Augmented Reality System

Abstract

To provide an augmented reality system.SOLUTION: The augmented reality (AR) system comprises a user interface component, one or more sensors which generate sensor data representing a part of an environment in which a user is located, and a memory. The memory stores therein object association data associating the user with one or more objects in the environment and object location data indicating each location of each of the one or more objects. The AR system determines a position of the user, determines an updated location of one object out of the one or more objects in dependence on the generated sensor data and the determined position of the user, updates the stored object location data to indicate the determined updated location of the one object out of the one or more objects, and outputs information depending on the updated location of the one object out of the one or more objects via the user interface.SELECTED DRAWING: Figure 3

Description

[Technical Field]

[0001] The present invention relates to Augmented Reality (AR) systems, and has particular, but not exclusive, relevance to the use of AR systems to keep track of a user's belongings. [Background technology]

[0002] Losing belongings can cause serious inconvenience, for example, because one leaves belongings behind when leaving a public place, and can be costly if the belongings cannot be found or recovered. This problem can be particularly prevalent for people with memory problems, for example, resulting from dementia, and / or for small belongings whose absence may not be easily noticed. For example, it is common for people to leave their smartphone, a pair of headphones, a set of keys, a wallet and / or a purse at their home.

[0003] Augmented reality (AR) devices, such as AR headsets, that provide visual information to enhance a user's experience of an environment are becoming smaller, lighter, and have form factors that enable extended and / or everyday use. Summary of the Invention

[0004] According to a first aspect, an augmented reality (AR) system is provided. The AR system includes a user interface, one or more sensors configured to generate sensor data representing a portion of an environment in which a user of the AR system is located, and a memory. The memory is configured to store object association data associating the user with one or more objects in the environment and object position data indicating a respective position of each of the one or more objects. The AR system is configured to determine a position of the user, determine an updated position of one of the one or more objects in response to the generated sensor data and the determined position of the user, update the stored object position data to indicate the determined updated position of the one of the one or more objects, and output information in response to the updated position of the one of the one or more objects via the user interface.

[0005] According to a second aspect, there is provided a computer-implemented method including storing object association data associating one or more objects in an environment with a user of an AR system, receiving sensor data representative of a portion of the environment in which the user is located, determining a location of the user, determining a location of one of the one or more objects associated with the user in response to the received sensor data and the determined location of the user, storing object position data indicative of the determined location of the one of the one or more objects, and outputting information in response to the determined location of the one of the one or more objects via a user interface of the AR system.

[0006] Further features and advantages of the present invention will become apparent from the following description of preferred embodiments of the invention, given by way of example only, made with reference to the accompanying drawings, in which: [Brief explanation of the drawings]

[0007] [Figure 1] FIG. 1 is a schematic block diagram illustrating an augmented reality (AR) system according to an embodiment. [Figure 2] 1 illustrates an example of an AR system including a pair of smart glasses. [Figure 3] FIG. 1 is a flow diagram illustrating a computer-implemented method for generating information for output via a user interface of an AR system, according to an embodiment. [Figure 4A] 3 illustrates an example of information being output by the AR system of FIG. 2. [Figure 4B] 3 illustrates an example of information being output by the AR system of FIG. 2. [Figure 5] FIG. 1 is a flow diagram illustrating a computer-implemented method for generating an alert for output via a user interface of an AR system, according to an embodiment. [Figure 6A] 3 illustrates an example of an alert being output by the AR system of FIG. 2. [Figure 6B] 3 illustrates an example of an alert being output by the AR system of FIG. 2. [Figure 6C] 3 illustrates an example of an alert being output by the AR system of FIG. 2. [Figure 7] FIG. 10 is a flow diagram illustrating a further computer-implemented method for generating an alert for output via a user interface of an AR system, according to an embodiment. [Figure 8] FIG. 1 is a flow diagram illustrating a computer-implemented method for generating data associating a user of an AR device with objects in an environment, according to an embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0008] 1 illustrates an example of an AR system 100. AR system 100 may be embodied as a single AR device, such as a headset, a pair of smart glasses, or any other type of suitable wearable device. AR system 100 may alternatively include multiple devices connected by wired or wireless means. Particular embodiments are described in more detail with reference to FIG. 2.

[0009] The AR system 100 includes one or more sensors 102 configured to generate sensor data representing a portion of an environment in which a user of the AR system 100 is located. The sensor 102 may be a component of a single device, such as an AR headset, or may be a component of multiple connected devices. The sensor 102 includes one or more cameras for generating image data representing a portion of the environment that falls within the field of view of the one or more cameras. The field of view may be defined vertically and / or horizontally depending on the number and position of the cameras. For example, the cameras may be configured to face substantially the same direction as the head of a user wearing the AR headset, in which case the field of view of the one or more cameras may include all or a portion of the user's field of view. Alternatively, the field of view may include an area wider than completely surrounding the user, for example. The cameras may include stereo cameras, which enable the AR system to derive depth information indicating distances to objects in the environment using stereo matching. The sensor 102 may alternatively or additionally include a depth sensor for generating the depth information, such as an infrared camera, a sound navigation and ranging (sonar) transceiver, and / or a light detection and ranging (LIDAR) system. The AR system 100 can be configured to combine image data and associated depth information to generate a three-dimensional representation of a portion of an environment, for example in RGB-D format, and / or as a point cloud or volumetric representation.

[0010] The sensors 102 include position sensors for determining the location and / or orientation (collectively referred to as position or attitude) of a user of the AR system 100. The position sensors may include a Global Positioning System (GPS) module, one or more accelerometers, and / or a Hall-effect magnetometer (electronic compass) for determining orientation. The AR system 100 can additionally or alternatively determine or refine the estimated location of the user by analyzing image data and / or depth information using simultaneous location and mapping (SLAM) techniques.

[0011] The AR system 100 includes a user interface 104 through which a user can interact with the AR system 100. The user interface 104 includes input and output devices that may be components of a single AR device or may be components of multiple connected devices. The output devices are configured to output user information and include one or more displays for providing visual information to the user to enhance the user's experience of the environment. The one or more displays may include opaque displays configured to generate and display image data corresponding to a representation of a portion of the environment generated using one or more cameras and / or depth sensors, with additional information or virtual objects overlaid or otherwise combined with the generated representation of the environment. Additionally or alternatively, the one or more displays may include transparent displays onto which the user can directly observe the environment and onto which information or virtual objects are projected, for example, using wave guidance or laser scanning display technology.

[0012] The output device may include one or more loudspeakers, for example attached to earpieces or headphones, thereby enabling the AR system 100 to output information to the user in the form of audio, which may include, for example, synthesized or pre-recorded voices, beeps, buzzers, clicks, music, or any other sound suitable for conveying information to the user.

[0013] The output device may further include a haptic output device configured to generate forces that cause movement of part or all of the AR system 100, including, for example, vibrations, clicks, or other actions that can be detected by sensing a user's touch. In one example, the AR headset can send signals to an additional device, such as a smartwatch, fitness tracker, bracelet or other wearable device, or smartphone, to provide the additional device with a haptic output for the user.

[0014] The input devices of the user interface 104 are configured to receive information from a user of the AR system 100. The input devices may include one or more microphones for capturing speech or other sounds made by the user. For example, the input devices may include a microphone array that enables the AR system 100 to determine a direction relative to an audio source, thereby enabling the AR system 100 to distinguish sounds made by the user from other sounds in the environment. The AR system 100 may further be configured to perform speech recognition and respond to verbal commands from the user.

[0015] The input device may include one or more eye tracking sensors configured to track the orientation and / or movement of the user's eyes. The eye tracking sensor may be, for example, an optical eye tracking sensor that can track the orientation of the eyes by analyzing images of the eyes generated by a camera facing the eyes. The eye tracking sensor can generate eye tracking data that enables AR system 100 to determine what part of the environment or what object within the environment the user is currently looking at. The eye tracking sensor may also be used to determine when the user blinks or closes their eyes, which AR system 100 can use as an input signal.

[0016] The input device may further include a button or a touch input device. For example, the AR system 100 may include one or more scroll wheels, touch-sensitive areas, or a trackpad. As mentioned above, the input device may be part of the AR device that houses the sensor 102, or may be part of a separate, remote device.

[0017] In some embodiments, one or more cameras of AR system 100 may further function as user input devices, for example, to facilitate gesture recognition. Additionally, an accelerometer and / or electronic compass can be used to determine when a user nods or shakes their head.

[0018] The AR system 100 includes a memory 108 and a processing circuit 110. The memory 108 and the processing circuit 110 may be part of the AR device that houses the sensor 102. Alternatively, portions of the memory 108 and the processing circuit 110 may be part of one or more separate devices, such as a dedicated computing device, a smartphone, a tablet or laptop computer, a desktop computer, a server, or one or more devices in a networked system. In embodiments, certain data storage and processing tasks are performed locally in the AR device, while other data storage and processing tasks are performed remotely. In this manner, the data storage and processing performed by the AR device can be kept to a necessary minimum, thereby allowing the AR device to have a size, weight, and form factor that is practical and attractive for long-term and / or everyday use of the AR device.

[0019] The memory circuitry 108 includes non-volatile and volatile random-access memory (RAM), such as static random-access memory (SRAM) and dynamic random-access memory (DRAM), as well as non-volatile storage, for example in the form of one or more solid-state drives (SSDs). Other types of memory, such as removable storage, synchronous DRAM, etc., may be included.

[0020] Processing circuitry 110 may include various processing units, including a central processing unit (CPU), a graphics processing unit (GPU), and / or a specialized neural processing unit (NPU) for efficiently performing neural network operations. In the present invention, neural networks may be used for specific tasks, including object detection and SLAM, as described in more detail below. Processing circuitry 110 may also include other specialized processing units, such as an application specific integrated circuit (ASIC), a digital signal processor (DSP), or a field programmable gate array (FPGA).

[0021] The memory 108 carries machine-readable instructions in the form of program code that, when executed by the processing circuitry 110, causes the AR system 100 to perform the method described below. The memory 108 is also configured to store additional data for use in performing the method. The additional data in this example includes sensor data generated by the one or more sensors 102, object association data associating a user of the AR system 100 with one or more physical objects in the environment, and object position data indicating a respective position of each of the one or more objects.

[0022] 2 shows an example of an AR system 200 that includes a pair of smart glasses 202 and an associated app on a smartphone 203. In this example, the smart glasses 202 and smartphone 203 are paired during a pairing process, but in other examples, the smart glasses 202 may be paired with another type of device, such as a smartwatch or tablet computer, or may function without being paired with any additional devices.

[0023] The smart glasses 202 include a central frame portion 204 and two folding arms 206a, 206b, with the central portion 204 serving as a support for two lenses 208a, 208b. The central frame portion 204 and arms 206a, 206b house various sensors and user interface components, as described below. The lenses 208a, 208b in this example are neutral, but in other examples, the lenses may be corrective lenses matching a particular user's prescription and / or may be tinted, for example, in the case of smart sunglasses. Each of the lenses 208a, 208b is a transparent display upon which a corresponding projection component 210a, 210b is configured to display information for the user.

[0024] The central frame portion 204 houses two forward-facing cameras 212a, 212b with a combined field of view that approximately corresponds to the user's field of view. The AR system 200 is configured to analyze image data generated by the cameras 212a, 212b using stereo matching to determine depth information. The central frame portion 204 also houses a microphone array 214 for receiving audio input from the user and optical eye tracking sensors 216a, 216b for tracking the orientation and movement of the user's right and left eyes, respectively. The arms 206a, 206b house a communications module including a dedicated power source, processing circuitry, and memory circuitry, as well as a global positioning system (GPS) receiver, an electronic compass, an accelerometer, and an antenna for wireless communication with a smartphone 203 running an associated app. It should be noted that while the smart glasses 202 in this example include the necessary components for the smart glasses 202 to function independently of the smartphone 203, the smartphone 203 also includes certain components, such as a GPS receiver and an accelerometer, that are equivalent to components of the smart glasses 202. Where appropriate, the AR system 200 can use components of the smartphone 203 in place of equivalent components of the smart glasses 202, for example, to conserve battery power of the smart glasses 202.

[0025] The AR system 200 is configured to determine the user's position (i.e., location and orientation) using the smart glasses 202's on-board GPS receiver and electronic compass and / or by processing image data from the cameras 212a, 212b using SLAM. The AR system 200 may be configured to continuously monitor the user's position, or alternatively, to determine the user's position only when a specific event is detected (e.g., when user movement is detected by an accelerometer).

[0026] The smart glasses 202 can be configured according to the user's preferences, for example, using an app on the smartphone 203 or using the input devices of the smart glasses 202 directly. For example, the user can use the app to select the type of information displayed on the lenses 208a, 208b and whether the smart glasses 202 continuously monitor the user's location as described above. The app has associated storage on the smartphone 203, which can be used to store data used by the AR system 200, in addition to the memory circuitry of the smart glasses 202. The AR system 200 can further utilize the processing power of the smartphone 203 to perform certain resource-intensive processing tasks, such as SLAM. Sharing the storage and processing requirements of the AR system 200 between the smart glasses 202 and the smartphone 203 allows the size, weight, and form factor of the smart glasses 202 to be comparable to that of a regular pair of eyeglasses, thereby allowing the user to wear the smart glasses 202 comfortably for extended periods and on a daily basis.

[0027] 3 illustrates an example of a method performed by AR system 200 in accordance with the present invention. Although the method is described with reference to AR system 200, it will be understood that the same method can be performed by any suitable embodiment of AR system 100 without departing from the scope of the present invention.

[0028] The AR system 200 stores 302 object association data associating a user of the AR system 200 with one or more objects in the environment. The AR system 200 may store association data for multiple users. In this example, the object association data can be viewed using an app on the smartphone 203. The AR system 200 stores object association data for user John Doe, indicating that John Doe is associated with three objects: a set of keys, a wallet, and the smartphone 203. In this example, the app displays a table under the user's name, with rows corresponding to different objects associated with the user. The column titled "O" contains an icon representing the object. The column titled "P" indicates whether the user currently has the object. The column titled "L" indicates the location of the object. Methods for generating, storing, and updating object association data are described with reference to FIG. 8.

[0029] The AR system 200 receives 304 sensor data representing a portion of an environment in which a user of the AR system 200 is located. In this example, the sensor data includes image data generated by the forward-facing cameras 212 a, 212 b. The cameras 212 a, 212 b are configured to continuously generate frames of image data for analysis by the AR system 200. The frames of image data are generated at a rate high enough to capture events occurring in the environment, but slow enough to allow the AR system 200 to analyze the image data in real time. In other examples, the sensor data may include other types of data, such as depth information.

[0030] The AR system 200 determines 306 the location of the user of the AR system 200. Depending on the information available to the AR system 200, the determined location may be a global location or a local location relative to a local coordinate system within a room or other vicinity in which the user is located. In this example, the AR system 200 is configured to determine the user's location and orientation using an on-board GPS receiver, accelerometer, and / or electronic compass of the smart glasses 202 and / or by processing image data from the cameras 212a, 212b. For example, the AR system 200 may use a GPS receiver and electronic compass to determine the user's approximate location and orientation, and then use SLAM to determine a refined location and orientation of whether the user is in a suitable location (e.g., whether the user is indoors). Alternatively, if the AR system 200 cannot determine the user's global location using a GPS receiver, the AR system 200 may use SLAM to determine the local location. Additionally or alternatively, the AR system 200 may use computer vision techniques, such as scene recognition, to determine the user's location. For example, the AR system 200 can use scene recognition to determine that the user is in a pub or restaurant, or in the user's home.

[0031] At 308, the AR system 200 determines the position of one of the objects indicated as associated with the user in response to the sensor data received at 306 and the user's position determined at 308. In this example, the AR system 200 processes image data generated by the cameras 212a, 212b using object detection. Upon detecting one of the objects associated with the user, the AR system 200 determines the position of the detected object relative to the user. In this example, the position of the detected object relative to the user is determined in three dimensions using depth information derived using stereo matching. The AR system 200 then determines the position of the object in response to the determined position of the user. The position of the object may be a global position, a local position relative to a local coordinate system, or a combination of both. For example, if the user's position determined at 306 is a global position, the AR system 200 can determine the global position of the object. If the user's position determined at 306 is a local position, the AR system 200 can determine the local position of the object.

[0032] In addition to determining the location of the detected object, the AR system 200 determines whether the user is currently holding the detected object, for example, whether the user is currently holding the object, carrying the object in a pocket, or wearing the object if the object is an item of clothing. To this end, the AR system 200 is configured to recognize when the object is being picked up or placed down by the user. Because the cameras 212a and 212b are front-facing and have a field of view roughly equivalent to the user's field of view, when the user picks up or places down the object, it is highly likely that the object will be within the field of view of the cameras 212a and 212b. In this embodiment, to recognize when the object is being picked up or placed down, a machine learning classifier based on a convolutional neural network is trained to identify when the user is holding the object in their hand. If, in a sequence of images captured by cameras 212a, 212b, it is determined that an object remains in the user's hand while leaving the field of view of cameras 212a, 212b, AR system 200 determines that the object has been picked up by the user and therefore determines that the user is holding the object. In contrast, if, in a sequence of images, the object is first determined to be in the user's hand and then determined to be no longer in the user's hand but still within field of view, AR system 200 can determine that the object has been placed or positioned in the environment and therefore determines that the user is no longer holding the object. It will be understood that other methods may be used to identify when a user is picking up or placing an object. More generally, if AR system 200 detects an object in the environment and the object is not being held by the user, AR system 200 determines that the user is not currently holding the detected object. At 310, AR system 200 stores object position data indicating the determined position of the object.The object location data may include precise global or local coordinates of the determined location of a given object, including, for example, latitude, longitude, and altitude. Other representations are possible without departing from the scope of the invention. For example, if the surface of the Earth is divided into a 3-meter square grid, it can be shown that every 3-meter square can be uniquely identified using a permutation of three words in the English language.

[0033] In addition to a coordinate representation of the object's determined location, the AR system 200 can determine a name or other identifier, such as a zip code or postcode, of the updated location, for example, by interfacing with local or cloud-based mapping software and / or by recognizing a user-specified location, such as "home," "work," etc. By using the determined user location and analyzing image data received from cameras 212a, 212b, the AR system 200 can determine a very specific identifier for the object's location, such as, for example, "on my desk at work" or "above the bar at Eagle Pub." In this example, the object location data also indicates whether the user is currently holding the object.

[0034] In a further example, the AR system may store object position data hierarchically. For example, the AR system may recognize when a first object is located within a second object, such as a bag, thereby forming an association between the positions of the first and second objects. The object position data will then indicate that the position of the first object matches the position of the second object until the AR system 200 determines that the positions of the two objects are no longer associated. Similarly, the object position data may indicate that a first object, such as a credit card, is located within a second object, such as a purse, which is located within a third object, such as a handbag. The AR system 200 forms associations between the positions of the three objects such that the object position data indicates that all three positions match the position of the handbag until the AR system 200 determines that the positions of the three objects are no longer associated.

[0035] The AR system 200 outputs information in response to the stored object position data at 312. The information may indicate, for example, the most recently determined position of one of the objects associated with the user. In one embodiment, the AR system 200 outputs the information in response to a request from the user. The request may include a verbal request received via the microphone array 214, in which case the AR system 200 can use speech recognition and natural language processing to identify the request and / or determine the content of the request. A verbal request may include, for example, a user asking, "Where are my keys?" The request may additionally or alternatively include a gesture, in which case the AR system 200 can identify the request by analyzing image data received from the cameras 212a, 212b. The request may be received via any other suitable input method.

[0036] The AR system 200 may output information to the user through any user interface of the AR system 200, for example, by displaying information on the lenses 208a, 208b of the smart glasses 202. FIG. 4A shows an example in which information indicating the location of an object associated with the user is displayed. The displayed information includes an icon 402 representing the object (a set of keys), an arrow 404 indicating the direction to the object's most recently determined location, and text indicating the distance to the object's most recently determined location. FIG. 4B shows another example in which information indicating the location of an object associated with the user is displayed. The displayed information includes an icon 406 representing the object (a wallet) and a text description of the object's most recently determined location. In a further example, the information may include an image captured by the cameras 212a, 212b indicating the location where the AR system 200 last detected the object.

[0037] As an alternative to displaying information on lenses 208a, 208b, AR system 200 can output information via an app on smartphone 203. In the example of FIG. 2, the app displays a table showing object position data for several objects associated with a user. The column titled "P" indicates whether user John Doe currently possesses each object ("Y" indicates that the user currently possesses the object, and "N" indicates that the user does not currently possess the object). In this case, John Doe does not currently possess a set of keys or a wallet, but he does currently possess smartphone 203. The column titled "L" indicates the respective location of the objects. In this example, it has been determined that John Doe currently possesses smartphone 203, so the table does not display the location of smartphone 203.

[0038] In the above-described embodiments, information is displayed visually to a user of AR system 200. In other embodiments, information may be communicated to a user via any other suitable method, for example, as audio, including synthesized or pre-recorded speech.

[0039] FIG. 5 illustrates a further method performed by AR system 200 (or any other suitable embodiment of AR system 100). Actions 502-510 of FIG. 5 are identical to actions 302-310 of FIG. 3. At 512, AR system 200 detects or predicts separation of a user from one of the objects associated with the user. Such separation may be detected, for example, when an object is within the field of view of cameras 218a, 218b and either the object or the user moves such that the object is no longer within view. For example, separation may be predicted if the angular separation of the object from the axis of cameras 212a, 212b exceeds a predetermined threshold, indicating that the user has rotated away from the object.

[0040] The AR system 200 may predict the separation of an object from a user when the user places an object in a specific location. For example, if the AR system 200 detects that a user places a set of keys on a surface in a public place, the AR system 200 can predict the separation of the set of keys from the user. The AR system 200 may further use the eye tracking sensors 216 a, 216 b to determine whether the user is looking directly at the object while placing it down. If the user is not looking directly at the object, the AR system 200 can determine that the user is not focused on the object and, accordingly, predict the separation of the object from the user.

[0041] The AR system 200 may further predict or detect separation of an object from the user when the object is outside the field of view of the cameras 212a, 212b. For example, the AR system 200 may detect that the user is moving away from the object from the user's position determined in 506 and, accordingly, detect separation between the user and the object. Separation between the user and the object may be detected, for example, when it is determined that the distance between the user and the object increases beyond a predetermined threshold. Alternatively, the AR system 200 may determine that the user is leaving the premises where the object is located without possessing the object, thereby detecting separation between the user and the object. For example, the AR system 200 may recognize that the user has placed a set of keys on a table in a restaurant and store object position data indicating the location of the keys. At a later point in time, the AR system 200 may determine that the user has left the restaurant without possessing the set of keys, thereby detecting separation between the user and the set of keys.

[0042] Separation can also be detected or predicted when the AR system 200 determines that an object has moved while within the field of view of the cameras 212a, 212b. The object may move, for example, if someone other than the user picks up the object (which may indicate that the object has been stolen) or if the object falls off a surface. The AR system 200 can also detect the position or movement of an object while it is not within the field of view of the cameras 212a, 212b. For example, the AR system 200 may detect the movement of a connected device, such as the smartphone 203, based on wireless signals transmitted by the connected device, such as Bluetooth or Wi-Fi signals.

[0043] The AR system 200 generates an alert at 514 in response to the detection or predicted separation of the object from the user and outputs the alert via a user interface at 516. In this example, the alert is a visual alert projected onto the lenses 208 a, 208 b of the smart glasses 202. In other examples, the alert may alternatively or additionally have an audio component, such as a pre-recorded or synthesized voice or any other sound. The alert may identify the object and / or include information indicating the object's location as indicated by the object position data. In some examples, different levels of alert may be generated in response to different events. For example, the AR system 200 may generate a first alert when it predicts separation of the user from the object in response to the user rotating away from the object so that the object is no longer within the field of view of the cameras 212 a, 212 b. The AR system 200 may then generate a second alert when it detects separation of the user from the object, for example, when the user moves away from or leaves the premises in which the object is located. The first alert may be a relatively unobtrusive visual alert, such as an arrow displayed on the periphery of the lenses 208a, 208b, while the second alert may be more intrusive, such as an arrow displayed on the center of the lenses 208a, 208b and / or including an audio component. The AR system 200 may generate a higher priority alert, such as an audio component, if the AR system 200 detects that an object is being stolen. FIGS. 6A-6C illustrate an example in which an alert is generated for a user of the AR system 200. In FIG. 6A, the AR system 200 identifies the user's smartphone 203 on a table 604. In this example, a bounding box 602 is displayed on the lenses 208b to indicate to the user that the AR system 200 has identified the smartphone 203. The stored object location data is updated to indicate the location of the smartphone 203 on the table 604.In Figure 6B, the user begins moving in the direction indicated by arrow A. In Figure 6C, the user continues moving in the direction indicated by arrow A, so that smartphone 203 is no longer within the field of view of cameras 212a, 212b. In this example, AR system 200 detects the user's separation from smartphone 203 when the position of smartphone 203, as indicated by the object position data, moves from within the field of view of cameras 212a, 212b to outside the field of view of cameras 212a, 212b. Accordingly, AR system 200 generates and displays an alert on lens 208b in the form of arrow 606 pointing toward the position of smartphone 203 as indicated by the object position data.

[0044] FIG. 7 illustrates a further method performed by AR system 200 (or any other suitable embodiment of AR system 100). Operations 702-712 of FIG. 7 are identical to operations 502-512 of FIG. 5, except that, at 702, AR system 200 further stores privileged location data indicating one or more privileged locations. The privileged locations may include, for example, the user's home and / or the user's workplace. In this example, the user can enter the privileged locations via an app on smartphone 202, either by manually entering location details or by indicating that the user is currently located at the privileged location. In other examples, the AR system can automatically identify a location as the user's home, for example, by analyzing the user's locations over time. In this example, the privileged location data indicates a precise set of coordinates for each privileged location, but in other examples, the privileged location data may include additional information, such as the layout of a home or other premises, generated using SLAM, for example.

[0045] If separation between the user and one of the objects represented in the object association data is detected or predicted at 712, the AR system 200 determines at 714 whether the position of the object represented by the object position data corresponds to one of one or more privileged positions. In this example, if the object's position is less than a threshold distance from the specified coordinates of the privileged position, the object's position is determined to correspond to the privileged position. In other examples, for example, if the privileged positions include the layout of a home or other premises, the AR system can accurately determine whether the object's position is within the premises and therefore corresponds to a privileged position.

[0046] In this example, if the AR system 200 determines that the object's location corresponds to a privileged location, the AR system 200 takes no further action regarding the object's detection or predicted separation from the user. In particular, if the object is left in a privileged location, the AR system 200 prevents alerting the user. For example, a user may choose to leave one or more objects at home, in which case it is undesirable for the AR system 200 to alert the user when the user leaves home. Even if a user accidentally leaves an object at home, this may be less inconvenient and / or costly than accidentally leaving an object elsewhere. If the AR system 200 determines that the object is not located in a privileged location, the AR system generates and outputs an alert at 716 and 718, as described in 514 and 516 of FIG. 5 .

[0047] 7, AR system 200 is configured to generate an alert when it determines that an object is not located in a privileged location; however, AR system 200 may instead be configured to generate an alert when it determines that an object is located in a privileged location. Furthermore, different privileged locations may be applicable to different objects. For example, a user may want to be notified if they leave their home without an object (e.g., a set of keys or a smartphone) that belongs to a specified set of objects, but may not want to be notified if they leave their home without an object that does not belong to the specified set of objects.

[0048] 8 illustrates an example of a method performed by AR system 200 (or any other suitable embodiment of AR system 100) to store object association data associating a user of AR system 200 with a new item previously unknown to AR system 200. AR system 200 receives image data generated by cameras 212a, 212b at 802. In this example, an app on smartphone 203 prompts the user to hold a new object in the field of view of cameras 212a, 212b at different orientations such that cameras 212a, 212b capture images of the object at various different orientations.

[0049] The AR system 200 learns the appearance of the object at 804. In this example, the AR system 200 processes images of the object at different orientations and uses the images of the object at different orientations as training data to train a classifier using supervised learning. The AR system 200 thereby learns to identify the object from various angles and can therefore recognize the object when it appears in subsequent images. In this example, the AR system 200 associates a unique identifier with the object, and upon detecting the object in an image, the AR system 200 is trained to determine the unique identifier as well as the coordinates of a bounding box containing the object. In this example, the AR system 200 is pre-trained to identify a particular general class of object, e.g., "smartphone" or "key." This allows the AR system 200 to efficiently detect objects of a general class, reducing the difficulty of training the AR system 200 to learn the appearance of a specific object. This principle is called transfer learning.

[0050] Once the object's appearance is learned, the AR system 200 stores object association data indicating the association between the user and the object at 806. The object association data in this example includes a unique identifier for the object, an icon or image representing the object, and optionally a name for the object.

[0051] In the example of FIG. 8 , the AR system 200 learns the appearance of an object after prompting the user to hold the object in view of the cameras 212 a, 212 b. In other examples, the AR system can automatically generate object association data, for example, if the AR system detects a particular object that frequently appears in the user's home or another privileged location. In other examples, the user may place a unique AR tag or marker on the object that the AR system can be trained to detect. In yet another example, the AR system can learn other characteristics of the object, for example, the identity of the connected device obtained during the wireless pairing process. The object association may then include the device's identity as an alternative to, or in addition to, the device's appearance. The AR system can then identify the device based on the wireless signal transmitted by the device.

[0052] The above embodiments should be understood as illustrative examples of the present invention. Further embodiments of the present invention are contemplated. For example, an AR system may be trained to recognize specific people other than the user of the AR system (e.g., members of the user's family or friends). The AR system may identify one of the recognized people picking up an object and store object position data indicating that the object is in the person's possession. Furthermore, if the AR system identifies a person other than the user picking up an object, the AR system determines whether the user is one of the recognized people. If the person is recognized, the AR system takes no further action. If the person is not recognized, the AR system generates an alert indicating that the person may steal the object. The AR system may also be configured to capture and store an image of the person who may steal the object for use in later identification of the person. Additionally or alternatively, generating an alert may depend on whether the person picks up an object in a privileged position. For example, if the AR system identifies an object being picked up by another person in the user's home, the AR system may not take any further action. For example, an AR system with a wide field of view that completely surrounds the user is particularly well-suited for such applications.

[0053] It will be understood that any feature described in connection with any one embodiment may be used alone or in combination with other features described, and may also be used in combination with any other one or more features of any of the embodiments, or in any combination of other embodiments. Furthermore, equivalents and modifications such as those described above may also be used without departing from the scope of the invention as defined in the appended claims.

Claims

1. 1. An augmented reality (AR) system, comprising: A user interface; one or more sensors configured to generate sensor data including first image data representative of a portion of an environment in which a user of the AR system is located at a first time point and second image data representative of a portion of the environment in which the user of the AR system is located at a second time point; A memory, object association data associating the user with a plurality of objects in the environment, the plurality of objects including a first object and a second object, the first object being capable of being located within the second object; and object position data indicating the position of each of the plurality of objects; and a memory configured to store The AR system comprises: determining that the first object is located inside the second object by detecting the first object and the second object in the first image data; determining a location of the user at the second time; determining a position of the second object in response to the determined position of the user at the second time by detecting the second object in the second image data; updating the stored object position data to indicate the determined position of the second object and the association between the first object and the second object; outputting information via the user interface in response to the updated stored object position data, the information indicating that the first object is present at the determined position of the second object. It is configured as follows: Augmented reality (AR) system.

2. determining the location of the second object includes determining whether the user is currently holding the second object; the object position data indicating whether the user is currently in possession of the second object; The AR system of claim 1 .

3. Detecting or predicting separation of the second object from the user in response to the updated stored position data; generating an alert in response to the detected or predicted separation of the second object from the user; outputting the alert via the user interface; 3. The AR system according to claim 1, wherein the AR system is configured as follows:

4. the memory is further configured to store privileged location data indicative of one or more privileged locations; generating the alert is dependent on whether the determined location of the second object corresponds to any of the one or more privileged locations. The AR system according to claim 3 .

5. configured to use the determined location of the user to detect or predict when the user has left one of the one or more privileged locations; generating the alert is dependent on the detection or prediction of the user leaving the one of the one or more privileged locations; The AR system according to claim 4 .

6. configured to use the second image data to detect pick-up of the second object by a person other than the user; the detecting or predicting the separation of the second object from the user is dependent on the detected picking up of the second object by the person other than the user. The AR system according to claim 3 .

7. The AR system of claim 6 , configured to capture an image of the person other than the user upon detecting the pick-up of the second object by the person other than the user.

8. The AR system of claim 3 , wherein the alert includes information indicating the determined position of the second object.

9. The AR system according to any one of claims 1 to 8, wherein the outputting of the information via the user interface is in response to a request from the user.

10. The AR system of claim 9 , wherein the request from the user is derived using at least one of speech recognition and gesture recognition.

11. 11. The AR system of claim 1, configured to update the stored object association data in response to further sensor data generated by the one or more sensors to associate the user with further objects in the environment.

12. the further sensor data includes further image data; said updating said stored object association data includes processing said further image data to learn an appearance of said further object. The AR system of claim 11.

13. 13. The AR system of claim 11 or 12, wherein the updating of the object association data is dependent on input from the user.

14. The AR system of claim 13 , wherein the input from the user is derived using at least one of speech recognition and gesture recognition.

15. The AR system of any one of claims 1 to 14, configured to determine the location of the user using simultaneous localization and mapping, SLAM.

16. 16. The AR system of claim 1, wherein the user interface includes one or more displays, and the AR system is configured to visually output the information on the one or more displays.

17. 1. A computer-implemented method comprising: object association data associating a plurality of objects in an environment with a user of the AR system, the plurality of objects including a first object and a second object, the first object being capable of being located within the second object; and object position data indicating the position of each of the plurality of objects; Remembering the receiving first image data representing a portion of the environment in which the user is located at a first time; determining that the first object is located within the second object by processing the first image data to detect the first object and the second object; receiving second image data representative of a portion of the environment in which the user is located at a second time; and determining a location of the user at the second time; determining a position of the second object in response to the determined position of the user at the second time by detecting the second object in the second image data; updating the stored object position data to indicate the determined position of the second object and an association between the first object and the second object; outputting, via a user interface of the AR system, information in response to the updated stored object position data, the information indicating that the first object is present at the determined position of the second object; and 11. A computer-implemented method comprising:

18. determining the location of the second object includes determining whether the user is currently holding the second object; the updated stored object position data indicates whether the user is currently in possession of the second object; 18. The method of claim 17.

19. Detecting or predicting separation of the second object from the user in response to the determined position of the second object; generating an alert in response to the detected or predicted separation of the second object from the user for output via the user interface of the AR system; and 19. The method of claim 17 or 18, comprising:

20. 20. The method of any one of claims 17 to 19, comprising updating the stored object association data in response to further sensor data generated by the one or more sensors to associate the user with further objects in the environment.

21. the further sensor data includes further image data, and updating the stored object association data includes processing the further image data to learn an appearance of the further object.

21. The method of claim 20.

22. A computer program product comprising machine-readable instructions that, when executed by a computing system, cause the computing system to perform the method of any one of claims 17 to 21.

Images