled to a processor such that the processor can read information from, and write information to, the storage medium. In the alternative, the storage medium may be integral to the processor.
The methods disclosed herein comprise one or more steps or actions for achieving the described method. The method steps and/or actions may be interchanged with one another without departing from the scope of the claims. In other words, unless a specific order of steps or actions is specified, the order and/or use of specific steps and/or actions may be modified without departing from the scope of the claims.
The functions described may be implemented in hardware, software, firmware, or any combination thereof. If implemented in hardware, an example hardware configuration may comprise a processing system in a device. The processing system may be implemented with a bus architecture. The bus may include any number of interconnecting buses and bridges depending on the specific application of the processing system and the overall design constraints. The bus may link together various circuits including a processor, machine-readable media, and a bus interface. The bus interface may be used to connect a network adapter, among other things, to the processing system via the bus. The network adapter may be used to implement signal processing functions. For certain aspects, a user interface (e.g., keypad, display, mouse, joystick, etc.) may also be connected to the bus. The bus may also link various other circuits such as timing sources, peripherals, voltage regulators, power management circuits, and the like, which are well known in the art, and therefore, will not be described any further.
The processor may be responsible for managing the bus and general processing, including the execution of software stored on the machine-readable media. The processor may be implemented with one or more general-purpose and/or special-purpose processors. Examples include microprocessors, microcontrollers, DSP processors, and other circuitry that can execute software. Software shall be construed broadly to mean instructions, data, or any combination thereof, whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise. Machine-readable media may include, by way of example, random access memory (RAM), flash memory, read only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable Read-only memory (EEPROM), registers, magnetic disks, optical disks, hard drives, or any other suitable storage medium, or any combination thereof. The machine-readable media may be embodied in a computer-program product. The computer-program product may comprise packaging materials.
In a hardware implementation, the machine-readable media may be part of the processing system separate from the processor. However, as those skilled in the art will readily appreciate, the machine-readable media, or any portion thereof, may be external to the processing system. By way of example, the machine-readable media may include a transmission line, a carrier wave modulated by data, and/or a computer product separate from the device, all which may be accessed by the processor through the bus interface. Alternatively, or in addition, the machine-readable media, or any portion thereof, may be integrated into the processor, such as the case may be with cache and/or general register files. Although the various components discussed may be described as having a specific location, such as a local component, they may also be configured in various ways, such as certain components being configured as part of a distributed computing system.
The processing system may be configured as a general-purpose processing system with one or more microprocessors providing the processor functionality and external memory providing at least a portion of the machine-readable media, all linked together with other supporting circuitry through an external bus architecture. Alternatively, the processing system may comprise one or more neuromorphic processors for implementing the neuron models and models of neural systems described herein. As another alternative, the processing system may be implemented with an application specific integrated circuit (ASIC) with the processor, the bus interface, the user interface, supporting circuitry, and at least a portion of the machine-readable media integrated into a single chip, or with one or more field programmable gate arrays (FPGAs), programmable logic devices (PLDs), controllers, state machines, gated logic, discrete hardware components, or any other suitable circuitry, or any combination of circuits that can perform the various functionality described throughout this disclosure. Those skilled in the art will recognize how best to implement the described functionality for the processing system depending on the particular application and the overall design constraints imposed on the overall system.
The machine-readable media may comprise a number of software modules. The software modules include instructions that, when executed by the processor, cause the processing system to perform various functions. The software modules may include a transmission module and a receiving module. Each software module may reside in a single storage device or be distributed across multiple storage devices. By way of example, a software module may be loaded into RAM from a hard drive when a triggering event occurs. During execution of the software module, the processor may load some of the instructions into cache to increase access speed. One or more cache lines may then be loaded into a general register file for execution by the processor. When referring to the functionality of a software module below, it will be understood that such functionality is implemented by the processor when executing instructions from that software module.
If implemented in software, the functions may be stored or transmitted over as one or more instructions or code on a computer-readable medium. Computer-readable media include both computer storage media and communication media including any medium that facilitates transfer of a computer program from one place to another. A storage medium may be any available medium that can be accessed by a computer. By way of example, and not limitation, such computer-readable media can comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. Also, any connection is properly termed a computer-readable medium. For example, if the software is transmitted from a website, server, or other remote source using a coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared (IR), radio, and microwave, then the coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of medium. Disk and disc, as used herein, include compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk, and Blu-ray® disc where disks usually reproduce data magnetically, while discs reproduce data optically with lasers. Thus, in some aspects computer-readable media may comprise non-transitory computer-readable media (e.g., tangible media). In addition, for other aspects computer-readable media may comprise transitory computer-readable media (e.g., a signal). Combinations of the above should also be included within the scope of computer-readable media.
Thus, certain aspects may comprise a computer program product for performing the operations presented herein. For example, such a computer program product may comprise a computer-readable medium having instructions stored (and/or encoded) thereon, the instructions being executable by one or more processors to perform the operations described herein. For certain aspects, the computer program product may include packaging material.
Further, it should be appreciated that modules and/or other appropriate means for performing the methods and techniques described herein can be downloaded and/or otherwise obtained by a user terminal and/or base station as applicable. For example, such a device can be coupled to a server to facilitate the transfer of means for performing the methods described herein. Alternatively, various methods described herein can be provided via storage means (e.g., RAM, ROM, a physical storage medium such as a compact disc (CD) or floppy disk, etc.), such that a user terminal and/or base station can obtain the various methods upon coupling or providing the storage means to the device. Moreover, any other suitable technique for providing the methods and techniques described herein to a device can be utilized.
It is to be understood that the claims are not limited to the precise configuration and components illustrated above. Various modifications, changes and variations may be made in the arrangement, operation and details of the methods and apparatus described above without departing from the scope of the claims.
This application claims the benefit under 35 U.S.C. §119(e) to U.S. Provisional Patent Application No. 62/051,131, entitled “EVENT BASED DOWN SAMPLING,” filed on Sep. 16, 2014, the disclosure of which is expressly incorporated herein by reference in its entirety
| Number | Date | Country | |
|---|---|---|---|
| 62051131 | Sep 2014 | US |
