Augmented or mixed reality is a technology that allows virtual imagery to be mixed with a user's actual view of the real world. A near-eye display may be worn by a user to view the mixed imagery of virtual and real objects. A near-eye display displays virtual imagery in the user's field of view. However, the user's field of view is not stationary as a user moves his or her head. Furthermore, what the user is looking at in the field of view changes as the user shifts his or her eyes, even if his or her head does not move.
The technology provides various embodiments for gaze determination within a see-through, near-eye mixed reality display device. Gaze is sometimes referred to as a line of sight from the user's eye to an object, real or virtual, at which the user is looking. Embodiments are provided for determining gaze based on glint data or a combination of both glint and image data. In some embodiments, a gaze determination coordinate system based on predetermined positioning of at least one light sensor and illuminators on the display device provides a three dimensional (3D) spatial relationship between the display device and each respective eye. A gaze vector for each eye may be determined based on the 3D spatial relationship. Based on gaze vectors for both eyes, a point of gaze may be determined which indicates one or more objects, real or virtual, at which a user is gazing, or more commonly stated as, at which the user is looking.
The technology provides an embodiment of a mixed reality display system with gaze determination. The system embodiment comprises a see-through, near-eye display device including a respective display optical system for each eye positioned to be seen through by the respective eye. For each eye, an image generation unit is attached to the see-through display device for generating at least one virtual image for display. Additionally, the embodiment includes a respective arrangement of gaze detection elements positioned on the display device for forming a spatial relationship between the gaze detection elements and each eye. The gaze detection elements include a set of illuminators for generating glints on the respective eye, and at least one respective sensor for capturing light reflected from the respective eye and generating data representing the captured reflected light. The system embodiment includes memory for storing software and the data. A processor is communicatively coupled to the at least one respective sensor for receiving and storing the data representing the captured reflected light in the accessible memory. The processor determines a gaze vector for each respective eye based on the data representing the captured reflected light and a point of gaze based on the gaze vectors in a three-dimensional (3D) user field of view.
The technology provides an embodiment of a method for determining gaze in a see-through, near-eye mixed reality display system. The method embodiment comprising determining boundaries of a gaze detection coordinate system based on positions of glints detected on a user eye, positions on the near-eye display system of illuminators for generating the glints; and a position of at least one sensor for detecting the glints. The method further comprises determining a gaze vector for each eye based on reflected eye data including the glints; determining a point of gaze based on the gaze vectors for the two eyes in a three-dimensional (3D) user field of view including real and virtual objects, and identifying any object at the point of gaze in the 3D user field of view.
Another system embodiment is also provided by the technology for a mixed reality display system with gaze determination based on glints. The system comprises a see-through, near-eye display device including a respective display optical system for each eye positioned to be seen through by the respective eye and a respective image generation unit for each eye attached to the see-through display device for generating at least one virtual image for display. The system further comprises a set of infra-red (IR) illuminators for producing glints for each eye and at least one respective sensor is positioned on the near-eye display device at a predetermined position to detect the glints and generate glint data to be stored in a memory accessible by a processor which determines a gaze vector for each eye based on the glint data. A point of gaze in a three-dimensional user field of view is determined by the software controlled processor based on the gaze vectors.
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, nor is it intended to be used as an aid in determining the scope of the claimed subject matter.
a is a flowchart illustrating a method embodiment for generating a set of training data sets for a comparison based determination of gaze.
b is a flowchart illustrating a method embodiment for determining gaze based on the training data sets.
c is a flowchart of an interpolation method embodiment which may be used with the comparison step of
The technology provides various embodiments for gaze determination within a see-through, near-eye, mixed reality display device. Gaze is sometimes referred to as a line of sight or a visual axis of an eye. The visual axis extends from the fovea, sometimes referred to as the foveal centralis, of the retina through the center of the pupil. Extending the visual axis from the fovea through the pupil and a see-through lens for each eye, one can determine a point of gaze in a user's field of view which may include images of virtual objects, and an actual direct view of the real world.
The use of the term “actual direct view” refers to the ability to see real world objects directly with the human eye, rather than seeing created image representations of the objects. For example, looking through glass at a room allows a user to have an actual direct view of the room, while viewing a video of a room on a television is not an actual direct view of the room. Based on the context of executing software, for example, a gaming application, the system can project images of virtual objects, sometimes referred to as virtual images, on the display that are viewable by the person wearing the see-through display device while that person is also viewing real world objects through the display.
Geometry of one or more gaze detection elements with respect to the visible portion of a human eye forms a basis for various embodiments of gaze determination. In some embodiments, data from only glints may be used to track changing intensities due to different reflectivities on parts of the eye like the sclera, sometimes referred to as the white section of the eye, the pupil and the iris. A glint is a very small and often very bright reflection of light from a light source off of a surface of the cornea of an eye. The glint is an image of the light source, typically a narrow beam source focused on the eye. In some embodiments, a training or calibration gaze data set for the glints may be used in comparisons for detecting a current gaze.
Other embodiments use both image data of the eye and data representing glints in the context of a geometry of the illuminators and at least one image sensor to determine boundaries of a three-dimensional (3D) spatial relationship between positions of parts of the eye and a respective system of gaze detection elements. Examples of such parts of the eye are a center of a cornea determined based on glint data, a center of a pupil determined from image data of an eye, and a center of rotation of the eye a position of which is estimated based on the position of the cornea center. For accuracy considerations in gaze determination purposes, the center of rotation of the eyeball may be considered fixed. A gaze vector for the respective eye is determined based on the cornea center, pupil center, and center of rotation which form an optical axis for the respective eye. An angle offset may be applied to the optical axis in order to obtain a visual axis for the eye which may be selected as the gaze vector.
Different gaze detection techniques may be used within the same system. For example, due to obstructions of the eye or update processing time, less computationally intensive techniques, like a version of the approach based on correlating glint intensity values with pupil position, may be used more frequently in combination with more computationally intensive techniques run with longer time intervals in between like a version of determining the gaze vector based on the 3D spatial relationship between the cornea center, pupil center, center of rotation and a gaze detection system of optical elements. Changes in the spatial relationship including depth changes between the eye and the gaze detection elements can be determined also as an indicator triggering recalibration of the system, for example in embodiments using a training gaze data set.
In the embodiments discussed below, the see-through display device is in a set of eyeglasses but other head mounted display (HMD) formats and near-eye display holders suitable for consumer, everyday use can be used as well.
Head mounted display device 2, which in one embodiment is in the shape of eyeglasses in a frame 115, is worn on the head of a user so that the user can see through a display, embodied in this example as a display optical system 14 for each eye, and thereby have an actual direct view of the space in front of the user. Frame 115 provides a support for holding elements of the system in place as well as a conduit for electrical connections. In this embodiment, frame 115 provides a convenient eyeglass frame as support for the elements of the system discussed further below. In other embodiments, other support structures can be used. An example of such a structure is a visor or goggles. The frame 115 includes a temple or side arm for resting on each of a user's ears. Temple 102 is representative of an embodiment of the right temple. Nose bridge 104 of the frame includes a microphone 110 for recording sounds and transmitting audio data to processing unit 4.
Hub computing system 12 may be a computer, a gaming system or console, or the like. According to an example embodiment, the hub computing system 12 may include hardware components and/or software components such that hub computing system 12 may be used to execute applications such as gaming applications, non-gaming applications, or the like. In one embodiment, hub computing system 12 may include a processor such as a standardized processor, a specialized processor, a microprocessor, or the like that may execute instructions stored on a processor readable storage device for performing the processes described herein.
Hub computing system 12 further includes one or more capture devices, such as capture devices 20A and 20B. In other embodiments, more or less than two capture devices can be used to capture the room or other physical environment of the user. Capture devices 20A and 20B may be, for example, cameras that visually monitor one or more users and the surrounding space such that gestures and/or movements performed by the one or more users, as well as the structure of the surrounding space, may be captured, analyzed, and tracked to perform one or more controls or actions within an application and/or animate an avatar or on-screen character. An application may be executing on hub computing system 12, the display device 2, as discussed below on a mobile device 5 or a combination of these.
Hub computing system 12 may be connected to an audiovisual device 16 such as a television, a monitor, a high-definition television (HDTV), or the like that may provide game or application visuals. In some instances, the audiovisual device 16 may be a three-dimensional display device. For example, hub computing system 12 may include a video adapter such as a graphics card and/or an audio adapter such as a sound card that may provide audiovisual signals associated with the game application, non-game application, etc. The audiovisual device 16 may receive the audiovisual signals from hub computing system 12 and may then output the game or application visuals and/or audio associated with the audiovisual signals. According to one embodiment, the audiovisual device 16 may be connected to hub computing system 12 via, for example, an S-Video cable, a coaxial cable, an HDMI cable, a DVI cable, a VGA cable, component video cable, RCA cables, etc. In one example, audiovisual device 16 includes internal speakers. In other embodiments, audiovisual device 16, a separate stereo or hub computing system 12 is connected to external speakers 22.
Furthermore, as in the hub computing system 12, gaming and non-gaming applications may execute on a processor of the mobile device 5 which user actions control or which user actions animate an avatar as may be displayed on a display 7 of the device 5. The mobile device 5 also provides a network interface for communicating with other computing devices like hub computing system 12 over the Internet or another communication network via a wired or wireless communication medium using a wired or wireless communication protocol. A remote network accessible computer system like hub computing system 12 may be leveraged for processing power and remote data access by a processing unit 4 like mobile device 5. Examples of hardware and software components of a mobile device 5 such as may be embodied in a smartphone or tablet computing device are described in
As noted above, in some embodiments, gaze detection of each of a user's eyes is based on a three dimensional coordinate system of gaze detection elements on a near-eye, mixed reality display device like the eyeglasses 2 in relation to one or more human eye elements such as a cornea center, a center of eyeball rotation and a pupil center. Examples of gaze detection elements which may be part of the coordinate system including glint generating illuminators and at least one sensor for capturing data representing the generated glints. As discussed in the embodiment of
In the example of
In one example, a visible light camera also commonly referred to as an RGB camera may be the sensor, and an example of an optical element or light directing element is a visible light reflecting mirror which is partially transmissive and partially reflective. The visible light camera provides image data of the pupil of the user's eye, while IR photodetectors 152 capture glints which are reflections in the IR portion of the spectrum. If a visible light camera is used, reflections of virtual images may appear in the eye data captured by the camera. An image filtering technique may be used to remove the virtual image reflections if desired. An IR camera is not sensitive to the virtual image reflections on the eye.
In other examples, the at least one sensor 134 is an IR camera or a position sensitive detector (PSD) to which IR radiation may be directed. For example, a hot reflecting surface may transmit visible light but reflect IR radiation. The IR radiation reflected from the eye may be from incident radiation of the illuminators 153, other IR illuminators (not shown) or from ambient IR radiation reflected off the eye. In some examples, sensor 134 may be a combination of an RGB and an IR camera, and the optical light directing elements may include a visible light reflecting or diverting element and an IR radiation reflecting or diverting element. In some examples, a camera may be small, e.g. 2 millimeters (mm) by 2 mm. An example of such a camera sensor is the Omnivision OV7727. In other examples, the camera may be small enough, e.g. the Omnivision OV7727, e.g. that the image sensor or camera 134 may be centered on the optical axis or other location of the display optical system 14. For example, the camera 134 may be embedded within a lens of the system 14. Additionally, an image filtering technique may be applied to blend the camera into a user field of view to lessen any distraction to the user.
In the example of
As mentioned above, in some embodiments which calculate a cornea center as part of determining a gaze vector, two glints, and therefore two illuminators will suffice. However, other embodiments may use additional glints in determining a pupil position and hence a gaze vector. As eye data representing the glints is repeatedly captured, for example at 30 frames a second or greater, data for one glint may be blocked by an eyelid or even an eyelash, but data may be gathered by a glint generated by another illuminator.
Control circuits 136 provide various electronics that support the other components of head mounted display device 2. More details of control circuits 136 are provided below with respect to
The display device 2 provides an image generation unit which can create one or more images including one or more virtual objects. In some embodiments, a microdisplay may be used as the image generation unit. A microdisplay assembly 173 comprises light processing elements and a variable focus adjuster 135. An example of a light processing element is a microdisplay unit 120. Other examples include one or more optical elements such as one or more lenses of a lens system 122 and one or more reflecting elements such as surfaces 124a and 124b in
Mounted to or inside temple 102, the microdisplay unit 120 includes an image source and generates an image of a virtual object. The microdisplay unit 120 is optically aligned with the lens system 122 and the reflecting surface 124 or reflecting surfaces 124a and 124b as illustrated in the following figures. The optical alignment may be along an optical axis 133 or an optical path 133 including one or more optical axes. The microdisplay unit 120 projects the image of the virtual object through lens system 122, which may direct the image light, onto reflecting element 124 which directs the light into lightguide optical element 112 as in
The variable focus adjuster 135 changes the displacement between one or more light processing elements in the optical path of the microdisplay assembly or an optical power of an element in the microdisplay assembly. The optical power of a lens is defined as the reciprocal of its focal length, e.g. 1/focal length, so a change in one effects the other. The change in focal length results in a change in the region of the field of view, e.g. a region at a certain distance, which is in focus for an image generated by the microdisplay assembly 173.
In one example of the microdisplay assembly 173 making displacement changes, the displacement changes are guided within an armature 137 supporting at least one light processing element such as the lens system 122 and the microdisplay 120 in this example. The armature 137 helps stabilize the alignment along the optical path 133 during physical movement of the elements to achieve a selected displacement or optical power. In some examples, the adjuster 135 may move one or more optical elements such as a lens in lens system 122 within the armature 137. In other examples, the armature may have grooves or space in the area around a light processing element so it slides over the element, for example, microdisplay 120, without moving the light processing element. Another element in the armature such as the lens system 122 is attached so that the system 122 or a lens within slides or moves with the moving armature 137. The displacement range is typically on the order of a few millimeters (mm). In one example, the range is 1-2 mm. In other examples, the armature 137 may provide support to the lens system 122 for focal adjustment techniques involving adjustment of other physical parameters than displacement. An example of such a parameter is polarization.
For more information on adjusting a focal distance of a microdisplay assembly, see U.S. patent Ser. No. 12/941,825 entitled “Automatic Variable Virtual Focus for Augmented Reality Displays,” filed Nov. 8, 2010, having inventors Avi Bar-Zeev and John Lewis and which is hereby incorporated by reference.
In one example, the adjuster 135 may be an actuator such as a piezoelectric motor. Other technologies for the actuator may also be used and some examples of such technologies are a voice coil formed of a coil and a permanent magnet, a magnetostriction element, and an electrostriction element.
There are different image generation technologies that can be used to implement microdisplay 120. For example, microdisplay 120 can be implemented using a transmissive projection technology where the light source is modulated by optically active material, backlit with white light. These technologies are usually implemented using LCD type displays with powerful backlights and high optical energy densities. Microdisplay 120 can also be implemented using a reflective technology for which external light is reflected and modulated by an optically active material. The illumination is forward lit by either a white source or RGB source, depending on the technology. Digital light processing (DLP), liquid crystal on silicon (LCOS) and Mirasol® display technology from Qualcomm, Inc. are all examples of reflective technologies which are efficient as most energy is reflected away from the modulated structure and may be used in the system described herein. Additionally, microdisplay 120 can be implemented using an emissive technology where light is generated by the display. For example, a PicoP™ engine from Microvision, Inc. emits a laser signal with a micro mirror steering either onto a tiny screen that acts as a transmissive element or beamed directly into the eye (e.g., laser).
As mentioned above, the configuration of the light processing elements of the microdisplay assembly 173 create a focal distance or focal region in which a virtual object appears in an image. Changing the configuration changes the focal region for the virtual object image. The focal region determined by the light processing elements can be determined and changed based on the equation 1/S1+1/S2=1/f.
The symbol f represents the focal length of a lens such as lens system 122 in the microdisplay assembly 173. The lens system 122 has a front nodal point and a rear nodal point. If light rays are directed toward either nodal point at a given angle relative to the optical axis, the light rays will emerge from the other nodal point at an equivalent angle relative to the optical axis. In one example, the rear nodal point of lens system 122 would be between itself and the microdisplay 120. The distance from the rear nodal point to the microdisplay 120 may be denoted as S2. The front nodal point is typically within a few mm of lens system 122. The target location is the location of the virtual image to be generated by the microdisplay 120 in a three-dimensional physical space. The distance from the front nodal point to the target location of the virtual image may be denoted as S1. Since the image is to be a virtual image appearing on the same side of the lens as the microdisplay 120, sign conventions give that S1 has a negative value.
If the focal length of the lens is fixed, S1 and S2 are varied to focus virtual objects at different depths. For example, an initial position may have S1 set to infinity, and S2 equal to the focal length of lens system 122. Assuming lens system 122 has a focal length of 10 mm, consider an example in which the virtual object is to be placed about 1 foot or 300 mm into the user's field of view. S1 is now about −300 mm, f is 10 mm and S2 is set currently at the initial position of the focal length, 10 mm, meaning the rear nodal point of lens system 122 is 10 mm from the microdisplay 120. The new distance or new displacement between the lens 122 and microdisplay 120 is determined based on 1/(−300)+1/S2=1/10 with all in units of mm. The result is about 9.67 mm for S2.
In one example, one or more processors such as in the control circuitry, the processing unit 4, 5 or both can calculate the displacement values for S1 and S2, leaving the focal length f fixed and cause the control circuitry 136 to cause a variable adjuster driver 237 (see
The display optical system 14 in this embodiment has an optical axis 142 and includes a see-through lens 118 allowing the user an actual direct view of the real world. In this example, the see-through lens 118 is a standard lens used in eye glasses and can be made to any prescription (including no prescription). In another embodiment, see-through lens 118 can be replaced by a variable prescription lens. In some embodiments, see-through, near-eye display device 2 will include additional lenses.
The display optical system 14 further comprises reflecting surfaces 124a and 124b. In this embodiment, light from the microdisplay 120 is directed along optical path 133 via a reflecting element 124a to a partially reflective element 124b embedded in lens 118 which combines the virtual object image view traveling along optical path 133 with the natural or actual direct view along the optical axis 142 so that the combined views are directed into a user's eye, right one in this example, at the optical axis, the position with the most collimated light for a clearest view.
A detection area 139r of a light sensor is also part of the display optical system 14r. An optical element 125 embodies the detection area 139r by capturing reflected light from the user's eye received along the optical axis 142 and directs the captured light to the sensor 134r, in this example positioned in the bridge 104. As shown, the arrangement allows the detection area 139 of the sensor 134r to have its center aligned with the center of the display optical system 14. For example, if sensor 134r is an image sensor, sensor 134r captures the detection area 139, so an image captured at the image sensor is centered on the optical axis because the detection area 139 is. In one example, sensor 134r is a visible light camera or a combination of RGB/IR camera, and the optical element 125 includes an optical element which reflects visible light reflected from the user's eye, for example a partially reflective mirror.
In other embodiments, the sensor 134r is an IR sensitive device such as an IR camera, and the element 125 includes a hot reflecting surface which lets visible light pass through it and reflects IR radiation to the sensor 134r. An IR camera may capture not only glints, but also an infra-red or near-infra-red image of the user's eye including the pupil.
In other embodiments, the IR sensor device 134r is a position sensitive device (PSD), sometimes referred to as an optical position sensor. The position of detected light on the surface of the sensor is identified. A PSD can be selected which is sensitive to a wavelength range or about a predetermined wavelength of IR illuminators for the glints. When light within the wavelength range or about the predetermined wavelength of the position sensitive device is detected on the sensor or light sensitive portion of the device, an electrical signal is generated which identifies the location on the surface of the detector. In some embodiments, the surface of a PSD is divided into discrete sensors like pixels from which the location of the light can be determined. In other examples, a PSD isotropic sensor may be used in which a change in local resistance on the surface can be used to identify the location of the light spot on the PSD. Other embodiments of PSDs may also be used. By operating the illuminators 153 in a predetermined sequence, the location of the reflection of glints on the PSD can be identified and hence related back to their location on a cornea surface.
The depiction of the light directing elements, in this case reflecting elements, 125, 124, 124a and 124b in
The display optical system 14 includes other gaze detection elements in this embodiment. In this embodiment, attached to frame 115 and on the sides of lens 118, are at least two (2) but may be more, infra-red (IR) illuminating devices 153 which direct narrow infra-red light beams within a particular wavelength range or about a predetermined wavelength at the user's eye to each generate a respective glint on a surface of the respective cornea. In other embodiments, the illuminators and any photodiodes may be on the lenses, for example at the corners or edges. In this embodiment, in addition to the at least 2 infra-red (IR) illuminating device 153 are IR photodetectors 152. Each photodetector 152 is sensitive to IR radiation within the particular wavelength range of its corresponding IR illuminator 153 across the lens 118 and is positioned to detect a respective glint. As shown in
In the embodiment of
In one embodiment, glint reflections can estimate gaze based on a few data points of the intensity values detected for the glints, rather than processing much, much larger sets of image data of eyes. The position of the illuminators 153 on the eyeglass frame 115 or other support structure of a near-eye display device may be fixed so that the position of glints detected by one or more sensors is fixed in the sensor detection area. The cornea and hence the iris and the pupil rotate with the eyeball about a fixed center. The iris, pupil, and the sclera which is sometimes referred to as the white portion of the eyeball, move underneath the glint as the user's gaze changes. So a glint detected at a same sensor location may result in different intensity values due to different reflectivities associated with the different eye parts. As the pupil is a hole with tissue that absorbs most incoming light, the intensity value for it would be very low or near zero, while that for the iris would be a higher intensity value due to its higher reflectivity. An intensity value for the sclera may be highest as the sclera has the highest reflectivity. In some examples, an illuminator may be positioned as in
Lightguide optical element 112 transmits light from microdisplay 120 to the eye of the user wearing head mounted display device 2. Lightguide optical element 112 also allows light from in front of the head mounted display device 2 to be transmitted through lightguide optical element 112 to the user's eye thereby allowing the user to have an actual direct view of the space in front of head mounted display device 2 in addition to receiving a virtual image from microdisplay 120. Thus, the walls of lightguide optical element 112 are see-through. Lightguide optical element 112 includes a first reflecting surface 124 (e.g., a mirror or other surface). Light from microdisplay 120 passes through lens 122 and becomes incident on reflecting surface 124. The reflecting surface 124 reflects the incident light from the microdisplay 120 such that light is trapped inside a planar, substrate comprising lightguide optical element 112 by internal reflection.
After several reflections off the surfaces of the substrate, the trapped light waves reach an array of selectively reflecting surfaces 126. Note that only one of the five surfaces is labeled 126 to prevent over-crowding of the drawing. Reflecting surfaces 126 couple the light waves incident upon those reflecting surfaces out of the substrate into the eye of the user. More details of a lightguide optical element can be found in United States Patent Application Publication 2008/0285140, Ser. No. 12/214,366, published on Nov. 20, 2008, “Substrate-Guided Optical Devices” incorporated herein by reference in its entirety.
In this embodiment, as in
In the embodiments of
In the embodiments above, the specific number of lenses shown are just examples. Other numbers and configurations of lenses operating on the same principles may be used. Additionally, in the examples above, only the right side of the see-through, near-eye display 2 are shown. A full near-eye, mixed reality display device would include as examples another set of lenses 116 and/or 118, another lightguide optical element 112 for the embodiments of
Note that some of the components of
Camera interface 216 provides an interface to the two physical environment facing cameras 113 and each eye camera 134 and stores respective images received from the cameras 113, 134 in camera buffer 218. Display driver 220 will drive microdisplay 120. Display formatter 222 may provide information, about the virtual image being displayed on microdisplay 120 to one or more processors of one or more computer systems, e.g. 4, 12, 210 performing processing for the augmented reality system. Timing generator 226 is used to provide timing data for the system. Display out 228 is a buffer for providing images from physical environment facing cameras 113 and the eye cameras 134 to the processing unit 4. Display in 230 is a buffer for receiving images such as a virtual image to be displayed on microdisplay 120. Display out 228 and display in 230 communicate with band interface 232 which is an interface to processing unit 4.
Power management circuit 202 includes voltage regulator 234, eye tracking illumination driver 236, variable adjuster driver 237, photodetector interface 239, audio DAC and amplifier 238, microphone preamplifier and audio ADC 240, temperature sensor interface 242 and clock generator 244. Voltage regulator 234 receives power from processing unit 4 via band interface 232 and provides that power to the other components of head mounted display device 2. Illumination driver 236 controls, for example via a drive current or voltage, the illumination devices 153 to operate about a predetermined wavelength or within a wavelength range. Audio DAC and amplifier 238 receives the audio information from earphones 130. Microphone preamplifier and audio ADC 240 provides an interface for microphone 110. Temperature sensor interface 242 is an interface for temperature sensor 138. Power management unit 202 also provides power and receives data back from three axis magnetometer 132A, three axis gyro 132B and three axis accelerometer 132C. Power management unit 202 also provides power and receives data back from and sends data to GPS transceiver 144.
The variable adjuster driver 237 provides a control signal, for example a drive current or a drive voltage, to the adjuster 135 to move one or more elements of the microdisplay assembly 173 to achieve a displacement for a focal region calculated by software executing in the a processor 210 of the control circuitry 13, or the processing unit 4,5 or the hub computer 12 or both. In embodiments of sweeping through a range of displacements and, hence, a range of focal regions, the variable adjuster driver 237 receives timing signals from the timing generator 226, or alternatively, the clock generator 244 to operate at a programmed rate or frequency.
The photodetector interface 239 performs any analog to digital conversion needed for voltage or current readings from each photodetector, stores the readings in a processor readable format in memory via the memory controller 212, and monitors the operation parameters of the photodetectors 152 such as temperature and wavelength accuracy.
In one embodiment, wireless communication component 346 can include a Wi-Fi enabled communication device, Bluetooth communication device, infrared communication device, etc. The USB port can be used to dock the processing unit 4, 5 to hub computing device 12 in order to load data or software onto processing unit 4, 5, as well as charge processing unit 4, 5. In one embodiment, CPU 320 and GPU 322 are the main workhorses for determining where, when and how to insert virtual images into the view of the user.
Power management circuit 306 includes clock generator 360, analog to digital converter 362, battery charger 364, voltage regulator 366, see-through, near-eye display power source 376, and temperature sensor interface 372 in communication with temperature sensor 374 (located on the wrist band of processing unit 4). An alternating current to direct current converter 362 is connected to a charging jack 370 for receiving an AC supply and creating a DC supply for the system. Voltage regulator 366 is in communication with battery 368 for supplying power to the system. Battery charger 364 is used to charge battery 368 (via voltage regulator 366) upon receiving power from charging jack 370. Device power interface 376 provides power to the display device 2.
The axis 178 formed from the center of rotation 166 through the cornea center 164 to the pupil 162 is the optical axis of the eye. A gaze vector 180 is sometimes referred to as the line of sight or visual axis which extends from the fovea through the center of the pupil 162. The fovea is a small area of about 1.2 degrees located in the retina. The angular offset between the optical axis computed in the embodiment of
In the illustrated embodiment of
Through the display optical systems, 14l and 14r in the eyeglass frame 115, the user's field of view includes both real objects 190, 192 and 194 and virtual objects 182, 184, and 186. In this example, the cornea 168l of the left eye is rotated to the right or towards the user's nose, and the cornea 168r of the right eye is rotated to the left or towards the user's nose. Both pupils are gazing at a virtual object 186. Gaze vectors 180l and 180r from each eye enter the Panum's fusional region 195 in which virtual object 186 is located. The Panum's fusional region is the area of single vision in a binocular viewing system like that of human vision. The intersection of the gaze vectors 180l and 180r indicates that the user is looking at virtual object 186.
For a see-through mixed reality display device, the gaze vectors are determined to identify a point of gaze in a three-dimensional (3D) user field of view which includes both real objects, typically not under computer control, and virtual objects generated by an application. The gaze vectors may intersect at an object 10 feet away or at a distance effectively at infinity. The following figures briefly discuss embodiments for determining a 3D user field of view.
References to front facing image data are referring to image data from one or more front facing camera like camera 113 in
In some embodiments such as in
Each of the method embodiments of
The GPS image tracking application 454 identifies images of the user's location in one or more image database(s) 470 based on GPS data received from the processing unit 4,5 or other GPS units identified as being within a vicinity of the user, or both. Additionally, the image database(s) may provide accessible images of a location with metadata like GPS data and identifying data uploaded by users who wish to share their images. The GPS image tracking application provides distances between objects in an image based on GPS data to the depth image processing application 450. Additionally, the application 456 may perform processing for mapping and locating objects in a 3D user space locally and may interact with the GPS image tracking application for receiving distances between objects. Many combinations of shared processing are possible between the applications by leveraging network connectivity.
The method embodiment in
In step 609, the one or more processors determine whether there are less candidate glints than a predetermined number. For example, for four illuminators, four glints are expected but the predetermined number may be two. In the example of the rectangle as the geometric relationship, two glints which form a horizontal line or a diagonal line of a predetermined length may have been selected as candidates. There may be an eyelid or eyelash obstruction for the other glints. If there are less than the predetermined number of glints, the data sample set is dropped for further processing, and processing returns in step 611 to step 605 of a next data sample set. If there are not less candidates than a predetermined number, then step 613 determines whether there are more candidate glints that a predetermined number. If there are more candidates, in step 615, the one or more processors select as glints the predetermined number of candidates which most closely fit the predetermined geometrical relationship between the glints. For example, for the rectangle, which candidates most closely form the rectangle of the predetermined size and shape. If there are not more candidates than the number, the number of candidates matches the predetermined number of glints, and the candidates are selected as the glints in step 617.
Due to the geometry of the placement of illuminators for generating the glints as discussed above, the glints appear in the same locations, barring movement of the frame 115 with respect to the eye. Furthermore, as the positioning of the illuminators with respect to each other on the support structure of the frame 115 or lens 118 is fixed, the spatial relationship of the glints to each other in the image is fixed as well. As for size, as the glints are very small, the number of pixels making up the glint area on the sensor and in the sensed image would be correspondingly small. For example, if the image sensor of the camera has a 1000 pixels, each glint may take up less than ten pixels. Glints may be monitored in each image frame taken for example at 30 or 60 frames a second and an area may be identified as a glint from a number of frame samples. There may not be glint data in every frame. Sampling accommodates or smoothes out obstructions of glint, and pupil data, in different image frames such as due to factors like an eyelid or eyelash covering the glint and/or pupil. An image frame is an example of an image format.
Optionally, the one or more processors in step 618 determines a position of the fixed center of eyeball rotation with respect to the illuminators and the at least one sensor for the respective eye. This position determined in step 618 provides a depth distance between a fixed point, or one that can be approximated as fixed for accuracy considerations of gaze detection, and the display optical system. In effect, a depth axis has been defined for the gaze detection coordinate system. Changes detected along the depth axis may be used to indicate that the near-eye display system has moved and trigger determination of boundaries of the coordinate system again or re-calibration of training gaze data sets as discussed below.
{circumflex over (q)}i is a position of an illuminatori, the light of which produces glint ĝi (e.g. 174)
ĝi is the glint produced by illuminatori (153) on a cornea surface,
ô is a camera pupil center of the pin-hole camera model,
Îi is the image of glint ĝi on the image plane which is the detection area 139 of the camera sensor,
lengthi is the scalar distance or length from point ô to {circumflex over (q)}i,
Îi is the vector from the camera pupil center ô to the image Îi on the image sensor of the glint ĝi,
{circumflex over (Q)}i is the vector from the camera pupil center ô to the position {circumflex over (q)}i of illuminatori,
the {circumflex over (X)}i axis is defined along {circumflex over (Q)}i, in this example
and the {circumflex over (Z)}i axis of the coordinate system is such so that Îi which connects the image Îi of the glint ĝi on image plane 139 (detection area) lies in a plane formed by the {circumflex over (X)}i and {circumflex over (Z)}i axes.
{circumflex over (β)} is an angle formed in the {circumflex over (X)}i{circumflex over (Z)}i plane between a line 502 representing the incident ray of light from the illuminator (153) position {circumflex over (q)}i to the glint ĝi (174) on a cornea surface.
{circumflex over (α)} is the angle formed in the {circumflex over (X)}i{circumflex over (Z)}i plane between a line 504 representing the reflected ray from the glint ĝi to the camera pupil center of the camera, ô, which is also the origin of the coordinate system.
ĉ is the position of the cornea center which also lies in the {circumflex over (X)}i{circumflex over (Z)}i plane.
As the cornea is modeled as a sphere, r is the radius of the corneal sphere, and each glint ĝi is a point on the first or external surface of the sphere, so each glint is separated from the cornea center by the radius r. In the above example, the glint ĝi is modeled as a point on the exterior surface or first surface of the cornea. In such a model, the light of the illuminator is bouncing off the cornea in the same medium, air, of the same index of refraction as the reflected light of the glint directed back to the camera sensor.
As shown in
wherein
According to Hennessey, the center of the cornea ĉi can be defined in the coordinate system 500 in terms of the unknown parameter ĝix resulting in 3 equations for 4 unknowns (ĉix, ĉiy, ĉiz, ĝix) as follows:
Another two-dimensional plane including the cornea center, ĉ, another glint ĝi, the camera pupil center ô of the camera and a position {circumflex over (q)}i of another illuminator is also formed. The camera pupil center ô of the camera and the cornea center are the same in each plane although the camera pupil center ô position is known. This will result in 6 equations with 8 unknowns. In Hennessey, the gaze detection coordinate system is treated as an auxiliary coordinate system for which a rotation matrix {circumflex over (R)}i can transform points between the auxiliary coordinate systems for each plane and a single world coordinate system such as the third coordinate system which relates the position of the detection area 139 to the illuminators 153. A constraint exists in which the cornea center defined for each glint is the same in the world coordinate system, e.g. ĉ1=ĉ2 and 3 equations result for the different axis components, e.g., ĉ1x=ĉ2x, ĉ1y=ĉ2y, and ĉ1z=ĉ2z, thus providing 9 equations with 8 unknowns. Hennessey (p. 90) states to solve numerically for ĉ using a gradient descent algorithm. Thus, the position center 164 of the cornea 168 is defined with respect to the positions of the illuminators and the image plane or detection area 139.
The pupil appears as a circle in an image format such as an image frame of a camera having its detection area centered on the optical axis of the display when the pupil is looking straight ahead through the display. As the pupil changes its gaze and moves from the center of the image frame, the pupil appears as an ellipse, as a circle viewed from an angle appears as an ellipse. The width of the minor axis of the ellipse changes with gaze changes. A narrow ellipse to the left of the center of the image frame indicates the user is looking to the far right. A wider ellipse a distance less to the right of the center of the image frame indicates the user is looking left but not far left.
The center of the pupil is the center of the ellipse. The ellipse is fitted from detected edge points in the image. Because such edge points are noisy and not all of them are on the ellipse, the ellipse fitting process is repeated many times over randomly selected subsets of all edge points. The subset that is most consistent with all the edge points is used to obtain the final ellipse. The processor in step 646 performs an ellipse fitting algorithm on the average black pupil area for determining an ellipse representing the pupil, and in step 648 determines the center of the pupil by determining the center of the ellipse representing the pupil.
With the center of rotation, the cornea center and the pupil center identified, one can extend a ray from the center of rotation through the cornea and pupil centers to obtain an optical axis for the eye. However, as noted previously, a gaze vector in a human is the visual axis or line of sight from the fovea through the pupil center. Photoreceptors in the fovea region of the human retina are more densely packed than in the rest of the retina. This area provides the highest visual acuity or clearness of vision, and also provides stereoscopic vision of nearby objects. After determining the optical axis, a default offset angle may be applied so that the optical axis approximates the visual axis and is selected as the gaze vector.
In one embodiment, with the fixed positioning of the illuminators as a basis, the effect of different areas of the eye on reflectivity and hence on the amount or intensity of light reflected is used as a basis for gaze detection. Intensity data from either IR or visible light sensors may be used to determine gaze, so the reflectivity data may be based on IR based reflectivity or visible light reflectivity. For illustration, the sclera is more reflective than other areas of the eye like the pupil and the iris. If a user looks to the user's far left, an illuminator 153 located on the frame 115 at the user's far right causes a glint reflection on the right sclera of the user's right eye. PSD 134r or as in
The accuracy may not be as much as those based on images of the full eye, but may suffice for many applications. Additionally, such a gaze detection may be useful as an auxiliary or backup gaze detection technique. For example, during computationally intensive periods of generating complex virtual images, such a glint based technique relieves some processor overhead. Furthermore, such a glint-based technique can be executed many more times in a time period than an image based technique which processes more data or a computationally intensive but more accurate technique which may be run at a slower rate to recalibrate accuracy of gaze detection periodically. An example of a gaze detection technique which is both image based and more computationally intensive is one for determining a gaze vector with respect to inner parts of the eye based on glint data and pupil image data like the embodiments described in
As noted above, different methods with different accuracies may be employed at different periodic rates to trade accuracy for speed. A method embodiment based on glint intensity values such as that described in
Using training data sets for gaze determination relies on the assumption that the near-eye display device 2 with respect to the eye has not moved. If movement is detected, the training gaze data sets are to be recalibrated. A lighting change may also be a basis for recalibration.
A training gaze data set is acquired for each of a set of predetermined gaze directions. For example, training data sets may be obtained for different sections of the display optical system 14 through which the user's pupils gaze at a gaze or pupil angle. In one example, there are nine (9), one for each of the four (4) corners of the display optical system, a middle left side block or area, a middle right side block or area, a top middle block, a bottom middle block, and a center area. In the case of glints, a comparison of intensity values at the four glint positions for current data against training data sets may be used.
a is a flowchart illustrating a method embodiment for generating a set of training data sets for a comparison based determination of gaze. The method may be used to determine training sets for gaze angles based glint intensity value data representing pupil positions. The method embodiment is presented in an exemplar loop structure beginning at step 702 and ending at step 714. For each of a number of training gaze data sets, one or more processors of the control circuitry 136, the processing unit 4, the mobile device 5, a networked hub computing environment 12 alone or in combination, generate in step 704 a virtual image at a predetermined different position for the respective training data set for a predetermined time period in the user field of view. As previously discussed, the microdisplay 120 generates virtual images at different positions in the user field of view.
In step 706, data of each eye is captured during the predetermined time period based on glints. In step 708 from the captured data, the one or more processors determine data representing a pupil position, for example, a set of intensity values from a number of glints. In step 710, a gaze vector is determined based on the pupil position data and the predetermined different position of the virtual image in the user field of view. In the case of pupil and glint data being captured, a gaze vector may be determined based on the cornea center, pupil center and fixed center of eyeball rotation as discussed above with respect to the embodiments of
The box or other geometric shape of glints provides another example. A distance vector for each current glint from a training gaze data set of glint intensity values indicates a direction of intensity change as the glints are fixed barring movement of the coordinate system.
In step 734, the one or more processors select the training gaze data set with the smallest sum for its one or more distance vectors and in step 736 interpolates a position change vector between the pupil position of the selected training gaze data set to the pupil position of the current data. In step 738, the one or more processors estimate as the current gaze vector the gaze vector of the selected training gaze data set moved by the position change vector
Particularly when using training data for comparison, movement of the gaze detection coordinate system is a cause for recalibrating the training data sets. One may periodically predetermine the positions of the cornea center and fixed center of rotation to determine whether there has been a change in the spatial relationship between them and the illuminators and at least one sensor.
Other tests for movement may be performed based on a facial feature with a fixed characteristic in image data. In one embodiment, an eye camera may capture about 5 to 10 mm of area around the visible eyeball portion of the cornea bulge, eye white, iris and pupil so as to capture part of an eyelid and eyelashes. A positionally fixed facial feature like a mole or freckle on skin such as an eyelid or on the bottom rim of the skin encasing the lower eyeball may also be present in the image data of the eye. In image samples, the position of the mole or freckle may be monitored for a change in position. If the facial feature has moved up, down, right or left, a vertical or horizontal shift can be detected. If the facial feature appears larger or smaller, a depth change in the spatial relationship between eye and display device 2 can be determined. There may be a criteria range in the change of position to trigger recalibration of the training gaze data sets due to things like camera resolution, etc.
In another example, although lighting is a factor which changes the size of the pupil and the ratio of pupil area to visible iris area within the circumference or perimeter of the iris, the size of the perimeter or circumference of the iris does not change with gaze change or lighting change; hence, the perimeter or circumference is a fixed characteristic of the iris as a facial feature. Through ellipse fitting of the iris, the one or more processors can determine whether the iris has become larger or smaller in image data in accordance with criteria. If larger, the display device 2 with its illuminators 153 and at least one sensor 134 has moved closer in depth to the user's eye; if smaller, the display device 2 has moved farther away. A change in a fixed characteristic can trigger a recalibration of training data sets.
Mobile device 900 may include, for example, processors 912, memory 1010 including applications and non-volatile storage. The processor 912 can implement communications, as well as any number of applications, including the interaction applications discussed herein. Memory 1010 can be any variety of memory storage media types, including non-volatile and volatile memory. A device operating system handles the different operations of the mobile device 900 and may contain user interfaces for operations, such as placing and receiving phone calls, text messaging, checking voicemail, and the like. The applications 1030 can be any assortment of programs, such as a camera application for photos and/or videos, an address book, a calendar application, a media player, an internet browser, games, other multimedia applications, an alarm application, other third party applications, the interaction application discussed herein, and the like. The non-volatile storage component 1040 in memory 1010 contains data such as web caches, music, photos, contact data, scheduling data, and other files.
The processor 912 also communicates with RF transmit/receive circuitry 906 which in turn is coupled to an antenna 902, with an infrared transmitted/receiver 908, with any additional communication channels 1060 like Wi-Fi or Bluetooth, and with a movement/orientation sensor 914 such as an accelerometer. Accelerometers have been incorporated into mobile devices to enable such applications as intelligent user interfaces that let users input commands through gestures, indoor GPS functionality which calculates the movement and direction of the device after contact is broken with a GPS satellite, and to detect the orientation of the device and automatically change the display from portrait to landscape when the phone is rotated. An accelerometer can be provided, e.g., by a micro-electromechanical system (MEMS) which is a tiny mechanical device (of micrometer dimensions) built onto a semiconductor chip. Acceleration direction, as well as orientation, vibration and shock can be sensed. The processor 912 further communicates with a ringer/vibrator 916, a user interface keypad/screen, biometric sensor system 918, a speaker 1020, a microphone 922, a camera 924, a light sensor 926 and a temperature sensor 928.
The processor 912 controls transmission and reception of wireless signals. During a transmission mode, the processor 912 provides a voice signal from microphone 922, or other data signal, to the RF transmit/receive circuitry 906. The transmit/receive circuitry 906 transmits the signal to a remote station (e.g., a fixed station, operator, other cellular phones, etc.) for communication through the antenna 902. The ringer/vibrator 916 is used to signal an incoming call, text message, calendar reminder, alarm clock reminder, or other notification to the user. During a receiving mode, the transmit/receive circuitry 906 receives a voice or other data signal from a remote station through the antenna 902. A received voice signal is provided to the speaker 1020 while other received data signals are also processed appropriately.
Additionally, a physical connector 988 can be used to connect the mobile device 900 to an external power source, such as an AC adapter or powered docking station. The physical connector 988 can also be used as a data connection to a computing device. The data connection allows for operations such as synchronizing mobile device data with the computing data on another device.
A GPS transceiver 965 utilizing satellite-based radio navigation to relay the position of the user applications is enabled for such service.
The example computer systems illustrated in the figures include examples of computer readable storage media. Computer readable storage media are also processor readable storage media. Such media may include volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, cache, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, memory sticks or cards, magnetic cassettes, magnetic tape, a media drive, a hard disk, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can accessed by a computer.
CPU 801, memory controller 802, and various memory devices are interconnected via one or more buses (not shown). The details of the bus that is used in this implementation are not particularly relevant to understanding the subject matter of interest being discussed herein. However, it will be understood that such a bus might include one or more of serial and parallel buses, a memory bus, a peripheral bus, and a processor or local bus, using any of a variety of bus architectures. By way of example, such architectures can include an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MCA) bus, an Enhanced ISA (EISA) bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnects (PCI) bus also known as a Mezzanine bus.
In one implementation, CPU 801, memory controller 802, ROM 803, and RAM 806 are integrated onto a common module 814. In this implementation, ROM 803 is configured as a flash ROM that is connected to memory controller 802 via a PCI bus and a ROM bus (neither of which are shown). RAM 806 is configured as multiple Double Data Rate Synchronous Dynamic RAM (DDR SDRAM) modules that are independently controlled by memory controller 802 via separate buses (not shown). Hard disk drive 808 and portable media drive 805 are shown connected to the memory controller 802 via the PCI bus and an AT Attachment (ATA) bus 816. However, in other implementations, dedicated data bus structures of different types can also be applied in the alternative.
A graphics processing unit 820 and a video encoder 822 form a video processing pipeline for high speed and high resolution (e.g., High Definition) graphics processing. Data are carried from graphics processing unit (GPU) 820 to video encoder 822 via a digital video bus (not shown). Lightweight messages generated by the system applications (e.g., pop ups) are displayed by using a GPU 820 interrupt to schedule code to render popup into an overlay. The amount of memory used for an overlay depends on the overlay area size and the overlay preferably scales with screen resolution. Where a full user interface is used by the concurrent system application, it is preferable to use a resolution independent of application resolution. A scaler may be used to set this resolution such that the need to change frequency and cause a TV resync is eliminated.
An audio processing unit 824 and an audio codec (coder/decoder) 826 form a corresponding audio processing pipeline for multi-channel audio processing of various digital audio formats. Audio data are carried between audio processing unit 824 and audio codec 826 via a communication link (not shown). The video and audio processing pipelines output data to an A/V (audio/video) port 828 for transmission to a television or other display. In the illustrated implementation, video and audio processing components 820-828 are mounted on module 214.
In the implementation depicted in
MUs 840(1) and 840(2) are illustrated as being connectable to MU ports “A” 830(1) and “B” 830(2) respectively. Additional MUs (e.g., MUs 840(3)-840(6)) are illustrated as being connectable to controllers 804(1) and 804(3), i.e., two MUs for each controller. Controllers 804(2) and 804(4) can also be configured to receive MUs (not shown). Each MU 840 offers additional storage on which games, game parameters, and other data may be stored. In some implementations, the other data can include any of a digital game component, an executable gaming application, an instruction set for expanding a gaming application, and a media file. When inserted into console 800 or a controller, MU 840 can be accessed by memory controller 802. A system power supply module 850 provides power to the components of gaming system 800. A fan 852 cools the circuitry within console 800. A microcontroller unit 854 is also provided.
An application 860 comprising machine instructions is stored on hard disk drive 808. When console 800 is powered on, various portions of application 860 are loaded into RAM 806, and/or caches 810 and 812, for execution on CPU 801, wherein application 860 is one such example. Various applications can be stored on hard disk drive 808 for execution on CPU 801.
Gaming and media system 800 may be operated as a standalone system by simply connecting the system to monitor 16 (
Although the subject matter has been described in language specific to structural features and/or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.
Number | Date | Country | Kind |
---|---|---|---|
2,750,287 | Aug 2011 | CA | national |