Patent Application
20150049013
1. Field
The present disclosure relates generally to augmented reality (AR) devices, e.g.,
AR eyeglasses, having optical see-through head mounted displays (HMD) and eye tracking capability, and more particularly, to automatic calibration of the eye tracking capabilities of such AR devices. AR is a technology in which a user's view of the real world is enhanced with additional information generated from a computer model. The enhancements may include labels, 3D rendered models, or shading and illumination changes. AR allows a user to work with and examine the physical real world, while receiving additional information about the objects in it.
2. Background
AR devices typically include an optical see-through HMD and one or more user input mechanisms that allow users to simultaneously see and interact with their surroundings while interacting with applications, such as e-mail and media players. User input mechanisms may include one or more of gesture recognition technology, eye tracking technology, and other similar mechanisms.
AR devices with eye tracking capability may provide for visual activation of applications and selection of files and documents, wherein activation or selection occurs when a user is looking at a displayed object corresponding to the application, file or document. In such AR devices, the displayed object, e.g., application icon, is displayed at a coordinate location of the HMD, while the location on the HMD at which a user is looking or gazing is determined using eye tracking technology. The determined eye gaze location is identified by coordinates of the HMD and may be displayed on the HMD as a point. When the eye gaze coordinates of the user are within range of the displayed object, i.e., either at or sufficiently near the coordinates of the displayed object, activation or selection of the displayed object occurs.
Effective implementation of visual activation of applications and selection of files and documents through eye tracking necessarily depends on the accuracy of the eye tracking technology. Eye tracking accuracy varies depending on environmental factors, such as light conditions, and user factors, such as eye shape and nose height. In cases where eye tracking is inaccurate, the eye gaze location or point determined by the eye tracking technology may not correspond to the location where the user is looking. Accordingly, although a user may be looking at an application icon on the HMD in an attempt to launch the application, the AR device does not recognize the attempt because the eye gaze coordinate output by the eye tracking technology is not within range of the icon.
An AR device may present a point on the HMD corresponding to the eye gaze coordinate. In such instances, the user may become aware of an inaccuracy of the eye tracking technology upon noting that the eye point does not appear where the user is looking, and as a result, initiate calibration of the eye tracking technology. AR devices, however, may not display every eye gaze coordinate in order to reduce power consumption and to minimize display real estate occupation and user distraction. As a result, a user may not be aware that the eye gaze point output by the eye tracking technology does not accurately reflect where the user is looking.
In an aspect of the disclosure, a method, an apparatus, and a computer program product for calibrating an eye tracking system of a head mounted display are provided. An example apparatus displays a moving object in a scene visible through the head mounted display. The object is displayed progressively at a plurality of different points (P) at corresponding different times (T). While the object is at a first point of the plurality of different points in time, the apparatus determines whether an offset between the point P and an eye gaze point (E) satisfies a threshold. The eye-gaze point (E) corresponds to a point where a user is determined to be gazing by the eye tracking system. If the threshold is not satisfied, the apparatus performs a calibration of the eye tracking system when the object is at a second point of the plurality of different points in time. The apparatus then repeats the determining step when the object is at a third point of the plurality of different points in time.
The detailed description set forth below in connection with the appended drawings is intended as a description of various configurations and is not intended to represent the only configurations in which the concepts described herein may be practiced. The detailed description includes specific details for the purpose of providing a thorough understanding of various concepts. However, it will be apparent to those skilled in the art that these concepts may be practiced without these specific details. In some instances, well known structures and components are shown in block diagram form in order to avoid obscuring such concepts.
Several aspects related to automatic calibration of the eye tracking capabilities of AR devices will now be presented with reference to various apparatus and methods. These apparatus and methods will be described in the following detailed description and illustrated in the accompanying drawings by various blocks, modules, components, circuits, steps, processes, algorithms, etc. (collectively referred to as “elements”). These elements may be implemented using electronic hardware, computer software, or any combination thereof. Whether such elements are implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system.
By way of example, an element, or any portion of an element, or any combination of elements may be implemented with a “processing system” that includes one or more processors. Examples of processors include microprocessors, microcontrollers, digital signal processors (DSPs), field programmable gate arrays (FPGAs), programmable logic devices (PLDs), state machines, gated logic, discrete hardware circuits, and other suitable hardware configured to perform the various functionality described throughout this disclosure. One or more processors in the processing system may execute software. Software shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executables, threads of execution, procedures, functions, etc., whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise.
Accordingly, in one or more exemplary embodiments, the functions described may be implemented in hardware, software, firmware, or any combination thereof. If implemented in software, the functions may be stored on or encoded as one or more instructions or code on a computer-readable medium. Computer-readable media includes computer storage media. Storage media may be any available media that can be accessed by a computer. By way of example, and not limitation, such computer-readable media can comprise a random-access memory (RAM), a read-only memory (ROM), an electrically erasable programmable ROM (EEPROM), compact disk ROM (CD-ROM) or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. Disk and disc, as used herein, includes CD, laser disc, optical disc, digital versatile disc (DVD), and floppy disk where disks usually reproduce data magnetically, while discs reproduce data optically with lasers. Combinations of the above should also be included within the scope of computer-readable media.
The processing system 106 and the eye tracking components provide eye tracking capability. Depending on the eye tracking technology being employed, eye tracking components may include one or both of eye cameras and infra-red emitters, e.g. diodes. The processing system 106 and the scene camera 108 provide gesture tracking capability.
The feedback devices 110 provide perception feedback to the user in response to certain interactions with the AR device. Feedback devices 110 may include a speaker or a vibration device. Perception feedback may also be provided by visual indication through the HMD.
The transceiver 112 facilitates wireless communication between the processing system 106 and remote devices, systems or networks. For example, the AR device may communicate with remote servers through the transceiver 112 for purposes of remote processing, such as on-line searches through remote search engines.
As mention above, the AR device 100 allows a user to view real-world scenes through optical see-through HMDs together with content displayed on the HMDs. For example, with reference to
User interaction with the AR device 100 is provided by one or more user input mechanisms, such as a gesture tracking module or an eye-gaze tracking module. Gesture tracking is provided by the scene camera 108 in conjunction with a gesture tracking module of the processing system 106. With gesture tracking, a user may attempt to activate an application by placing his finger on an application icon 204, 206, 208 in the field of view of the AR device. The scene camera 108 captures an image of the finger and sends the image to the gesture tracking module. The gesture tracking module processes the image and determines coordinates of a gesture point corresponding to where the user is pointing. The processing system 106 compares the coordinate location of the gesture point to the coordinate location of the icon on the display. If the locations match, or are within a threshold distance of each other, the processing system 106 determines that the user has selected the icon 204, 206, 208 and accordingly, launches the application.
Eye-gaze tracking is provided by the eye tracking components (not visible) in conjunction with an eye tracking module of the processing system 106. A user may attempt to activate an application by gazing at an application icon 204, 206, 208 in the field of view of the AR device. The eye tracking components capture images of the eyes, and provide the images to the eye tracking module. The eye tracking module processes the images and determines coordinates of an eye-gaze point corresponding to where the user is looking. The processing system 106 compares the coordinate location of the eye-gaze point to the coordinate location of the icon on the display. If the locations match, or are within a threshold distance of each other, the processing system 106 determines that the user has selected the icon 204, 206, 208 and accordingly, launches the application. Often, such eye-gaze based launching is coupled with another form of input, e.g., gesture, to confirm the user's intention of launching the application.
The AR device 300 includes an on-board processing system 310, which in turn includes one or more of an eye tracking module 312 and a gesture tracking module 314. As described further below, an object selection module 316 processes the outputs of the one or more tracking modules to determine user interactions and tracking module accuracy. A tracking calibration module 318 calibrates the one or more tracking modules if the tracking module is determined to be inaccurate.
The on-board processing system 310 may also include a scene camera calibration module 320, a graphical user interface (GUI) adjustment module 322, and a perception feedback module 324. The scene camera calibration module 320 calibrates the AR device so that the AR content is aligned with real world objects. The GUI adjustment module 322 may adjust the parameters of GUI objects displayed on the HMD to compensate for eye-tracking or gesture-tracking inaccuracies detected by the object selection module 316. Such adjustments may precede, supplement, or substitute for the actions of the tracking calibration module 318. The feedback module 324 controls one or more feedback devices 326 to provide perception feedback to the user in response to one or more types of user interactions. For example, the feedback module a feedback device 326 to output sound when a user selects an icon in the field of view using a gesture or eye gaze.
The AR device 300 further includes memory 328 for storing program code to implement the foregoing features of the on-board processing system 310. A communications module 330 and transceiver 332 facilitate wireless communications with remote devices, systems and networks. For example, in one implementation, an image of a real-world object may be captured by the scene camera 308 and transmitted by the communications module 330 and the transceiver 332 to a remote search engine, with subsequent search results being received by the transceiver.
With further respect to eye tracking capability, the diodes 304 and eye cameras 306, together with the eye tracking module 312, provide eye tracking capability as generally described above. In the example implementation of
The scene camera 308, together with the gesture tracking module 314, provide gesture tracking capability using a known technology as generally described above. In the example implementation of
The object selection processor 316 functions to determine whether interactions of the user, as characterized by one or more of the eye tracking module 312 and the gesture tracking module 314, correspond to a selection of an object, e.g., application icon, displayed on the HMD 302 and visible in the field of view. If an interaction does correspond to a selection by the user, for example, a selection of an icon to launch an application 334, the object selection processor 316 outputs a command to the application.
With respect to object selection based on eye gaze, the object selection processor 316 receives eye gaze coordinates representing an eye gaze point (E) from the eye tracking module 312. In one configuration, the eye gaze coordinates output by the eye tracking module 312 are with respect to the origin and direction of an eye coordinate system. The eye coordinate system is defined by the eye tracking module 312 and usually has an origin corresponding to the center of the user's respective right or left eye. In this configuration, the object selection processor 316 transforms the eye gaze coordinates to a screen coordinate system, and optionally outputs the transformed eye gaze coordinates as an eye gaze point (E) for display on the HMD. The screen coordinate system is defined by the HMD and usually has an origin corresponding to the center of the respective right or left HMD. In another configuration, the transformation of the eye gaze coordinate to the screen coordinate system of the HMD may be performed by the eye tracking module 306.
The object selection processor 316 compares the eye gaze coordinates of eye gaze point (E) to the object coordinates of an object point (P), for example, by determining the distance between the point (E) and point (P). The object point (P) may correspond to the center of an application icon displayed on the HMD. If the eye gaze point (E) is determined to be at or near the object point (P), then the object selection processor 316 determines that a selection has occurred, and an appropriate action follows. For example, if the object point (P) represents and application icon, the command may open the application.
With respect to object selection based on gesture, the object selection processor 316 receives gesture coordinates representing a gesture point (G) from the gesture tracking module 314. In one configuration, the gesture coordinates output by the gesture tracking module 314 are with respect to the origin and direction of gesture coordinate system. The gesture coordinate system is defined by the gesture tracking module 314 and usually has an origin corresponding to the center of the scene camera 308. In this configuration, the object selection processor 316 transforms the gesture coordinates to a screen coordinate system, and optionally outputs the transformed gesture coordinates as gesture point (G) for display on the HMD. The screen coordinate system is defined by the HMD and usually has an origin corresponding to the center of the respective right or left HMD. In another configuration, the transformation of the gesture coordinate to the screen coordinate system of the HMD may be performed by the gesture tracking module 314.
The object selection processor 316 compares the gesture coordinates of gesture point (G) to the object coordinates of an object point (P), for example, by determining the distance between the point (G) and point (P). The object point (P) may correspond to the center of an application icon displayed on the HMD. If the gesture point (G) is determined to be at or near the object point (P), then the object selection processor 316 determines that a selection has occurred, and an appropriate action follows. For example, if the object point (P) represents and application icon, the command may open the application.
As mentioned above, the accuracy of eye tracking technology varies depending on environmental factors, such as light conditions, and user factors, such as eye shape and nose height. Inaccurate eye tracking may result in poor user experience in that a user's attempt to activate an application may go undetected by the processing system 106.
There are two problems with the foregoing calibration process: First, the calibration process is user initiated in that the user uses some input technique, such as a gesture detectable by the gesture recognition system or a touch detectable by a touch sensor to tell the system that she is looking into the target dot and ready for calibration. Second, completion of the calibration process is user dependent in that the user decides whether she is satisfied with the calibration accuracy or not. If the user is not satisfied, the calibration process need to be is repeated. On the other hand, if the user goes forward with poor eye tracking accuracy, the eye tracking based user input system may not work well, especially when eye tracking is used for triggering some events, thereby resulting in poor user experience. Disclosed herein is a technique which removes the above two problems and ensures good eye tracking accuracy before the user moves to next phase in the system.
The object is displayed progressively at a plurality of different object points (P) on the HMD at corresponding different times (T). Progressively, in this context means that an object is displayed in a manner so as to present to the wearer of the AR device a primary point of focus. For example, in one implementation, only one object is displayed on the scene at a time so as to avoid distracting the user with multiple points of focus. The object is displayed at a location for a brief period of time, then reappears after another brief period of time at a different location. Depending on the periods of time, the object may appear to be progressing, or moving across the HMD.
Returning to
The AR device may determine the offset based coordinates of the accuracy point (A) 702 and the eye gaze point (E) 706. For example, the HMD may be characterized by a screen coordinate system. Such as a Cartesian coordinate system having an origin at the center of the HMD. The accuracy point (A) 702 may be characterized by coordinates that represent a location of the object within the screen coordinate system. Likewise, the eye gaze point (E) 706 may be characterized by coordinates that represent a location of the eye gaze within the screen coordinate system. In this case, the AR device derives the offset, e.g., the distance between the points, from the coordinates of the accuracy point (A) and the eye gaze point (E).
At step 606, the AR device determines if the threshold is satisfied. The threshold may be satisfied when the offset between the accuracy point (A) 702 and the eye-gaze point (E) 706 is less than a predetermined value. In one implementation, the predetermined values range from 0 to 10 pixels.
If the AR device determines that the threshold is satisfied, the process stops. If the threshold is not satisfied, then at step 608, the AR device performs a calibration of the eye tracking module for a calibration point (C) 704 corresponding to one of the plurality of different object points (P). The calibration point (C) 704 and accuracy point (A) 702 may be different object points (P).
In an example calibration, the AR device obtains coordinates of the calibration point (C) that represent a location of the object within the screen coordinate system of the HMD, and coordinates of the eye gaze point (E) that also represent a location of the eye gaze within the screen coordinate system. The AR device then determines an offset between the calibration point (C) and the eye gaze point (E) based on the respective coordinates, and calibrates the eye tracking module by adjusting the coordinates of subsequent eye gaze points (E) output by the eye tracking module based on the offset. For example, if the offset between the calibration point (C) and the eye gaze point (E) along the x coordinate of the HMD display is 10 pixels, then the eye tracking module adjusts the x coordinate output of subsequent eye gaze points (E) by 10 pixels.
After calibration, the process returns to step 604, where the AR device determines whether an offset between a subsequent accuracy point (A) 708 and a subsequent eye gaze point (E) 710 satisfies a threshold. The subsequent accuracy point (A) 708 corresponds to one of the plurality of different object points (P). If the threshold is not satisfied at step 606, then the calibration of step 608 is performed at a calibration point (C) 712 corresponding to one of the plurality of different object points (P). The 604. 606 and 608 may be repeated until the threshold is satisfied.
The moving object module 804 displays a moving object in a scene visible through a head mounted display module 806. The moving object module is configured to display the object progressively at a plurality of different points (P) at corresponding different times (T). The offset determination module 806 determines, at a first point of the plurality of different points in time, whether an offset between the point P and an eye gaze point (E) satisfies a threshold. The eye-gaze point (E) corresponds to a point where a user is determined to be gazing by the eye tracking system module 810. The tracking calibration module 812 performs a calibration of the eye tracking system at a second point of the plurality of different points in time if the threshold is not satisfied. The offset determination module 808 determines, at a third point of the plurality of different points in time, whether an offset between the point P and an eye gaze point (E) satisfies a threshold.
The AR devices, as illustrated in
The processing system 914 includes a processor 904 coupled to a computer-readable medium/memory 906. The processor 904 is responsible for general processing, including the execution of software stored on the computer-readable medium/memory 906. The software, when executed by the processor 904, causes the processing system 914 to perform the various functions described supra for any particular apparatus. The computer-readable medium/memory 906 may also be used for storing data that is manipulated by the processor 904 when executing software. The processing system further includes at least one of the modules 804, 806, 808 and 810. The modules may be software modules running in the processor 904, resident/stored in the computer readable medium/memory 906, one or more hardware modules coupled to the processor 904, or some combination thereof.
In one configuration, the apparatus 802/802′ includes means for automatically displaying a moving object in a scene visible through the head mounted display, where the object is displayed progressively at a plurality of different points (P) at corresponding different times (T). The apparatus 802/802′ also includes means for, at a first point of the plurality of different points in time, determining whether an offset between the point P and an eye gaze point (E) satisfies a threshold, wherein the eye-gaze point (E) corresponds to a point where a user is determined to be gazing by the eye tracking system. The apparatus 802/802′ further includes means for performing a calibration of the eye tracking system at a second point of the plurality of different points in time, if the threshold is not satisfied, and means for repeating the determining step at a third point of the plurality of different points in time. The aforementioned means may be one or more of the aforementioned modules of the apparatus 802 and/or the processing system 914 of the apparatus 802′ configured to perform the functions recited by the aforementioned means.
Returning to
The eye gaze tracker 804 monitors the eye gaze of the user while the user is gazing at a first point P1 of the plurality of different points in time. For example, with reference to
This eye gaze point (E) may or may not be displayed in the scene. In either case, the processor 808 then determines whether an offset between the point P1 and the eye gaze point (E) satisfies a threshold. If the threshold is not satisfied, the calibrator 810 performs a calibration of the eye tracking system. The calibration is performed at a second point P2 of the plurality of different points in time. For example, with reference to
After calibration, the system performs an accuracy check at a third point P3 of the plurality of different points in time. To this end, the eye gaze tracker 804 monitors the eye gaze of the user while the user is gazing at a first point P3. The eye tracker 804 determines an eye-gaze point (E) where the user is gazing. The processor 808 then determines whether an offset between the point P3 and the eye gaze point (E) satisfies the threshold. The foregoing accuracy check and calibration are repeated at successive points until the threshold is satisfied.
In the eye tracking calibration process disclosed herein, a user is presented with a dynamically moving dot and asked to follow it. For example, at time tn, the virtual dot is presented in Pn position, at time tn+1, the virtual dot is presented in Pn+1 position and so on. At time tn, the eye tracking system reports users gaze point En. The system automatically initiates calibration at time tn with the dot in Pn position and eye tracking model reported dot En position and it measures the calibration offset f at time tn+1 with Pn+1 and En+1. The offset is defined
f=abs(dist(Pn+1,En+1))
If the offset is less than predefined threshold, calibration is done; otherwise, it initiates calibration in the next point Pn+1+k with eye tracking reported point En+1+k and repeat the same method.
It is understood that the specific order or hierarchy of steps in the processes disclosed is an illustration of exemplary approaches. Based upon design preferences, it is understood that the specific order or hierarchy of steps in the processes may be rearranged. Further, some steps may be combined or omitted. The accompanying method claims present elements of the various steps in a sample order, and are not meant to be limited to the specific order or hierarchy presented.
The previous description is provided to enable any person skilled in the art to practice the various aspects described herein. Various modifications to these aspects will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other aspects. Thus, the claims are not intended to be limited to the aspects shown herein, but is to be accorded the full scope consistent with the language claims, wherein reference to an element in the singular is not intended to mean “one and only one” unless specifically so stated, but rather “one or more.” The word “exemplary” is used herein to mean “serving as an example, instance, or illustration.” Any aspect described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other aspects.” Unless specifically stated otherwise, the term “some” refers to one or more. Combinations such as “at least one of A, B, or C,” “at least one of A, B, and C,” and “A, B, C, or any combination thereof” include any combination of A, B, and/or C, and may include multiples of A, multiples of B, or multiples of C. Specifically, combinations such as “at least one of A, B, or C,” “at least one of A, B, and C,” and “A, B, C, or any combination thereof” may be A only, B only, C only, A and B, A and C, B and C, or A and B and C, where any such combinations may contain one or more member or members of A, B, or C. All structural and functional equivalents to the elements of the various aspects described throughout this disclosure that are known or later come to be known to those of ordinary skill in the art are expressly incorporated herein by reference and are intended to be encompassed by the claims. Moreover, nothing disclosed herein is intended to be dedicated to the public regardless of whether such disclosure is explicitly recited in the claims. No claim element is to be construed as a means plus function unless the element is expressly recited using the phrase “means for.”
This application claims the benefit of U.S. Provisional Application Ser. No. 61/867,569, entitled “Eye Tracking Auto Calibration for Optical See-Through HMD With Augmented Reality and Eye Tracking” and filed on Aug. 19, 2013, which is expressly incorporated by reference herein in its entirety.
| Number | Date | Country | |
|---|---|---|---|
| 61867569 | Aug 2013 | US |
