`571-272-7822
`
`
`Paper # 26
`
` Entered: November 28, 2022
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`__________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`__________
`
`APPLE, INC., LG ELECTRONICS, INC.,
`LG ELECTRONICS U.S.A., INC., AND GOOGLE LLC,
`Petitioner,
`
`v.
`
`GESTURE TECHNOLOGY PARTNERS, LLC,
`Patent Owner.
`__________
`
`IPR2021-00923
`Patent 8,194,924 B2
`__________
`
`Record of Oral Hearing
`Held: September 14, 2022
`__________
`
`Before PATRICK R. SCANLON, GREGG I. ANDERSON, and
`BRENT M. DOUGAL, Administrative Patent Judges.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`IPR2021-00923
`Patent 8,194,924 B2
`
`
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONER:
`
`
`PAUL HART
`ADAM SEITZ
`Of: Erise IP, P.A.
`adam.seitz@eriseip.com
`paul.hart@eriseip.com
`
`
`
`ON BEHALF OF THE PATENT OWNER:
`
`
`TODD LANDIS
`JOHN WITTENZELLNER
`Of: Williams Simons & Landis PLLC
`tlandis@wsltrial.com
`johnw@wsltrial.com
`
`
`
`
`The above-entitled matter came on for hearing Wednesday,
`September 14, 2022, commencing at 2:10 p.m. EDT, via Video-conference.
`
`
`
` 2
`
`
`
`
`
`
`
`
`IPR2021-00923
`Patent 8,194,924 B2
`
`
`P-R-O-C-E-E-D-I-N-G-S
`
`2:10 p.m.
`
`JUDGE DOUGAL: Okay. This is the oral hearing for IPR 2021-
`
`00923, U.S. Patent 8,194,924 between Apple et al and Gesture Technology
`
`Partners.
`
`As before, I am Judge Brent Dougal. I have with me Judges Scanlon
`
`and Anderson.
`
`We will start with appearances. Who do we have from petitioner?
`
`MR. HART: Thank you, Your Honor. Paul Hart for petitioner,
`
`Apple, Inc. Also in the room is Adam Seitz.
`
`JUDGE DOUGAL: All right. Thank you, Mr. Hart. And who do we
`
`have today for patent owner?
`
`MR. WITTENZELLNER: Good afternoon, Your Honor. My name is
`
`John Wittenzellner with the law firm Williams, Simons and Landis, PLLC.
`
`Here on behalf of the patent owner, I'm joined by my colleague, Mr.
`
`Todd Landis, lead counsel in this proceeding.
`
`JUDGE DOUGAL: Great, thank you. Petitioner, would you like to
`
`reserve any time for rebuttal?
`
`MR. HART: Yes, Your Honor. I'll do 20 minutes for rebuttal.
`
`JUDGE DOUGAL: Okay. And patent owner, would you like to
`
`reserve any time for rebuttal?
`
`MR. WITTENZELLNER: I would also like to reserve 20 minutes,
`
`Your Honor. Thank you.
`
`JUDGE DOUGAL: Okay. Thank you. Well, as this is, I guess at
`
`least for now, the last of a series of hearings.
`
`I'll refer back to our previous discussions in other hearings about our
`
`desire that you are heard.
`
`
`
` 3
`
`
`
`
`
`1
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`27
`
`
`
`
`IPR2021-00923
`Patent 8,194,924 B2
`
`
`And we want to make sure you're heard. If you have any questions or
`
`have any problems with that, please reach out to the contact numbers that
`
`you have. And with that, petitioner, please begin.
`
`MR. HART: Thanks very much, Your Honor. As with all these
`
`proceedings between the parties here, the general issue is gesture based
`
`control of a device.
`
`You need to the '924 patent that is the subject of this proceeding the
`
`claims focus on devices with two cameras that face in different directions
`
`and enable the user to implement a gesture command before one of those
`
`cameras.
`
`DX2 provides a summary of the grounds in this proceeding. The
`
`primary combination at issue proposes that a POSITA would've been
`
`motivated to combine the starkly similar teachings of Mann and Numazaki.
`
`Both reference teach camera-equipped PDAs and camera-equipped
`
`watches.
`
`Both accept user gestures as controlling inputs. Mann's devices are
`
`used to covertly capture video of a subject using the subject facing camera.
`
`To initiate these covert recordings, Mann requires the user to perform
`
`a touch baced gesture on the display of its devices.
`
`Numazaki as we've discussed at length in all these proceedings
`
`teaches controlling its devices using no touch gestures performed over or in
`
`front of a user facing camera.
`
`The petition establishes that Numazaki's no touch gestures improve
`
`the covert nature of Mann's system by allowing a user to casually swipe a
`
`finger over the device to initiate the video recording.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`
`
` 4
`
`
`
`
`
`
`
`
`IPR2021-00923
`Patent 8,194,924 B2
`
`
`DX3 provides a summary of the remaining disputes between the
`
`parties.
`
`The first is a dispute we've seen in every proceeding here, namely
`
`whether Numazaki's figure 2 falls outside the claims because it uses two
`
`cameras rather than one.
`
`The board rejected its argument and institution, and patent owner
`
`provides no reason to depart from that preliminary conclusion.
`
`Next, the parties dispute whether a POSITA would've been motivated
`
`to replace Mann's touch based gestures with Numazaki's no touch gestures.
`
`Relying on Dr. Bederson, the petition established that no touch
`
`gestures provide numerous benefits, including that they are less likely to
`
`draw the subject's attention when the video recording is initiated.
`
`Turning to the third dispute regarding claim 2, the petition establishes
`
`that a POSITA would've understood Mann's wristwatch telephone and its
`
`communication equipped PDA device satisfy the claimed mobile phone
`
`element.
`
`Turning to ground 2, patent owner and its expert argue that
`
`incorporating Amir's pupil detection functionality on demand is too
`
`complex.
`
`But they ignore critical evidence in support, including one, the
`
`Numazaki already teaches multiple cameras incorporated into devices that
`
`are watches and PDAs as Mann's devices are, and two, that Amir teaches its
`
`technology can be easily incorporated into, quote, extremely compact
`
`packages.
`
`The final dispute between the parties is whether Mann, Amir, and
`
`Aviv are analogous art.
`
`
`
` 5
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`
`
`
`IPR2021-00923
`Patent 8,194,924 B2
`
`
`The petition establishes that each describes a camera equip device
`
`with a variety of computer input functionalities, teachings that fall squarely
`
`within the field of endeavor of the '924 patent.
`
`DX4 through DX9 provide a brief overview of the proposed
`
`combination here and motivations to make that combination.
`
`On DX4, we see Mann's watch embodiment. On the left, it's a
`
`wristwatch with two cameras, one facing the user and one facing a subject.
`
`Among its disclosed functionalities as mentioned, it teaches that a
`
`user can initiate a covert video recording of the subject without the subject's
`
`knowledge.
`
`Mann teaches a touchscreen display on a small portion of its watch
`
`face.
`
`In DX4 on the left image there, we can see it's component 320 that is
`
`the touchscreen interface in Mann's wristwatch.
`
`And within that small touchscreen space, there is a watch screen or a
`
`watch -- wristwatch circular pattern that mimics a wristwatch on which the
`
`user can make note or touch based gestures in order to control the device,
`
`performing functions such as initiating or ending a video recording.
`
`In DX5, we see Numazaki's watch embodiment from its eight-
`
`embodiment.
`
`It also is a camera equipped watch that allows the user to exert control
`
`of gestures.
`
`But Numazaki distinguishing it from Mann teaches no touch gestures
`
`performed over that user facing camera.
`
`The petition as I mentioned proposes replacing Mann's touchscreen
`
`gestures with Numazaki's no touch gestures.
`
`
`
` 6
`
`
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`
`
`
`IPR2021-00923
`Patent 8,194,924 B2
`
`
`As discussed in the petition, Mann and Numazaki also include PDA
`
`devices.
`
`Their disclosures largely track the watch disclosures. Both the
`
`watches and the PDAs have been fully mapped in the petition.
`
`In DX6, we see a figure we're very familiar with at this time, figure 2
`
`from Numazaki.
`
`Figure 2 forms the basis of Numazaki's gesture recognition. As
`
`discussed in the other hearings, a first sensor 109 captures an image of the
`
`gesture when illuminated.
`
`Sensor 110 captures an image of the gesture when the light is off.
`
`The difference between those images is produced by the figure 2
`
`structure.
`
`And that is an image that precisely represents just the illuminated
`
`hand, that gesturing hand with all the extraneous background information
`
`removed.
`
`And that process, the two sensor structure that is described by
`
`Numazaki is just to improve the fidelity or the precision of the gesture
`
`detection functionality.
`
`The petition set forth a number of motivations to combine Mann and
`
`Numazaki as I've described.
`
`DX7 is the most prominent of those motivations to combine in the
`
`papers here.
`
`It explains that Numazaki's no touch gestures improve the covert
`
`nature of Mann.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`
`
` 7
`
`
`
`
`
`
`
`
`IPR2021-00923
`Patent 8,194,924 B2
`
`
`Dr. Bederson explained that they are less conspicuous than Mann's
`
`touch based controls and run less risk of a subject noticing that the user has
`
`initiated a recording.
`
`Again, that is a primary concern in Mann that the subject does not
`
`know the user is recording them, that they have started recording them.
`
`So one of the primary goals of Mann is to avoid the subject learning
`
`that a recording has been initiated.
`
`Dr. Bederson also explained that no touch gestures require less focus
`
`than Mann's gestures.
`
`Mann's gestures must be performed on a very small portion of an
`
`already small touchscreen requiring a significant amount of the user's
`
`attention in order to accurately exert that touch based control.
`
`The ability to casually swipe a finger over the device decreases as
`
`Numazaki provides, decreases the chances that the subject will notice a
`
`recording has been initiated.
`
`In DX8, two additional components of the motivation to combine
`
`here.
`
`We first see Numazaki's expressed motivation. It teaches that no
`
`touch gestures avoid obstructing the display on the device that is being
`
`controlled.
`
`Dr. Bederson explained that this benefit is also realized by Mann's
`
`devices with Numazaki's no touch gestures.
`
`He also explains that a separate benefit is realized in Mann, namely
`
`that no touch gestures avoid the user's finger swiping over the lens of the
`
`camera, the user facing camera.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`
`
` 8
`
`
`
`
`
`
`
`
`IPR2021-00923
`Patent 8,194,924 B2
`
`
`He explains that the user's finger touching the lens of the user facing
`
`camera over time will decrease fidelity.
`
`Essentially, a build up of oil and dirt will eventually decrease the
`
`effectiveness of that camera.
`
`DX9 addresses the expectation of success. The similarities between
`
`Mann and Numazaki strongly support an expectation of success here.
`
`Both Mann and Numazaki teach camera equipped watches and PDAs.
`
`Both teach converting user gestures into commands. And Numazaki
`
`establishes that it is technically straightforward to incorporate gesture
`
`recognizing cameras and devices that are both watch and PDA form factors.
`
`DX11 and -- sorry, DX10 and DX11 address the first substantive
`
`dispute between the parties which centers on Numazaki's figure 2.
`
`Starting with DX10, we see the claim language here requires only that
`
`a control function is based on the output of the camera and the combination
`
`that is the output of Numazaki's unit 102.
`
`Patent owner argues that because Numazaki uses a second camera to
`
`improve the fidelity of gesture detection that it falls outside the scope of
`
`these claims.
`
`Again, we've seen this argument in every one of these four
`
`proceedings between these parties.
`
`And DX11, we see the board correctly rejected this narrow view of
`
`the challenged claims.
`
`They correctly found not only that the claim is open ended but also
`
`that the control function is simply based on an output of the first camera and
`
`is not limited to only that output.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`
`
` 9
`
`
`
`
`
`
`
`
`IPR2021-00923
`Patent 8,194,924 B2
`
`
`Patent owner provides the board no reason to deviate from its
`
`preliminary finding here.
`
`Numazaki's gesture recognition unit 102 teaches precisely what's
`
`required by the claims.
`
`It processes gesture control based on the output of a first camera.
`
`That a second camera is used to improve the precision of this
`
`detection is irrelevant to the claim language.
`
`And as we discussed earlier today in the prior proceeding it's not just
`
`the individual image sensors that sensors that satisfy the claim limitations.
`
`But unit 102 as a whole, including its two separate sensors, also
`
`satisfies these claim limitations.
`
`In every scenario, 102 as a whole or either the individual units, image
`
`sensor units, the gesture is detected based on the output of those image
`
`sensors.
`
`DX12 through DX17 addressed the parties' dispute about whether no
`
`touch gestures improve the covert nature of Mann's system.
`
`This is probably the main dispute, the dispute that has taken the most
`
`pages in the papers here.
`
`DX12 reiterates the main points advanced by petitioner's expert, Dr.
`
`Bederson, in support of this covert improvement.
`
`He explained that no touch gestures as taught by Numazaki are less
`
`conspicuous than Mann's touch based controls and run less risk of the
`
`subject noticing that a video has been initiated.
`
`He also explained as I mentioned before that they require less focus.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`
`
`
`
`10
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`
`IPR2021-00923
`Patent 8,194,924 B2
`
`
`The fact that Mann requires small gestures in a small screen requires a
`
`lot of focus of the user which makes it more likely that the subject will
`
`notice they are initiating some process on the device.
`
`We turn to DX13. Patent owner's primary argument is that physically
`
`interacting with watches and PDAs, it's expected.
`
`And because it's expected, it would not draw unwanted attention.
`
`It also suggests that the no touch gestures that are the subject of the
`
`proposed combination here would be elaborate or intriguing, something that
`
`would be more likely to draw attention.
`
`Taking that second point first, that's a flawed premise. There is no
`
`evidence in the record that the no touch gestures proposed in the petition
`
`would be elaborate or intriguing.
`
`They are simply finger swipes in a predetermined direction, casual
`
`finger swipes over the face of the device that is captured by Numazaki's
`
`imaging sensors.
`
`Returning to DX14, the main problem with patent owner's argument is
`
`that it gets the analysis exactly backwards.
`
`The goal is to avoid the subject knowing that a recording has been
`
`initiated.
`
`To do so, the goal is to avoid indicating that the user has interacted
`
`with a device at all, that any process has been initiated on the watch or the
`
`PDA.
`
`As Dr. Bederson explained and as patent owner admits, Mann's touch
`
`based gestures are easily recognizable as users interacting with the devices.
`
`It is known to interact with a watch or a PDA through a screen or
`
`through a stylus.
`
`
`
`
`
`11
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`
`IPR2021-00923
`Patent 8,194,924 B2
`
`
`In doing so indicates to the subject that some process may have been
`
`initiated by the user.
`
`It's for this reason they draw more attention and are less covert that
`
`Numazaki's no touch finger swipes.
`
`DX15, Dr. Bederson elaborated on the user attention point in his
`
`supplemental declaration.
`
`He explained that Mann describes a very small touchscreen that
`
`demands a significant amount of focus of the user in order to accurately
`
`control the device to start or stop video recordings.
`
`In fact, Mann teaches that the rectangular touch portion of the watch
`
`display is only 0.7 inches on the diagonal.
`
`And as we saw in the earlier figures, the display that replicates a
`
`watch, circular display that replicates a watch comprises only a portion of
`
`that 0.7 inches on the diagonal rectangle.
`
`So inside an already small rectangle, Mann's device display is an even
`
`smaller circle.
`
`And it is within that circle that the user must precisely implement a
`
`touch based gesture in order to control the device.
`
`Numazaki's no touch gestures on the other hand merely require the
`
`user to swipe a finger over the watch base in a predetermined direction.
`
`This significantly simpler motion requires far less attention of the user
`
`and accordingly improves the covert nature of the process.
`
`In DX16, patent owner advances a new argument and its surreply that
`
`is limited to Mann's PDA embodiment.
`
`It asserts that -- excuse me, that Mann intends for users to actively
`
`engage with a device to use the stylus while covertly recording a subject.
`
`
`
`
`
`12
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`
`IPR2021-00923
`Patent 8,194,924 B2
`
`
`From this, it suggests that Numazaki's no touch gestures which do
`
`away with a stylus contradict Mann's intent.
`
`Now as a procedural point, patent owner could've advanced this in the
`
`patent owner response and did not.
`
`As a result, petitioner and its expert were deprived an opportunity to
`
`respond.
`
`So the board should not credit this argument as a result of it being late
`
`introduced into the matter.
`
`But even if the board were to consider the argument, it's simply based
`
`on a mischaracterization of Mann.
`
`The first quote that we see on DX16 from Mann that patent owner
`
`includes in its surreply states -- it's addressing scenarios in which the video
`
`is continuously captured to avoid missing any content.
`
`This is irrelevant to the combination. The combination is all about the
`
`initiation of a covert video recording.
`
`If the video is consistently recording or continuously recording, there
`
`is no need for a gesture to initiate it, and it simply falls outside the scope of
`
`what this case is about.
`
`The second statement merely proposes that pretending to write on
`
`PDA may be an effective way of hiding the display if the device is
`
`configured to display the captured subject during the co-recording.
`
`Again, no party and no expert has argued that hiding the display is a
`
`requirement here or would be beneficial.
`
`In fact, Mann at page 4 also discloses that the display can be
`
`configured such that it is only visible to the user and not the subject, doing
`
`away with any need to hide it by pretending to write with a stylus.
`
`
`
`
`
`13
`
`
`
`
`IPR2021-00923
`Patent 8,194,924 B2
`
`
`In sum, this late introduced argument is simply not supported by the
`
`record and turns on a mischaracterization of Mann's teachings.
`
`DX17, the final argument on the covert issue suggests that Dr.
`
`Bederson's opinion should be given little weight because he is not an expert
`
`in, quote, things of a covert nature.
`
`Now, it's important to properly frame the inquiry here. While the
`
`motivation to combine does indeed turn on improving the covert nature of
`
`Mann's device, it is the computer -- human computer interaction or HCI that
`
`dictates how covert such an interaction can be.
`
`The key considerations include how much focus is required of the
`
`user, how casually or inconspicuously a user can perform a covert gesture to
`
`initiate a covert recording.
`
`While the goal is to remain covert, these issues are without question
`
`dictated by the specific HCI limitations of a system.
`
`Dr. Bederson's experience is a precise fit for those issues, and it
`
`precisely aligns with the definition of a POSITA proposed by petitioner and
`
`accepted by patent owner in this case.
`
`That definition proposes that a POSITA would have experience in the
`
`field of human computer interaction.
`
`As we see in DX17 which summarizes Dr. Bederson's qualifications
`
`in this matter, he is widely recognized as a leader in the realm of technology
`
`in the HCI space.
`
`He was a member and past director of the Human Computer
`
`Interaction Lab, one of the best known and oldest HCI labs.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`
`
`
`
`14
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`
`IPR2021-00923
`Patent 8,194,924 B2
`
`
`He's received numerous awards for his HCI contributions. And
`
`outside of the HCI space, he also has direct experience with the specific
`
`technology implemented by the proposed combination.
`
`The sensing systems, the applications and mobile devices, the list goes
`
`on and one.
`
`Dr. Bederson is a precise fit for the issues in this case, not only on the
`
`hardware technology side but also on the HCI side.
`
`That really is what dictates the covert questions that are before the
`
`board.
`
`DX18 through DX20 address Numazaki's benefits that no touch
`
`gestures avoid blocking the display and avoid a user's finger touching the
`
`camera lens which degrades fidelity over time.
`
`DX18, as I mentioned earlier, Numazaki expressly motivates the
`
`combination by noting that no touch gestures avoid blocking the display, a
`
`known detriment with touch screen devices at the time.
`
`In response in DX19, the surreply -- patent owner's surreply advances
`
`a unique argument.
`
`It argues that even if Numazaki solves the obstructed display problem
`
`in the context of cover video recordings that benefit can be ignored because
`
`the proposed combination does not eradicate touch inputs entirely.
`
`In other words, patent owner asked this board to ignore a concrete
`
`benefit of the proposed combination simply because the petition does not
`
`explain how to realize that benefit in all contexts unrelated to the invalidity
`
`challenge.
`
`Patent owner provides no authority for this unique and narrow take on
`
`the law nor is petitioner aware of any such authority.
`
`
`
`
`
`15
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`
`IPR2021-00923
`Patent 8,194,924 B2
`
`
`In DX20, we see that patent owner is advancing the same argument
`
`with respect to the benefit the proposed combination adds that it avoids the
`
`user touching the camera lens and degrading fidelity over time.
`
`Again, there's no need that a petition solve a problem in all contexts,
`
`in all uses in a device.
`
`The focus of the petition here is on covert video recordings and how
`
`Numazaki's processes improved those recordings.
`
`The petition did not address nor was it obligated to address other uses
`
`of Mann's devices.
`
`Because there's no legal support for this new critique, the board
`
`should accord it no weight.
`
`DX21 through DX23 addressed patent owner's argument that
`
`Numazaki's figure 2 images are suitable for humans.
`
`Now we didn't get much detail on this in the patent owner response.
`
`As we see in DX21, patent owner did argue that Numazaki's figure 2
`
`is unsuitable for capturing video of the user.
`
`They site their expert, Mr. Occhiogrosso. But he merely stated that its
`
`output is a reflected light image that is unsuitable for humans, that average
`
`users would be confused upon viewing the output of figure 2.
`
`He provided no rational whatsoever for this opinion. In response,
`
`DX22, we see Dr. Bederson's supplemental declaration.
`
`He explained that the whole point of Numazaki's figure 2 is to
`
`improve the precision of the image.
`
`It uses the second sensor to remove extraneous information so that the
`
`resulting image captures only the illuminated object, improving the fidelity
`
`of the output.
`
`
`
`
`
`16
`
`
`
`
`IPR2021-00923
`Patent 8,194,924 B2
`
`
`He also explained that there is no basis on which to conclude that
`
`figure 2 produces an image unsuitable for humans.
`
`In DX23, patent owner's surreply introduces a new technical argument
`
`that neither party's expert has had an opportunity to analyze or address.
`
`Patent owner suggests that Numazaki's reflected light images, the
`
`output of figure 2, would be invisible to humans because Numazaki uses an
`
`infrared illumination, a lighting unit based on infrared light which is not
`
`visible to humans when illuminated from that lighting source.
`
`Now as I noted, this is a new argument with no support. Neither
`
`parties' expert has addressed it and we have not had an opportunity to
`
`address it.
`
`For that reason, it should be accorded no weight. But even if the
`
`board were to consider this argument, it's technically wrong.
`
`Infrared imaging is a hugely popular field with many applications.
`
`One very popular example is nighttime wildlife photography where
`
`visible light would scare the subjects.
`
`I've looked at the images, infrared, nighttime wildlife photography.
`
`They are beautiful, precise, absolutely intended for human viewing.
`
`Another example has been in the press widely lately. The James
`
`Webb space telescope operates on an infrared light and is currently
`
`producing images of space that are significantly more precise than
`
`predecessor telescopes that operate on the visible spectrum.
`
`Again, because patent owner's new theory is not only unsupported in
`
`the record but also contradicts the actual size of infrared imaging. It should
`
`be rejected here.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`
`
`
`
`17
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`
`IPR2021-00923
`Patent 8,194,924 B2
`
`
`The slides 24 and 25 address dependent claim 2's requirement that the
`
`device is a m mobile phone.
`
`Starting with DX24, the petition established a few points regarding
`
`Mann's watch and PDA embodiments.
`
`Number one, that there's no dispute that a watch and a PDA are
`
`mobile devices.
`
`Number two, Mann expressly characterizes the watch as a wristwatch
`
`telephone.
`
`Number three, Dr. Bederson explained that a POSITA would have
`
`agreed that the watch is a wristwatch telephone primarily because Mann
`
`teaches the watch includes wireless communication functionality.
`
`It is a communication, wireless communication capable portable
`
`device.
`
`Four, Mann teaches that its PDA also includes wireless
`
`communication functionality.
`
`Dr. Bederson analyzed that teaching and explained that it would also
`
`be considered a mobile phone for the same reason as the watch.
`
`They both include wireless communication capabilities. The watch is
`
`expressly described as a phone.
`
`And one of skill in the art would conclude that the PDA is as well.
`
`No patent owner does not dispute that Mann's devices are mobile or
`
`that the watch expressly characterizes a phone.
`
`Instead, it seeks to distinguish these teachings by asking the board to
`
`import numerous limitations into claim 2.
`
`Specifically as we see in DX25, patent owner asked the board to
`
`import that the mobile phone is a cellular phone specifically, that it includes
`
`
`
`
`
`18
`
`
`
`
`IPR2021-00923
`Patent 8,194,924 B2
`
`the ability to dial numbers, that it includes the ability to answer voice calls,
`
`and that it include a speaker and a microphone allowing the user to
`
`participate in voice calls.
`
`With all these features read into the claim, patent owner asked the
`
`board to exclude Mann's teaching because Mann does not expressly describe
`
`these low level details of placing and receiving voice calls.
`
`Importantly, patent owner admits that the '924 also does not describe
`
`voice calls.
`
`It is entirely focuses, exclusively focused on data communications.
`
`Patent owner in its surreply argue that the implementation details of
`
`voice calls are inherent in the disclosure of a phone and that the '924 patent
`
`need not disclosure based on that inherency.
`
`But that's a logically flawed argument. If the '924 patent may not
`
`describe voice calls because they are inherent in the disclosure of a phone,
`
`those functionalities are also inherent in the prior art's disclosure of a phone.
`
`Mann unambiguously describes mobile devices with communication
`
`capabilities, and it expressly describes its watch as a telephone.
`
`That is all the claim requires here. The record strongly supports
`
`petitioner's conclusion that both devices in Mann satisfy the mobile phone
`
`limitations of dependent claim 2.
`
`DX26 through DX28 addressed disputes concerning round two and
`
`the petitioner's proposal to implement Amir's pupil detection functionality.
`
`And the point of including Amir's pupil detection functionality is to
`
`ensure Mann's device captures the subject when the subject is both facing
`
`the camera and with its eyes open.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`
`
`
`
`19
`
`
`
`
`IPR2021-00923
`Patent 8,194,924 B2
`
`
`In addition to the covert nature of Mann's device, one of Mann's goals
`
`as discussed in a petition is to identify subjects.
`
`So it's subject identifying goals here. And as we describe in the
`
`petition, ensuring the user is facing the camera with our eyes open improves
`
`the chances that that subject's identifying characteristics, their facial
`
`characteristics from which their identity can be discerned or captured by the
`
`images.
`
`DX26 provides an overview of patent owner's primary argument here
`
`based on Mr. Occhiogrosso.
`
`His primary argument is a conclusory statement that incorporating
`
`both Amir's pupil detection and Numazaki's gesture detection into Mann's
`
`devices is too technically difficult to be obvious.
`
`As with prior opinions, Mr. Occhiogrosso ignores key evidence and
`
`support.
`
`Number one, he ignores that Numazaki's own teachings demonstrate
`
`the technical feasibility of incorporating multiple cameras and watches in
`
`PDA.
`
`As I've stated multiple times in this hearing, Numazaki's devices
`
`include multiple cameras.
`
`And Numazaki expressly teaches that those devices can include
`
`watches and PDAs.
`
`He also ignores that Amir expressly describes its hardware as a
`
`minimal to, quote, extremely compact overall packages, end quote.
`
`All of these disclosures in Numazaki and Amir are directed to very
`
`small devices like watches and PDAs.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`
`
`
`
`20
`
`
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`25
`
`26
`
`
`IPR2021-00923
`Patent 8,194,924 B2
`
`
`They all assume it is technically feasible to do so because only
`
`petitioner's expert has considered the full relevant record including
`
`Numazaki's teachings and Amir's teachings on this point.
`
`The board should give a little credit to Mr. Occhiogrosso's conclusory
`
`pushback on the technical feasibility.
`
`DX27 highlights patent owner's response and its surreply. In the
`
`POR, in the patent owner response, patent owner argued that Amir's
`
`functionality requires a user to look directly at the camera.
`
`They reason that there was nothing in the record suggesting a user
`
`would look directly at the watch or the PDA in Mann's system because
`
`they're not supposed to know that the video is being recorded at all.
`
`Petitioner's replay establish that there is no requirement that the
`
`subject look at the camera.
`
`Amir's functionality simply detects open eyes, not that the subject is
`
`staring directly into the camera.
`
`Patent owner abandoned that argument in surreply and shifted gears to
`
`suggest that Mann's devices may be positioned so below a subject's fact that
`
`Amir's functionality could not tell whether the subject's eyes are open. And
`
`in that case, it may not work properly.
`
`As with a lot of these arguments, there's simply no basis on which to
`
`assume that such a problem would exist in the real world.
`
`Were that device placed so low that it could not see the subject's eyes,
`
`it would be a misuse of the device and the mere possibility that a user could
`
`misuse the subject device does not defeat obviousness here.
`
`In DX28, p