Gesture recognition using multiple antenna

Abstract

Various embodiments wirelessly detect micro gestures using multiple antenna of a gesture sensor device. At times, the gesture sensor device transmits multiple outgoing radio frequency (RF) signals, each outgoing RF signal transmitted via a respective antenna of the gesture sensor device. The outgoing RF signals are configured to help capture information that can be used to identify micro-gestures performed by a hand. The gesture sensor device captures incoming RF signals generated by the outgoing RF signals reflecting off of the hand, and then analyzes the incoming RF signals to identify the micro-gesture.

Description

BACKGROUND

This background description is provided for the purpose of generally presenting the context of the disclosure. Unless otherwise indicated herein, material described in this section is neither expressly nor impliedly admitted to be prior art to the present disclosure or the appended claims.

As computing devices evolve with more computing power, they are able to evolve how they receive input commands or information. One type of evolving input mechanism relates to capturing user gestures. For instance, a user can attach a first peripheral device to their arm or hand that reads muscle activity, or hold a second peripheral device that contains an accelerometer that detects motion. In turn, these peripherals then communicate with a receiving computing device based upon a detected gesture. With these types of peripheral devices, a user physically connects the peripheral device to a corresponding body part that performs the gesture. However, this constrains the user, in that the user must not only acquire these peripheral devices, but must couple them to the receiving computing device. Thus, it would be advantageous to capture various gestures without attaching a peripheral device to the user.

SUMMARY

This Summary is provided to introduce a selection of concepts in a simplified form that are further described below in the Detailed Description. This Summary is not intended to identify key features or essential features of the claimed subject matter.

Various embodiments wirelessly detect micro gestures using multiple antennas of a gesture sensor device. At times, the gesture sensor device transmits multiple outgoing radio frequency (RF) signals, each outgoing RF signal transmitted via a respective antenna of the gesture sensor device. The outgoing RF signals are configured to help capture information that can be used to identify micro-gestures performed by a hand. The gesture sensor device captures incoming RF signals generated by the outgoing RF signals reflecting off of the hand, and then analyzes the incoming RF signals to identify the micro-gesture.

One or more embodiments provide a device configured to identify a micro-gesture associated with a target object, the device comprising: at least two antennas to respectively transmit a plurality of outgoing radio frequency (RF) signals, each antenna configured to: transmit a respective outgoing radio frequency (RF) signal of the plurality of outgoing RF signals; and receive an incoming RF signal generated by at least one transmitted outgoing RF signal of the plurality of outgoing RF signals reflecting off the target object; a digital signal processing component configured to: process a first set of data originating from incoming RF signals to extract information about the target object; and a machine-learning component configured to: receive the information extracted by the digital signal processing component; and process the information to identify the micro-gesture.

At least one embodiment provides a method for identifying a micro-gesture performed by a hand, the method comprising: transmitting a plurality of outgoing RF signals, each outgoing RF signal transmitted on a respective antenna of a plurality of antennas; capturing at least two incoming RF signals generated by at least two outgoing RF signals of the plurality of outgoing RF signals reflecting off the hand; and processing the at least two captured incoming RF signals to identify the micro-gesture performed by the hand.

At least one embodiment provides a device for detecting a micro-gesture performed by a hand, the device comprising: a gesture sensor component comprising: at least two antenna, each respective antenna associated with a respective transceiver path; at least one processor; one or more computer-readable storage devices; and one or more Application Programming Interfaces (APIs) stored on the one or more computer-readable storage devices which, responsive to execution by the at least one processor, configure the gesture sensor component to detect the micro-gesture by causing the gesture sensor component to perform operations comprising: transmitting a plurality of outgoing radio frequency (RF) signals, each outgoing radio frequency (RF) signal being transmitted on a respective antenna of the at least two antenna; receiving at least two incoming RF signals originating from at least two outgoing RF signals of the plurality of outgoing RF signals reflecting off the hand; and processing the at least two RF signals to detect the micro-gesture by processing data from at least two respective transceiver paths.

At least one embodiment provides a device configured to identify a micro-gesture associated with a target object, the device comprising: means for transmitting each respective outgoing radio frequency (RF) signal of a plurality of outgoing RF signals on a respective antenna; means for receiving incoming RF signals generated by at least some of the plurality of outgoing RF signals reflecting off the target object; means for processing a first set of data originating from incoming RF signals to extract information about the target object; and means for processing the information to identify the micro-gesture.

BRIEF DESCRIPTION OF THE DRAWINGS

Various aspects of wireless micro-gesture detection are described with reference to the following drawings. The same numbers are used throughout the drawings to reference like features and components:

FIG. 1 illustrates an example environment that employs wireless micro-gesture detection in accordance with one or more embodiments;

FIG. 2 illustrates an example implementation of a computing device of FIG. 1 in greater detail in accordance with one or more embodiments;

FIG. 3 illustrates an example of general signal properties;

FIG. 4 illustrates an example of general signal properties;

FIG. 5 illustrates an example of a pipeline in accordance with one or more embodiments;

FIG. 6 illustrates an example flow diagram in accordance with one or more embodiments;

FIG. 7 illustrates an example device in which micro-gesture hand detection can be employed in accordance with one or more embodiments.

DETAILED DESCRIPTION

Overview

Various embodiments wirelessly detect or recognize micro-gestures, e.g., hand gestures, using multiple antennas of a gesture sensor device. These micro-gestures can be detected in free-space, without any attachment or peripheral device connected to a corresponding body part, such as a hand. For example, hand gestures can be detected by a component of a computing device to which input via a hand gesture is directed. In turn, detected hand gestures can be forwarded as input commands or information to the computing device, a software application executing on the computing device, and the like. Other embodiments wirelessly detect macro-gestures or movements, such as a hand wave, a head turn, and so forth. At times, the gesture sensor device transmits each respective RF signal of the plurality of RF signals on a respective antenna of the multiple antennas. In some embodiments, how the RF signals are configured can enhance the quality of information that can be extracted relative to information extracted when using a random RF signal. In turn, the enhanced information can be used to identify micro-gestures of a hand or portions of the hand. For instance, some embodiments provide enough detection resolution to identify micro-gestures performed by finger(s) of a hand while the body of the hand is stationary. Sometimes, the configuration of the RF signals is based upon information extraction techniques, such as those that use radar signals. Upon transmitting the RF signals, the gesture sensor device captures, using the multiple antennas, RF signals generated by the transmitted RF signals reflecting off of the hand. In some cases, the gesture sensor device employs a pipeline that receives raw input data representing the captured RF signals, and extracts information about the hand at different levels of resolution with respective stages contained within the pipeline. The pipeline can employ various types of algorithms, such as digital signal processing algorithms and machine-learning algorithms. Some embodiments provide an ability to modify how the gesture sensor device configures the transmitted RF signals, what type of information the gesture sensor extracts, what algorithms are employed to extract the information, how the information is interpreted, and so forth.

In the following discussion, an example environment is first described in which various embodiments can be employed. Following this is a discussion of example RF signal propagation properties and how they can be employed in accordance with one or more embodiments. After this, wireless micro-gesture techniques are described. Finally, an example device is described in which various embodiments of wireless hand gesture detection can be employed.

Example Environment

FIG. 1 illustrates an example environment 100 in which wireless hand gesture detection can be employed. The example environment 100 includes a computing device 102 having a gesture sensor component 106 capable of wirelessly sensing gestures performed by hand 104. In this example environment, computing device 102 is illustrated as a mobile device, but it is to be appreciated that this is merely for discussion purposes, and that other devices can be utilized without departing from the scope of the claimed subject matter.

Gesture sensor component 106 represents functionality that wirelessly captures characteristics of a target object, illustrated here as hand 104. In this example, gesture sensor component 106 is a hardware component 106 of computing device 102. In some cases, gesture sensor component 106 not only captures characteristics about hand 104, but can additionally identify a specific gesture performed by hand 104 from other gestures. Any suitable type of characteristic or gesture can be captured or identified, such as an associated size of the hand, a directional movement of the hand, a micro-gesture performed by all or a portion of the hand (i.e., a single-tap gesture, a double-tap gesture, a left-swipe, a forward-swipe, a right-swipe, a finger making a shape, etc.), and so forth. The term micro-gesture is used to signify a gesture that can be identified from other gestures based on differences in movement using a scale on the order of millimeters to sub-millimeters. Alternately or additionally, gesture sensor component 106 can be configured to identify gestures on a larger scale than a micro-gesture (i.e., a macro-gesture that is identified by differences with a coarser resolution than a micro-gesture, such as differences measured in centimeters or meters).

Hand 104 represents a target object that gesture sensor component 106 is in process of detecting. Here, hand 104 resides in free-space with no devices attached to it. Being in free-space, hand 104 has no physical devices attached to it that couple to, or communicate with, computing device 102 and/or gesture sensor component 106. While this example is described in the context of detecting hand 104, it is to be appreciated that gesture sensor component 106 can be used to capture characteristics of any other suitable type of target object.

Signals 108 generally represent multiple RF signals transmitted and received by gesture sensor component 106. In some embodiments, gesture sensor component 106 transmits radar signals, each on a respective antenna, that are directed towards hand 104. As the transmitted signals reach hand 104, at least some reflect back to gesture sensor component 106 and are processed, as further described below. Signals 108 can have any suitable combination of energy level, carrier frequency, burst periodicity, pulse width, modulation type, waveform, phase relationship, and so forth. In some cases, some or all of the respective signals transmitted in signals 108 differs from one another to create a specific diversity scheme, such as a time diversity scheme that transmits multiple versions of a same signal at different points in time, a frequency diversity scheme that transmits signals using several different frequency channels, a space diversity scheme that transmits signals over different propagation paths, and so forth.

Having generally described an environment in which wireless hand gesture detection may be implemented, now consider FIG. 2, which illustrates an example implementation of computing device 102 of FIG. 1 in greater detail. As discussed above, computing device 102 represents any suitable type of computing device in which various embodiments can be implemented. In this example, various devices include, by way of example and not limitation: smartphone 102-1, laptop 102-2, television 102-3, desktop 102-4, tablet 102-5, and camera 102-6. It is to be appreciated that these are merely examples for illustrative purposes, and that any other suitable type of computing device can be utilized without departing from the scope of the claimed subject matter, such as a gaming console, a lighting system, an audio system, etc.

Computing device 102 includes processor(s) 202 and computer-readable media 204. Application(s) 206 and/or an operating system (not shown) embodied as computer-readable instructions on the computer-readable media 204 can be executed by the processor(s) 202 to provide some or all of the functionalities described herein.

Computer-readable media 204 also includes gesture sensor Application Programming Interfaces (APIs) 208 to provide programming access into various routines and tools provided by gesture sensor component 106. In some embodiments, gesture sensor APIs 208 provide high-level access into gesture sensor component 106 in order to abstract implementation details and/or hardware access from a calling program, request notifications related to identified events, query for results, and so forth. Gesture sensor APIs 208 can also provide low-level access to gesture sensor component 106, where a calling program can control direct or partial hardware configuration of gesture sensor component 106. In some cases, gesture sensor APIs 208 provide programmatic access to input configuration parameters that configure transmit signals (i.e., signals 108 of FIG. 1) and/or select gesture recognition algorithms. These APIs enable programs, such as application(s) 206, to incorporate the functionality provided by gesture sensor component 106 into executable code. For instance, application(s) 206 can call or invoke gesture sensor APIs 208 to register for, or request, an event notification when a particular micro-gesture has been detected, enable or disable wireless gesture recognition in computing device 102, and so forth. At times, gesture sensor APIs 208 can access and/or include low level hardware drivers that interface with hardware implementations of gesture sensor component 106. Alternately or additionally, gesture sensor APIs 208 can be used to access various algorithms that reside on gesture sensor component 106 to perform additional functionality or extract additional information, such as 3D tracking information, angular extent, reflectivity profiles from different aspects, correlations between transforms/features from different channels, and so forth.

Gesture sensor component 106 represents functionality that wirelessly detects micro-gestures performed by a hand. Gesture sensor component 106 can be implemented as a chip embedded within computing device 102. However, it is to be appreciated that gesture sensor component can be implemented in any other suitable manner, such as one or more Integrated Circuits (ICs), as a System-on-Chip (SoC), as a processor with embedded processor instructions or configured to access processor instructions stored in memory, as hardware with embedded firmware, a printed circuit board with various hardware components, or any combination thereof. Here, gesture sensor component 106 includes antennas 210, digital signal processing component 212, machine-learning component 214, and output logic component 216. In some embodiments, gesture sensor component 106 uses these various components in concert (such as a pipeline) to wirelessly detect hand gestures using radar techniques based on multiple signals.

Antennas 210 transmit and receive RF signals. As one skilled in the art will appreciate, this is achieved by converting electrical signals into electromagnetic waves for transmission, and vice versa for reception. Gesture sensor component 106 can include any suitable number of antennas in any suitable configuration. For instance, any of the antennas can be configured as a dipole antenna, a parabolic antenna, a helical antenna, a monopole antenna, and so forth. In some embodiments, antennas 210 are constructed on-chip (e.g., as part of an SoC), while in other embodiments, antennas 210 are components, metal, hardware, etc. that attach to gesture sensor component 106. The placement, size, and/or shape of antennas 210 can be chosen to enhance a specific transmission pattern or diversity scheme, such as a pattern or scheme designed to capture information about a micro-gesture performed by the hand, as further described above and below. In some cases, the antennas can be physically separated from one another by a distance that allows gesture sensor component 106 to collectively transmit and receive signals directed to a target object over different channels, different radio frequencies, and different distances. In some cases, antennas 210 are spatially distributed to support triangulation techniques, while in others the antennas are collocated to support beamforming techniques. While not illustrated, each antenna can correspond to a respective transceiver path that physically routes and manages the outgoing signals for transmission and the incoming signals for capture and analysis.

Digital signal processing component 212 generally represents functionality that digitally captures and processes a signal. For instance, digital signal processing component 212 performs sampling on RF signals received by antennas 210 to generate digital samples that represent the RF signals, and processes the digital samples to extract information about the target object. Alternately or additionally, digital signal processing component 212 controls the configuration of signals transmitted via antennas 210, such as configuring a plurality of signals to form a specific diversity scheme, such as a beamforming diversity scheme. In some cases, digital signal processing component 212 receives input configuration parameters that control an RF signal's transmission parameters (e.g., frequency channel, power level, etc.), such as through gesture sensor APIs 208. In turn, digital signal processing component 212 modifies the RF signal based upon the input configuration parameter. At times, the signal processing functions of digital signal processing component 212 are included in a library of signal processing functions or algorithms that are also accessible and/or configurable via gesture sensor APIs 208. Digital signal processing component 212 can be implemented in hardware, software, firmware, or any combination thereof.

Among other things, machine-learning component 214 receives information processed or extracted by digital signal processing component 212, and uses that information to classify or recognize various aspects of the target object, as further described below. In some cases, machine-learning component 214 applies one or more algorithms to probabilistically determine which gesture has occurred given an input signal and previously learned gesture features. As in the case of digital-signal processing component 212, machine-learning component 214 can include a library of multiple machine-learning algorithms, such as a Random Forrest algorithm, deep learning algorithms (i.e. artificial neural network algorithms, convolutional neural net algorithms, etc.), clustering algorithms, Bayesian algorithms, and so forth. Machine-learning component 214 can be trained on how to identify various gestures using input data that consists of example gesture(s) to learn. In turn, machine-learning component 214 uses the input data to learn what features can be attributed to a specific gesture. These features are then used to identify when the specific gesture occurs. In some embodiments, gesture sensor APIs 208 can be used to configure machine-learning component 214 and/or its corresponding algorithms.

Output logic component 216 represents functionality that uses logic to filter output information generated by digital signal processing component 212 and machine-learning component 214. In some cases, output logic component 216 uses knowledge about the target object to further filter or identify the output information. For example, consider a case where the target object is a hand repeatedly performing a tap gesture. Depending upon its configuration, output logic component 216 can filter the repeated tap gesture into a single output event indicating a repeated tap gesture, or repeatedly issue a single-tap gesture output event for each tap gesture identified. This can be based on knowledge of the target object, user input filtering configuration information, default filtering configuration information, and so forth. In some embodiments, the filtering configuration information of output logic component 216 can be modified via gesture sensor APIs 208.

Computing device 102 also includes I/O ports 218 and network interfaces 220. I/O ports 218 can include a variety of ports, such as by way of example and not limitation, high-definition multimedia (HDMI), digital video interface (DVI), display port, fiber-optic or light-based, audio ports (e.g., analog, optical, or digital), Universal Serial Bus (USB) ports, serial advanced technology attachment (SATA) ports, peripheral component interconnect (PCI) express based ports or card slots, serial ports, parallel ports, or other legacy ports. Computing device 102 may also include the network interface(s) 220 for communicating data over wired, wireless, or optical networks. By way of example and not limitation, the network interface(s) 220 may communicate data over a local-area-network (LAN), a wireless local-area-network (WLAN), a personal-area-network (PAN), a wide-area-network (WAN), an intranet, the Internet, a peer-to-peer network, point-to-point network, a mesh network, and the like.

Having described computing device 102 in accordance with one or more embodiments, now consider a discussion of using wireless detection of an object in accordance with one or more embodiments.

Propagation of RF Signals

As technology advances, users have an expectation that new devices will provide additional freedoms and flexibility over past devices. One such example is the inclusion of wireless capabilities in a device. Consider the case of a wireless mouse input device. A wireless mouse input device receives input from a user in the format of button clicks and movement in position, and wirelessly transmits this information to a corresponding computing device. The wireless nature obviates the need to have a wired connection between the wireless mouse input device and the computing device, which gives more freedom to the user with the mobility and placement of the mouse. However, the user still physically interacts with the wireless mouse input device as a way to enter input into the computing device. Accordingly, if the wireless mouse input device gets lost or is misplaced, the user is unable to enter input with that mechanism. Thus, removing the need for a peripheral device as an input mechanism gives additional freedom to the user. One such example is performing input to a computing device via a hand gesture.

Hand gestures provide a user with a simple and readily available mechanism to input commands to a computing device. However, detecting hand gestures can pose certain problems. For example, attaching a movement sensing device to a hand does not remove a user's dependency upon a peripheral device. Instead, it is a solution that simply trades one input peripheral for another. As an alternative, cameras can capture images, which can then be compared and analyzed to identify the hand gestures. However, this option may not yield a fine enough resolution to detect micro-gestures. An alternate solution involves usage of radar systems to transmit RF signals to a target object, and determine information about that target based upon an analysis of the reflected signal.

Various embodiments wirelessly detect hand gestures using multiple antenna. Each antenna can be configured to transmit a respective RF signal to enable detection of a micro-gesture performed by a hand. In some embodiments, the collective transmitted RF signals are configured to radiate a specific transmission pattern or specific diversity scheme. RF signals reflected off of the hand can be captured by the antenna, and further analyzed to identify temporal variations in the RF signals. In turn, these temporal variations can be used to identify micro-gestures.

Consider FIG. 3 which illustrates a simple example of RF wave propagation, and a corresponding reflected wave propagation. It is to be appreciated that the following discussion has been simplified, and is not intended to describe all technical aspects of RF wave propagation, reflected wave propagation, or detection techniques.

Environment 300a includes source device 302 and object 304. Source device 302 includes antenna 306, which is configured to transmit and receive electromagnetic waves in the form of an RF signal. In this example, source device 302 transmits a series of RF pulses, illustrated here as RF pulse 308a, RF pulse 308b, and RF pulse 308c. As indicated by their ordering and distance from source device 302, RF pulse 308a is transmitted first in time, followed by RF pulse 308b, and then RF pulse 308c. For discussion purposes, these RF pulses have the same pulse width, power level, and transmission periodicity between pulses, but any other suitable type of signal with alternate configurations can be transmitted without departing from the scope of the claimed subject matter.

Generally speaking, electromagnetic waves can be characterized by the frequency or wavelength of their corresponding oscillations. Being a form of electromagnetic radiation, RF signals adhere to various wave and particle properties, such as reflection. When an RF signal reaches an object, it will undergo some form of transition. Specifically, there will be some reflection off the object. Environment 300b illustrates the reflection of RF pulses 308a-308c reflecting off of object 304, where RF pulse 310a corresponds to a reflection originating from RF pulse 308a reflecting off of object 304, RF pulse 310b corresponds to a reflection originating from RF pulse 310b, and so forth. In this simple case, source device 302 and object 304 are stationary, and RF pulses 308a-308c are transmitted via a single antenna (antenna 306) over a same RF channel, and are transmitted directly towards object 304 with a perpendicular impact angle. Similarly, RF pulses 310a-310c are shown as reflecting directly back to source device 302, rather than with some angular deviation. However, as one skilled in the art will appreciate, these signals can alternately be transmitted or reflected with variations in their transmission and reflection directions based upon the configuration of source device 302, object 304, transmission parameters, variations in real-world factors, and so forth. Upon receiving and capturing RF pulses 310a-310c, source device 302 can then analyze the pulses, either individually or in combination, to identify characteristics related to object 304. For example, source device 302 can analyze all of the received RF pulses to obtain temporal information and/or spatial information about object 304. Accordingly, source device 302 can use knowledge about a transmission signal's configuration (such as pulse widths, spacing between pulses, pulse power levels, phase relationships, and so forth), and further analyze a reflected RF pulse to identify various characteristics about object 304, such as size, shape, movement speed, movement direction, surface smoothness, material composition, and so forth.

Now consider FIG. 4, which builds upon the above discussion of FIG. 3. FIG. 4 illustrates example environment 400 in which multiple antenna are used to ascertain information about a target object. Environment 400 includes source device 402 and a target object, shown here as hand 404. Generally speaking, source device 402 includes antennas 406a-406d to transmit and receive multiple RF signals. In some embodiments, source device 402 includes gesture sensor component 106 of FIG. 1 and FIG. 2, and antennas 406a-406d correspond to antennas 210. While source device 402 in this example includes four antennas, it is to be appreciated that any suitable number of antennas can be used. Each antenna of antennas 406a-406d is used by source device 402 to transmit a respective RF signal (e.g., antenna 406a transmits RF signal 408a, antenna 406b transmits RF signal 408b, and so forth). As discussed above, these RF signals can be configured to form a specific transmission pattern or diversity scheme when transmitted together. For example, the configuration of RF signals 408a-408d, as well as the placement of antennas 406a-406d relative to a target object, can be based upon beamforming techniques to produce constructive interference or destructive interference patterns, or alternately configured to support triangulation techniques. At times, source device 402 configures RF signals 408a-408d based upon an expected information extraction algorithm, as further described below.

When RF signals 408a-408d reach hand 404, they generate reflected RF signals 410a-410d. Similar to the discussion of FIG. 3 above, source device 402 captures these reflected RF signals, and then analyzes them to identify various properties or characteristics of hand 404, such as a micro-gesture. For instance, in this example, RF signals 408a-408d are illustrated with the bursts of the respective signals being transmitted synchronously in time. In turn, and based upon the shape and positioning of hand 404, reflected signals 410a-410d return to source device 402 at different points in time (e.g., reflected signal 410b is received first, followed by reflected signal 410c, then reflected signal 410a, and then reflected signal 410d). Reflected signals 410a-410d can be received by source device 402 in any suitable manner. For example, antennas 406a-406d can each receive all of reflected signals 410a-410d, or receive varying subset combinations of reflected signals 410a-410d (i.e. antenna 406a receives reflected signal 410a and reflected signal 410d, antenna 406b receives reflected signal 410a, reflected signal 410b, and reflected signal 410c, etc.). Thus, each antenna can receive reflected signals generated by transmissions from another antenna. By analyzing the various return times of each reflected signal, source device 402 can determine shape and corresponding distance information associated with hand 404. When reflected pulses are analyzed over time, source device 402 can additionally discern movement. Thus, by analyzing various properties of the reflected signals, as well as the transmitted signals, various information about hand 404 can be extracted, as further described below. It is to be appreciated that the above example has been simplified for discussion purposes, and is not intended to be limiting.

As in the case of FIG. 3, FIG. 4 illustrates RF signals 408a-408d as propagating at a 90° angle from source device 402 and in phase with one another. Similarly, reflected signals 410a-410d each propagate back at a 90° angle from hand 404 and, as in the case of RF signals 408a-408d, are in phase with one another. However, as one skilled in the art will appreciate, more complex transmission signal configurations, and signal analysis on the reflected signals, can be utilized, examples of which are provided above and below. In some embodiments, RF signals 408a-408d can each be configured with different directional transmission angles, signal phases, power levels, modulation schemes, RF transmission channels, and so forth. These differences result in variations between reflected signals 410a-410d. In turn, these variations each provide different perspectives of the target object which can be combined using data fusion techniques to yield a better estimate of hand 404, how it is moving, its 3-dimensional (3D) spatial profile, a corresponding micro-gesture, etc.

Having described general principles of RF signals which can be used in micro-gesture detection, now consider a discussion of various forms of information extraction that can be employed in accordance with one or more embodiments.

Wireless Detection of Micro-Gestures

The above discussion describes simple examples of RF signal transmission and reflection. In the case of using multiple antenna, it can be seen how transmitting a plurality of RF signals that have variations from one another results in receiving diverse information about a target object from the corresponding reflected signals. The diverse information can then be combined to improve detecting a characteristic or gesture associated with the target object. Accordingly, the system as a whole can exploit or optimize which signals are transmitted to improve the amount of information that can be extracted from the reflected signals. Some embodiments of a gesture sensor component capture raw data representative of signals reflected off a target object. In turn, digital-signal processing algorithms extract information from the raw data, which can then be fed to a machine-learning algorithm to classify a corresponding behavior of the target object. At times, the gesture sensor component utilizes a pipeline to identify or classify a micro-gesture.

FIG. 5 illustrates the various stages employed by an example pipeline 500 to identify micro-gestures using multiple antenna. In some embodiments, pipeline 500 can be implemented by various components of gesture sensor component 106 of FIGS. 1 and 2, such as antennas 210, digital signal processing component 212, machine-learning component 214, and/or output logic component 216. It is to be appreciated that these stages have been simplified for discussion purposes, and are not intended to be limiting. From one viewpoint, the stages can be grouped into two classifications: transmit side functionality 502 and receive side functionality 504. Generally speaking, the transmit side functionality in the pipeline does not feed directly into the receive side functionality. Instead, the transmit side functionality generates transmit signals which contribute to the reflected signals captured and processed by the receive side functionality, as further described above. Accordingly, the relationship between the transmit side functionality and the receive side functionality is indicated in pipeline 500 through the use of a dotted line to connect stage 506 of the pipeline with stage 508, rather than a solid line, since in various embodiments they are not directly connected with one another.

Stage 506 of the pipeline configures the RF transmit signals. In some cases, various transmission parameters are determined in order to generate the RF transmit signals. At times, the transmission parameters can be based upon an environment in which they are being used. For instance, the transmission parameters can be dependent upon a number of antenna available, the types of antenna available, a target object being detected, directional transmission information, a requested detection resolution, a long range object detection mode, a short range object detection mode, an expected receive-side digital signal processing algorithm, an expected receive-side machine-learning algorithm, physical antenna placement, and so forth. As noted above, the configuration of the RF transmit signals can be dependent upon an expected analysis on the receive side. Thus, the configuration of the RF transmit signals can change to support triangulation location detection methods, beamforming detection methods, and so forth. In some embodiments, the transmission parameters are automatically selected or loaded at startup (e.g., the RF transmit signal configurations are fixed). In other embodiments, these parameters are modifiable, such as through gesture sensor APIs 208 of FIG. 2.

At the start of receive side functionality 504, stage 508 performs signal pre-processing on raw data. For example, as an antenna receives reflected signal(s) (such as antennas 406a-406d receiving some or all of reflected signals 410a-410d of FIG. 4), some embodiments sample the signal(s) and generate a digital representation of the raw (incoming) signals. Upon generating the raw data, stage 508 performs pre-processing to clean up the signals or generate versions of the signals in a desired frequency band, or in a desired format. In some cases, pre-processing includes filtering the raw data to reduce a noise floor or remove aliasing, resampling the data to obtain to a different sample rate, generating a complex representation of the signal(s), and so forth. In some cases, stage 508 automatically pre-processes the raw data based upon default parameters, while in other cases the type of pre-processing is modifiable, such as through gesture sensor APIs 208 of FIG. 2.

Stage 510 transforms the received signal data into one or more different representations. Here, the signal(s) pre-processed by stage 508 are fed into stage 510. At times, stage 510 combines data from multiple paths (and corresponding antenna). The combined data can be any combination of “transmit paths”, “receive paths”, and “transmit and receive paths”. Any suitable type of data fusion technique can be used, such as weighted integration to optimize an heuristic (i.e., signal-to-noise (SNR) ratio, minimum mean square error (MMSE), etc.), beamforming, triangulation, etc. All respective paths can combined together, or various sub-combinations of paths can be made, to generate combined signal data. In some embodiments, stage 510 generates multiple combinations of signal data for different types of feature extraction, and/or transforms the signal data into another representation as a precursor to feature extraction. For example, some embodiments process the combined signal data to generate a 3 dimensional (3D) spatial profile of the target object. However, any suitable type of algorithm can be used to generate a transformed view or version of the raw data, such as an I/Q transformation that yields a complex vector containing phase and amplitude information related to the target object, a beamforming transformation that yields a spatial representation of target objects within range of a gesture sensor device, a Range-Doppler algorithm that yields target velocity and direction, a Range profile algorithm that yields target recognition information, a Micro-Doppler algorithm that yields high-resolution target recognition information, a Spectogram algorithm that yields a visual representation of the corresponding frequencies, and so forth. As described above, raw data can be processed in several ways to generate several transformations or combined signal data. At times, the same data can be analyzed or transformed in multiple ways. For instance, a same capture of raw data can be processed to generate a 3D profile, target velocity information, and target directional movement information. In addition to generating transformations of the raw data, stage 510 can perform basic classification of the target object, such as identifying information about its presence, a shape, a size, an orientation, a velocity over time, and so forth. For example, some embodiments use stage 510 to identify a basic orientation of a hand by measuring an amount of reflected energy off of the hand over time. These transformations and basic classifications can be performed in hardware, software, firmware, or any suitable combination. At times, the transformations and basic classifications are performed by digital signal processing component 212 and/or machine-learning component 214 of FIG. 2. In some cases, stage 510 automatically transforms the raw data or performs a basic classification based upon default parameters, while in other cases the transformations or classifications are modifiable, such as through gesture sensor APIs 208 of FIG. 2.

Stage 512 receives the transformed representation of the data from stage 510, and extracts or identifies feature(s) using the data. At times, feature extraction builds upon a basic classification identified in stage 510. Consider the above example in which stage 510 classifies a target object as a hand. Stage 512 can build from this basic classification to extract lower resolution features of the hand. In other words, if stage 512 is provided information identifying the target object as a hand, then stage 512 uses this knowledge to look for hand-related features (i.e., finger tapping, shape gestures, swipe movements, etc.) instead of head-related features, (i.e., an eye blink, mouthing a word, a head-shaking movement, etc.). As another example, consider a scenario where stage 510 transforms the raw signal data into a measure of the target object's velocity-over-time. In turn, this information can used by stage 512 to identify a finger fast-tap motion by using a threshold value to compare the target object's velocity of acceleration to the threshold value, a slow-tap feature, and so forth. Any suitable type of algorithm can be used to extract a feature, such as machine-learning algorithms implemented by machine-learning component 214, and/or digital signal processing algorithms implemented by digital signal processing component 212 of FIG. 2. Some embodiments simply apply a single algorithm to extract, identify, or classify a feature, while other embodiments apply multiple algorithms to extract a single feature or multiple features. Thus, different algorithms can be applied to extract different types of features on a same set of data, or different sets of data. In some cases, stage 512 searches for a default feature using default algorithm(s), while in other cases the applied algorithms and/or the feature being searched for is modifiable, such as through gesture sensor APIs 208 of FIG. 2.

Using feature extraction information generated by stage 512, stage 514 performs gesture recognition. For instance, consider a case where a finger tap feature has been extracted. Stage 514 uses this information to identify the feature as a double-click micro-gesture. At times, gesture recognition can be a probabilistic determination of which gesture has most likely occurred based upon the input information and how this information relates to one or more previously learned characteristics or features of various gestures. For example, a machine-learning algorithm can be used to determine how to weight various received characteristics to determine a likelihood these characteristics correspond to particular gestures (or components of the gestures). As in the case above, some embodiments apply a single algorithm to recognize a gesture, while other embodiments apply multiple algorithms to identify a single gesture or multiple gestures. This can include micro-gestures or macro-gestures. Further, any suitable type of algorithm can be used to identify a gesture, such as machine-learning algorithms implemented by machine-learning component 214, and/or digital signal processing algorithms implemented by digital signal processing component 212 of FIG. 2. In some cases, stage 514 uses default algorithm(s) to identify a gesture, while in other cases the applied algorithms and/or the gesture being identified is modifiable, such as through gesture sensor APIs 208 of FIG. 2.

Stage 516 filters output information generated by stage 514 and tracks this information over time. Referring back to the above example of identifying a finger tap micro-gesture, consider now the finger tap micro-gesture being performed (and identified) repeatedly. Depending upon the configuration of stage 514, stage 516 might receive multiple notifications when the micro-gesture is repeatedly identified. However, some recipients looking for a notification of a finger-tap micro-gesture might desire to be notified once when the finger tapping starts, and then a second time when the finger tapping stops. Thus, stage 516 can receive multiple input notifications, where each input notification corresponds to a respective instance of a (same) micro-gesture has been identified, and then filter the multiple input notifications into one output notification. Alternately or additionally, stage 516 can forward each input notification as respective output notifications. At times, stage 516 filters input information from stage 514 according to a set of parameters. In some embodiments, stage 516 filters the identifications or notifications using default parameters, while in other embodiments the filtering parameters are modifiable, such as through gesture sensor APIs 208 of FIG. 2.

Pipeline 500 provides an ability to detect micro-gestures (or macro-gestures) through the use of multiple antenna. This can include movements based on portions of a target object, rather than the whole target object. Consider again the example case of a target object that is a hand. On a whole, the hand, or portions of the hand, can be in a stationary position while other portions of the hand, such as one or more fingers, are moving. The above described techniques can be used to not only identify a stationary hand, but portions of the hand that are moving, such as two fingers rubbing together. Thus, a micro-gesture can entail identifying a first portion of the hand as being stationary, and identifying a second portion of the hand as having movement relative to the stationary portion. Using multiple antenna allows for the transmission of multiple RF signals, each with different configurations. In turn, when these signals reflect off a target object, different information can be extracted from each respective reflection, thus providing multiple “vantage point views” of a same target. When the analyzed together on a whole, these different views of the same target provide enough resolution to identify micro-gestures, as well as other types of gestures. For instance, a one antenna system cannot recognize a target object moving towards or away from the antenna due to the angular distance of the target object changing. However, in the same one antenna system, it becomes difficult to detect a horizontal, left to right movement across the sensor. Conversely, a multiple antenna system provides enough diversity in the various signals to make both types of directional detections. Accordingly, when pipeline 500 receives reflections of these various signals, the various stages extract different levels of resolution in the information. The first stage starts with a raw captured signal, then generates a transformed version of the raw signal. The transformed version can be used to extract a basic classification, which can then be used to extract higher-resolution feature identifications or classifications. A final stage can then be used to filtered notification outputs of the various feature identifications. Thus, pipeline 500 extracts information from each respective incoming signal, and combines the extracted information from each respective incoming signal to improve the micro-gesture identification process.

FIG. 6 is a flow diagram that describes steps in a method in accordance with one or more embodiments. The method can be implemented in connection with any suitable hardware, software, firmware, or combination thereof. In at least some embodiments, the method can be implemented by a suitably-configured system, such gesture sensor component 106 of FIGS. 1 and 2.

Step 602 determines transmission parameters for a plurality of outgoing RF signals. In some embodiments, transmission parameters for each respective outgoing RF signal are determined independently from one another, while in other embodiments, the transmission parameters for each respective outgoing RF signal are determined in dependence of one another to form a specific transmission pattern or specific diversity scheme, as further described above. The determination can be based upon an expected target object type, an expected operating environment, available hardware, anticipated feature or micro-gesture, expected feature-extraction algorithm, and so forth. In some cases, one or more gesture sensor APIs can be used to configure some or all of the transmission parameters for each outgoing RF signals.

Responsive to determining the transmission parameters, step 604 transmits the plurality of outgoing RF signals using the determined transmission parameters, such as by transmitting each respective outgoing RF signal on a respective antenna. In some embodiments, the direction of the transmissions can be based upon an expected location of a target object.

Step 606 captures incoming RF signals, such as RF signals reflected of the target object, using at least one antenna. As further described above, reflected RF signals originate from at least some of the plurality of transmitted RF contacting or connecting with a target object. While this example describes reflected signals, it is to be appreciated that other forms of signals can be captured, such as diffracted waves, refracted waves, scattered waves, and so forth. At times, all of the reflected RF signals are captured, while other times simply a portion of the RF signals are captured. Capturing an RF signal entails receiving at least a portion of the incoming RF signals using at least one antenna, but multiple antenna can be used. In some embodiments, capturing the reflected RF signals also includes digitizing the signals through a sampling process.

Responsive to capturing the incoming RF signals, step 608 processes the captured incoming RF signals to identify at least one micro-gesture or feature. For example, the processing can involve multiple stages of a pipeline, such as pipeline 500 of FIG. 5. The processing can be based upon a type of expected micro-gesture or feature (i.e., the type drives the processing algorithms used). In some embodiments, the processing extracts information at multiple levels of resolution, where each stage of processing refines or extracts information at a finer level of resolution than a previous stage. For instance, a first stage of the processing can extract a first set of information about the target object at a first level of resolution, such as classifying the target object as a hand. This first set of information can then be used as input to a second stage of processing. In turn, the second stage extracts a second level of information about the target object from the first set of information at a finer level of resolution that the first level of information, such as extracting a micro-gesture from multiple hand-related micro-gestures, and so forth. Alternately or additionally, the processing can include filtering an output based upon parameters, such as filtering how often notifications are sent when a micro-gesture has been detected.

Having considered various embodiments, consider now an example system and device that can be utilized to implement the embodiments described above.

Example Electronic Device

FIG. 7 illustrates various components of an example electronic device 700 that incorporates micro-gesture recognition using wireless techniques as describe with reference to FIGS. 1-6. Electronic device 700 may be implemented as any type of a fixed or mobile device, in any form of a consumer, computer, portable, user, communication, phone, navigation, gaming, audio, camera, messaging, media playback, and/or other type of electronic device, such as computing device 102 described with reference to FIGS. 1 and 2. In light of this, it is to be appreciated that various alternate embodiments can include additional components that are not described, or exclude components that are described, with respect to electronic device 700.

Electronic device 700 includes communication devices 702 that enable wired and/or wireless communication of device data 704 (e.g., received data, data that is being received, data scheduled for broadcast, data packets of the data, etc.). The device data 704 or other device content can include configuration settings of the device and/or information associated with a user of the device.

Electronic device 700 also includes communication interfaces 706 that can be implemented as any one or more of a serial and/or parallel interface, a wireless interface, any type of network interface, a modem, and as any other type of communication interface. The communication interfaces 706 provide a connection and/or communication links between electronic device 700 and a communication network by which other electronic, computing, and communication devices communicate data with electronic device 700.

Electronic device 700 includes one or more processors 708 (e.g., any of microprocessors, controllers, and the like) which process various computer-executable instructions to control the operation of electronic device 700 and to implement embodiments of the techniques described herein. Alternatively or in addition, electronic device 700 can be implemented with any one or combination of hardware, firmware, or fixed logic circuitry that is implemented in connection with processing and control circuits which are generally identified at 710. Although not shown, electronic device 700 can include a system bus or data transfer system that couples the various components within the device. A system bus can include any one or combination of different bus structures, such as a memory bus or memory controller, a peripheral bus, a universal serial bus, and/or a processor or local bus that utilizes any of a variety of bus architectures.

Electronic device 700 also includes computer-readable media 712, such as one or more memory components, examples of which include random access memory (RAM), non-volatile memory (e.g., any one or more of a read-only memory (ROM), flash memory, EPROM, EEPROM, etc.), and a disk storage device. A disk storage device may be implemented as any type of magnetic or optical storage device, such as a hard disk drive, a recordable and/or rewriteable compact disc (CD), any type of a digital versatile disc (DVD), and the like.

Computer-readable media 712 provides data storage mechanisms to store the device data 704, as well as various applications 714 and any other types of information and/or data related to operational aspects of electronic device 700. The applications 714 can include a device manager (e.g., a control application, software application, signal processing and control module, code that is native to a particular device, a hardware abstraction layer for a particular device, etc.).

Electronic device 700 also includes audio and/or video processing system 716 that processes audio data and/or passes through the audio and video data to audio system 718 and/or to display system 720 (e.g., a screen of a smart phone or camera). Audio system 718 and/or display system 720 may include any devices that process, display, and/or otherwise render audio, video, display, and/or image data. Display data and audio signals can be communicated to an audio component and/or to a display component via an RF link, S-video link, HDMI, composite video link, component video link, DVI, analog audio connection, or other similar communication link, such as media data port 722. In some implementations, audio system 718 and/or display system 720 are external components to electronic device 700. Alternatively or additionally, display system 720 can be an integrated component of the example electronic device, such as part of an integrated touch interface.

Electronic device 700 also includes gesture sensor component 724 that wirelessly identifies one or more features of a target object, such as a micro-gesture performed by a hand as further described above. Gesture sensor component 724 can be implemented as any suitable combination of hardware, software, firmware, and so forth. In some embodiments, gesture sensor component 724 is implemented as an SoC. Among other things, gesture sensor component 724 includes antennas 726, digital signal processing component 728, machine-learning component 730, and output logic component 732.

Antennas 726 transmit and receive RF signals under the control of gesture sensor component. Each respective antenna of antennas 726 can correspond to a respective transceiver path internal to gesture sensor component 724 that physical routes and manages outgoing signals for transmission and the incoming signals for capture and analysis as further described above.

Digital signal processing component 728 digitally processes RF signals received via antennas 726 to extract information about the target object. This can be high-level information that simply identifies a target object, or lower level information that identifies a particular micro-gesture performed by a hand. In some embodiments, digital signal processing component 728 additionally configures outgoing RF signals for transmission on antennas 726. Some of the information extracted by digital signal processing component 728 is used by machine-learning component 730. Digital signal processing component 728 at times includes multiple digital signal processing algorithms that can be selected or deselected for an analysis, examples of which are provided above. Thus, digital signal processing component 728 can generate key information from RF signals that can be used to determine what gesture might be occurring at any given moment.

Machine-learning component 730 receives input data, such as a transformed raw signal or high-level information about a target object, and analyzes the input date to identify or classify various features contained within the data. As in the case above, machine-learning component 730 can include multiple machine-learning algorithms that can be selected or deselected for an analysis. Among other things, machine-learning component 730 can use the key information generated by digital signal processing component 728 to detect relationships and/or correlations between the generated key information and previously learned gestures to probabilistically decide which gesture is being performed.

Output logic component 732 logically filters output information generated by digital signal processing component 728 and/or machine-learning component 730. Among other things, output logic component 732 identifies when received information is redundant, and logically filters the redundancy out to an intended recipient.

Electronic device 700 also includes gesture sensor APIs 734, which are illustrated as being embodied on computer-readable media 712. Gesture sensor APIs 734 provide programmatic access to gesture sensor component 724, examples of which are provided above. The programmatic access can range from high-level program access that obscures underlying details of how a function is implemented, to low-level programmatic access that enables access to hardware. In some cases, gesture sensor APIs can be used to send input configuration parameters associated with modifying operation of digital signal processing component 728, machine-learning component 730, output logic component 732, or any combination thereof, examples of which are provided above.

CONCLUSION

Various embodiments wirelessly detect micro gestures using multiple antenna of a gesture sensor device. At times, the gesture sensor device transmits multiple outgoing radio frequency (RF) signals, each outgoing RF signal transmitted via a respective antenna of the gesture sensor device. The outgoing RF signals are configured to help capture information that can be used to identify micro-gestures performed by a hand. The gesture sensor device captures incoming RF signals generated by the outgoing RF signals reflecting off of the hand, and then analyzes the incoming RF signals to identify the micro-gesture.

Although the embodiments have been described in language specific to structural features and/or methodological acts, it is to be understood that the various embodiments defined in the appended claims are not necessarily limited to the specific features or acts described. Rather, the specific features and acts are disclosed as example forms of implementing the various embodiments.

Claims

1. A method comprising: receiving, by a computing device, a plurality of incoming radio frequency (RF) signals generated by at least one outgoing RF signal reflecting off a target object;processing, by the computing device, a set of data originating from the incoming RF signals to determine shape information about the target object; andprocessing, by the computing device, the shape information about the target object to identify a gesture performed by the target object.

2. The method of claim 1, wherein: the set of data comprises raw data; andthe processing the set of data comprises applying at least one beamforming technique on the raw data.

3. The method of claim 1, further comprising: generating, by the computing device, raw data corresponding to the incoming RF signals; andtransforming, by the computing device, the raw data into at least one transformation,wherein: the transformation comprises one or more of: an I/Q transformation that yields one or more complex vectors containing phase and amplitude information,a beamforming transformation that yields a spatial representation of the target object,a range-Doppler transformation that yields velocity and direction information about the target object,a range profile transformation that yields recognition information about the target object,a micro-Doppler transformation that yields high-resolution recognition information about the target object, ora spectrogram transformation that represents corresponding frequencies of the incoming RF signals; andthe set of data comprises the transformation.

4. The method of claim 1, wherein the processing the set of data comprises first processing the set of data at a coarser resolution and subsequently processing the set of data at a finer resolution.

5. The method of claim 1, wherein the set of data includes information about a plurality of portions of the target object.

6. The method of claim 5, wherein the set of data further includes information about movement of the portions of the target object.

7. The method of claim 6, wherein the set of data further includes information about movement of the portions of the target object at a plurality of times.

8. The method of claim 5, wherein: the target object comprises a hand; andthe portions comprise respective fingers of the hand.

9. The method of claim 8, wherein the gesture comprises the hand forming a certain shape.

10. The method of claim 9, wherein the gesture further comprises one or more fingers in a certain orientation relative to other parts of the hand.

11. A device configured to identify a shape gesture performed by a target object, the device comprising: a processor; anda computer-readable media comprising instructions that, when executed by the processor, cause the device to: receive a plurality of incoming RF signals generated by at least one outgoing RF signal reflecting off the target object;generate signal data corresponding to the incoming RF signals;determine, based on the signal data, that the target object has performed the shape gesture, the shape gesture corresponding to a physical configuration or shape of the target object; andoutput an indication of the shape gesture.

12. The device of claim 11, wherein the device comprises a smartphone.

13. The device of claim 11, wherein the instructions further cause the device to transmit the at least one outgoing RF signal using one or more of: time diversity, frequency diversity, or spatial diversity.

14. The device of claim 11, wherein: the instructions further cause the processor to receive a plurality of other incoming RF signals generated by one or more other outgoing RF signals reflecting off the target object at respective previous times;the generation of the signal data comprises generating signal data for the incoming RF signals and other signal data for the other incoming RF signals; andthe determination that the target object is performing the shape gesture is based further on the other signal data.

15. The device of claim 11, further comprising a plurality of receive antennas, wherein: the receipt of the incoming RF signals comprises receiving the incoming RF signals via respective ones of the receive antennas; andthe generation of the signal data comprises applying at least one beamforming technique on the incoming RF signals.

16. The device of claim 11, wherein the determination that the target object is performing the shape gesture is based further on movement of portions of the target object.

17. The device of claim 16, wherein the determination that the target object is performing the shape gesture is based further on determining that the movement of the portions of the target object corresponds to a transformation of the target object into the physical configuration or shape.

18. The device of claim 11, wherein the target object is a hand.

19. The device of claim 18, wherein the shape gesture further comprises one or more fingers of the hand in a certain orientation relative to other parts of the hand.

20. The device of claim 18, wherein the shape gesture further comprises a finger of the hand forming a shape.

21. A device configured to identify a shape gesture performed by a target object, the device comprising: a processor; anda computer-readable media comprising instructions that, when executed by the processor, cause the device to: receive a plurality of incoming RF signals generated by at least one outgoing RF signal reflecting off the target object;generate signal data corresponding to the incoming RF signals;determine, based on the signal data and using at least one machine-learning algorithm, that the target object has performed the shape gesture, the shape gesture corresponding to a physical configuration or shape of the target object; andoutput an indication of the shape gesture.

22. The device of claim 21, wherein the machine-learning algorithm is trained using other signal data corresponding to other instances of the shape gesture.

23. The device of claim 21, wherein the machine-learning algorithm comprises one or more of a Random Forrest algorithm, a deep learning algorithm, an artificial neural network algorithm, a convolutional neural net algorithm, a clustering algorithm, or a Bayesian algorithm.

24. The device of claim 21, wherein: the signal data comprises raw data; andthe instructions further cause the device to transform, using at least one other machine-learning algorithm, the signal data into a different representation; andthe determination that the target object is performing the shape gesture is based further on the different representation.

25. The device of claim 21, wherein: the target object is a hand; andthe shape gesture further comprises one or more fingers of the hand in a certain orientation relative to other parts of the hand.

26. The device of claim 21, wherein: the signal data comprises raw data;the instructions further cause the device to transform the signal data into a different representation;the different representation comprises one or more of: an I/Q transformation that yields one or more complex vectors containing phase and amplitude information,a beamforming transformation that yields a spatial representation of the target object,a range-Doppler transformation that yields velocity and direction information about the target object,a range profile transformation that yields recognition information about the target object,a micro-Doppler transformation that yields high-resolution recognition information about the target object, ora spectrogram transformation that represents corresponding frequencies of the incoming RF signals; andthe determination that the target object has performed the shape gesture comprises inputting the different representation into the at least one machine-learning algorithm.

27. The device of claim 21, wherein the shape gesture is three-dimensional.

28. The device of claim 21, wherein the determination that the target object has performed the shape gesture comprises: determining, using a first stage, a classification of the target object; anddetermining, using a second stage, that the target object has performed the shape gesture.

29. The device of claim 28, wherein the second stage is based on the classification of the target object.

30. The device of claim 28, wherein the classification of the target object comprises a classification of the target object as a certain body part.

31. A device configured to identify a shape gesture performed by a target object, the device comprising: at least one transmit antenna;a plurality of receive antennas;a processor; anda computer-readable media comprising instructions that, when executed by the processor, cause the device to: receive a plurality of incoming RF signals generated by at least one outgoing RF signal reflecting off the target object, each of the incoming RF signals received by one of the receive antennas;generate signal data corresponding to the incoming RF signals;determine, based on the signal data, that the target object has performed the shape gesture, the shape gesture corresponding to a physical configuration or shape of the target object; andoutput an indication of the shape gesture.

32. The device of claim 31, wherein the receive antennas are spatially distributed to support triangulation between the device and portions of the target object.

33. The device of claim 31, wherein: the instructions further cause the device to determine a spatial profile of the target object; andthe determination that the target object has performed the shape gesture is based further on the spatial profile of the target object.

34. The device of claim 33, wherein the transmit antenna comprises a plurality of transmit antennas.

35. The device of claim 34, wherein the instructions further cause the device to transmit the at least one outgoing RF signal using constructive or destructive interference.

36. The device of claim 31, wherein: the signal data comprises raw data;the instructions further cause the device to transform the signal data into a different representation;the different representation comprises one or more of: an I/Q transformation that yields one or more complex vectors containing phase and amplitude information,a beamforming transformation that yields a spatial representation of the target object,a range-Doppler transformation that yields velocity and direction information about the target object,a range profile transformation that yields recognition information about the target object,a micro-Doppler transformation that yields high-resolution recognition information about the target object, ora spectrogram transformation that represents corresponding frequencies of the incoming RF signals; andthe determination that the target object has performed the shape gesture is based further on the different representation.

37. The device of claim 36, wherein the shape gesture comprises a swipe or tap gesture.

38. The device of claim 31, wherein the signal data comprises respective signal data for a plurality of portions of the target object.

39. The device of claim 38, wherein the signal data comprises movement information corresponding to the portions of the target object.

40. The device of claim 39, wherein: the target object is a hand; andthe portions of the target object are fingers of the hand.