`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 1 of 54
`
`EXHIBIT 1
`EXHIBIT 1
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 2 of 54
`
`Page 1
`
` IN THE UNITED STATES DISTRICT COURT
` FOR THE WESTERN DISTRICT OF TEXAS
` WACO DIVISION
`
` GENTEX CORPORATION and )
` INDIGO TECHNOLOGIES, LLC, )
` )
` Plaintiffs, )
` )
` THALES VISIONIX, INC., ) Case No. 6:21-cv-00755
` )
` Involuntary Plaintiff, )
` )
` vs. )
` )
` FACEBOOK, INC. and )
` FACEBOOK TECHNOLOGIES, )
` LLC, )
` )
` Defendants. )
`
` VIDEO-RECORDED DEPOSITION OF AARON BOBICK, PhD
` TAKEN ON BEHALF OF THE PLAINTIFFS
` MARCH 15, 2022
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`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
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 3 of 54
`
`Page 2
`
` I N D E X
`
` WITNESSES
`
` ALL WITNESSES PAGE
`
` For Plaintiff
`
` Aaron Bobick, PhD
`
` Examination by Mr. Harber 7
`
` EXHIBITS
`
` NO. PAGE
`
` Exhibit 1 Declaration of Dr. Aaron
`
` Bobick in Support of Defendants'
`
` Opening Claim Construction Brief 14
`
` Exhibit 2 U.S. Patent 6,757,068 65
`
` Exhibit 3 U.S. Patent 7,301,648 65
`
` Exhibit 4 Article, Localized feature
`
` selection to maximize
`
` discrimination 93
`
` Exhibit 5 U.S. Patent 6,922,632 136
`
` Exhibit 6 U.S. Patent 7,725,253 136
`
` Exhibit 7 Bobick errata sheet for
`
` Declaration paragraph 58 183
`
` (Exhibits attached to transcript.)
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`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
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 4 of 54
`
`Page 3
`
` IN THE UNITED STATES DISTRICT COURT
` FOR THE WESTERN DISTRICT OF TEXAS
` WACO DIVISION
` GENTEX CORPORATION and )
` INDIGO TECHNOLOGIES, LLC, )
` )
` Plaintiffs, )
` )
` THALES VISIONIX, INC.,) Case No. 6:21-cv-00755
` )
` Involuntary Plaintiff, )
` )
` vs. )
` )
` FACEBOOK, INC. and )
` FACEBOOK TECHNOLOGIES,)
` LLC, )
` )
` Defendants. )
`
` VIDEO-RECORDED DEPOSITION OF WITNESS,
` AARON BOBICK, produced, sworn and examined on the
` 15th day of March, 2022, between the hours of eight
` o'clock in the forenoon and six o'clock in the
` afternoon of that day, at the offices of Thompson
` Coburn, LLC, One US Bank Plaza, St. Louis,
` Missouri, before Tara Schwake, a Registered
` Professional Reporter, Certified Realtime Reporter,
` Certified Shorthand Reporter (IL), Certified Court
` Reporter (MO), and Notary Public within and for the
` State of Missouri.
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`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
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 5 of 54
`
`Page 4
`
` A P P E A R A N C E S
`
` F O R T H E P L A I N T I F F S :
`
` W I L L I A M S & C O N N O L L Y L L P
`
` 7 2 5 T w e l f t h S t r e e t , N . W .
`
` W a s h i n g t o n , D C 2 0 0 0 5
`
` ( 2 0 2 ) 4 3 4 - 5 0 0 0
`
` b y : M r . A d a m D . H a r b e r
`
` a h a r b e r @ w c . c o m
`
` F O R T H E D E F E N D A N T S :
`
` K I R K L A N D & E L L I S L L P
`
` 4 0 1 C o n g r e s s A v e n u e
`
` A u s t i n , T e x a s 7 8 7 0 1
`
` ( 5 1 2 ) 6 7 8 - 9 1 0 0
`
` b y : M s . B a i l e y M o r g a n W a t k i n s
`
` b a i l e y . w a t k i n s @ k i r k l a n d . c o m
`
` K I R K L A N D & E L L I S L L P
`
` 2 0 4 9 C e n t u r y P a r k E a s t , S u i t e 3 7 0 0
`
` L o s A n g e l e s , C a l i f o r n i a 9 0 0 6 7
`
` ( 3 1 0 ) 5 5 2 - 4 2 0 0
`
` b y : M r . E l l i s e n T u r n e r
`
` e l l i s e n . t u r n e r @ k i r k l a n d . c o m
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`1
`
`2 3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`1 0
`
`1 1
`
`1 2
`
`1 3
`
`1 4
`
`1 5
`
`1 6
`
`1 7
`
`1 8
`
`1 9
`
`2 0
`
`2 1
`
`2 2
`
`2 3
`
`2 4
`
`2 5
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 6 of 54
`
` C O U R T R E P O R T E R :
`
` T A R A S C H W A K E , C R R , R P R , C C R , C S R
`
` F o r V e r i t e x t L e g a l S o l u t i o n s
`
`Page 5
`
` V I D E O G R A P H E R :
`
` T I M O T H Y P E R R Y , C L V S
`
` F o r V e r i t e x t L e g a l S o l u t i o n s
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`1
`
`2
`
`3
`
`4 5
`
`6
`
`7
`
`8 9
`
`1 0
`
`1 1
`
`1 2
`
`1 3
`
`1 4
`
`1 5
`
`1 6
`
`1 7
`
`1 8
`
`1 9
`
`2 0
`
`2 1
`
`2 2
`
`2 3
`
`2 4
`
`2 5
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 7 of 54
`
`Page 26
`
` position orientation of the vehicle with respect to
`
` the world, and with respect to those objects.
`
` Q When you say "required estimating the
`
` position orientation of the vehicle with respect to
`
` the world," how was that estimation done?
`
` A Um, to the best of my recollection,
`
` because that was not the point of the -- what was
`
` written in the paper but it had to be implemented,
`
` we implemented a variety of Kalman filters to take
`
` information from the -- both the vehicle and the
`
` measurements of the objects in the imagery in order
`
` to estimate positions of the -- both the -- both
`
` the vehicle and the objects.
`
` Q Did you work on the implementation of
`
` the Kalman filter in that?
`
` A I wrote all the code.
`
` Q And what is a Kalman filter?
`
` A In general context, a Kalman filter
`
` is a specific mathematical formulation of -- that
`
` relates states of an object to some measurements,
`
` and being able to take those measurements
`
` iteratively over time and to continue to refine the
`
` state.
`
` It has some very explicit models
`
` built into it in terms of the nature of the sensor
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`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
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 8 of 54
`
`Page 27
`
` models, and -- and the basic Kalman filtering has
`
` specific limitations on how the state is related to
`
` the measurements.
`
` Q Why do -- let me ask this first.
`
` Would a person of ordinary skill in the art, at the
`
` time of the patents we're talking about in this
`
` case, be familiar with Kalman filtering?
`
` A Yes.
`
` Q And there is not one specific
`
` algorithm that is a Kalman filter; correct?
`
` MR. TURNER: Objection, vague.
`
` A So there is a set of equations that
`
` define the Kalman filter. So to the extent to
`
` which implementing those equations is the
`
` algorithm, then that would be the algorithm for
`
` implementing a Kalman filter.
`
` Q (BY MR. HARBER) Let me ask it this
`
` way. There's more than one type of Kalman filter;
`
` correct?
`
` MR. TURNER: Objection, vague.
`
` A Well, there are extensions to Kalman
`
` filters which are often referred to as extended
`
` Kalman filter, but Kalman filter, basic Kalman
`
` filtering itself, is a particular set of equations
`
` with a particular model.
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`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
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 9 of 54
`
`Page 28
`
` Q (BY MR. HARBER) But you could
`
` implement a Kalman filter to take in the data you
`
` were just talking about, the state estimates and
`
` sensor data, you could implement Kalman filters
`
` that will give you different results. Correct?
`
` MR. TURNER: Objection, vague,
`
` incomplete hypothetical.
`
` THE WITNESS: But an incomplete
`
` hypothetical, I should answer?
`
` MR. TURNER: Yes.
`
` A Okay. I mean, if you implement the
`
` equations that are of a Kalman filter and you use
`
` the same parameters, you'll get the same results.
`
` Q (BY MR. HARBER) So you're saying --
`
` is it your testimony that any Kalman filter, when
`
` given the same information, would produce the same
`
` results?
`
` A Any implementation of the basic
`
` Kalman filter equations, using the same parameters
`
` and measurements, would give the same results.
`
` Q What do you mean by same parameters
`
` and measurements?
`
` A Well, for example, you have to
`
` parameterize the noise matrices that represent the
`
` noise that you anticipate from the sensor. And
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`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
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 10 of 54
`
`Page 33
`
` filter is getting that input.
`
` MR. TURNER: Same objections.
`
` A The programmer of the Kalman filter
`
` would have to provide code -- well, let's be
`
` precise. The programmer is building the whole
`
` system in which the Kalman filter is sitting.
`
` Somewhere that system has to make a decision about
`
` where it's going to get its -- those parameters.
`
` So the program -- the programmer
`
` could -- would provide the code by which the thing
`
` then decides how to get those parameters.
`
` Q (BY MR. HARBER) And the output of
`
` the Kalman filter depends on the parameters that
`
` are input; correct?
`
` MR. TURNER: Objection, incomplete
`
` hypothetical.
`
` A The output of the Kalman filter
`
` depends upon the parameters that are in the Kalman
`
` filter.
`
` Q (BY MR. HARBER) So is that, you
`
` agree, that that's true?
`
` A I agree what's true?
`
` Q That the output of the Kalman filter
`
` depends on the inputs.
`
` A Well, the output of a Kalman filter
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`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
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 11 of 54
`
`Page 34
`
` clearly depends on the inputs to the Kalman filter.
`
` Q So depending on what the inputs are,
`
` the output of the Kalman filter could be different?
`
` A Sure.
`
` Q And -- and to be clear, by the output
`
` of the Kalman filter, we're talking about a state
`
` estimate for the object you are tracking, for
`
` example.
`
` MR. TURNER: Objection, misstates
`
` testimony, and it's an incomplete hypothetical.
`
` A Well, Kalman filter doesn't know
`
` anything about tracking or -- or -- all a Kalman
`
` filter is, is it's a way of filtering over time to
`
` achieve a -- an estimate of a state that is
`
` mathematically related to a set of measurements
`
` that are taken over time.
`
` So the output over time is the
`
` estimate of that state and statistical parameters
`
` about that estimate.
`
` Q (BY MR. HARBER) And the estimate of
`
` the state and the statistical parameters about that
`
` estimate will vary depending on what the inputs to
`
` the Kalman filter are; correct?
`
` MR. TURNER: Objection, incomplete
`
` hypothetical.
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`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
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 12 of 54
`
`Page 35
`
` A The -- the outputs -- I mean, Kalman
`
` filter produces outputs every time step. The
`
` outputs are a function of the totality of inputs
`
` that it has seen since it was turned on.
`
` Q (BY MR. HARBER) So would you agree
`
` that the estimate of the state and statistical
`
` parameters that are at the output of a Kalman
`
` filter will vary depending on the inputs to the
`
` Kalman filter?
`
` A The outputs of a Kalman filter and
`
` statistics of them are determined by the inputs to
`
` the Kalman filter.
`
` Q So they will be different depending
`
` on what the inputs are?
`
` A If you give a Kalman filter two
`
` different inputs, you will get two different
`
` outputs.
`
` Q And a person of ordinary skill would
`
` have known this at the time of the patents?
`
` A I would think so.
`
` Q And what is the reason, in a motion
`
` tracking system, that estimates are used?
`
` MR. TURNER: Objection, incomplete
`
` hypothetical, calls for speculation.
`
` A As opposed to what?
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`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
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 13 of 54
`
`Page 45
`
` context?
`
` A Well, I didn't say Kalman-like. I
`
` said "like a Kalman filter." Right? So what I
`
` meant by "like a Kalman filter" is a filter that
`
` has noise models associated with it and that has a
`
` way of integrating new information considering the
`
` noise models.
`
` Q And a person of ordinary skill in the
`
` art, before the priority date, would be familiar
`
` with a filter that has noise models associated with
`
` it and that has a way of integrating new
`
` information with noise models in a tracking system.
`
` Correct?
`
` A Well, since a Kalman filter is such
`
` an example, and since someone in the art would be
`
` familiar with Kalman filtering, then I think,
`
` tautologically, the answer there is yes.
`
` Q Would you agree that those of skill
`
` in the art before the priority date would have been
`
` familiar with other noise models other than a
`
` Kalman filter that did the same, performed the same
`
` function?
`
` MR. TURNER: Objection, vague.
`
` A I would imagine that they would be
`
` familiar with other stochastic models that
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`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
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 14 of 54
`
`Page 46
`
` integrated noisy measurements with noisy estimates.
`
` Q (BY MR. HARBER) What do you mean by
`
` stochastic models?
`
` A Sorry. Models that involve some form
`
` of a probability distribution or -- probability
`
` distribution over both the -- the noise of the
`
` measurements and the state estimates.
`
` Q We -- we got a little diverted when I
`
` was asking you about your -- your work with
`
` tracking systems. I think you had pointed in your
`
` CV to C 94, C 98, and J 25. Anything else that
`
` relates to motion tracking systems?
`
` A It's C -- C 97, 98, 99 are -- which
`
` are all by myself and Bob Bolles, are reporting on
`
` similar kinds of work but with different aspects of
`
` it. But in support of all of those was -- was the
`
` system that I had constructed for doing the -- the
`
` tracking.
`
` Q So are -- all of these publications
`
` are about the same system?
`
` A Well, the 97, 98, and 99 are about
`
` the same system.
`
` Q And can you describe that system
`
` again, please?
`
` A This was a system that process data
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`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
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 15 of 54
`
`Page 47
`
` that was captured on an autonomous land vehicle
`
` under manual control, and we -- and I constructed a
`
` system that would detect objects and populate them
`
` -- populate representations of them in a world
`
` coordinate system and develop representations of
`
` them.
`
` Q And am I correct that the sensors
`
` that were used there I think you said it was a -- a
`
` camera. Anything else?
`
` A Actually, so the sensors that we made
`
` use of were a laser range finder camera -- I mean,
`
` a laser range finder is a type of camera, it
`
` generates for you a 3D image, and we had inertial
`
` sensors on the vehicle that provided information as
`
` to what it believed the movement of the vehicle
`
` was.
`
` I think those were the -- the only
`
` sensors that we made use of for the geometry.
`
` Q Have you ever used an inertial sensor
`
` in any other context besides that one?
`
` A Yes.
`
` Q When was that?
`
` A I don't think it ever made it into
`
` the papers, but when we started doing work on
`
` gesture recognition, we had some sensors on the
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`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
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 16 of 54
`
`Page 68
`
` tracker to track a position of a
`
` localized feature associated with a
`
` limb of the user relative to the
`
` user's head. And the localized
`
` feature associated with the limb may
`
` include a hand-held object or a
`
` hand-mounted object or a point on the
`
` hand.
`
` Q (BY MR. HARBER) Now, did you
`
` understand that sentence that you just read when
`
` you first read it?
`
` MR. TURNER: Objection, vague.
`
` A I mean, what I understood was that
`
` this was something to do with tracking the
`
` relationship between something having to do with a
`
` person's hand and something to do with a person's
`
` head.
`
` Q (BY MR. HARBER) And if we look on
`
` page 6 of your Declaration Exhibit 1, paragraph 18,
`
` and I believe you read from part of this before, I
`
` want to read through this and I'll omit the -- the
`
` citations here, just -- just for ease of the
`
` record.
`
` You say, (Quote as read):
`
` The '068 and '648 Patents relate to
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`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
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 17 of 54
`
` position and/or orientation tracking?
`
`Page 69
`
` A I'm sorry, which --
`
` Q Paragraph 18.
`
` A Sorry, yes.
`
` Q Are you with me?
`
` A Yeah.
`
` Q And so I'll start again. You say
`
` that (Quote as read):
`
` The '068 and '648 Patents relate to
`
` position and/or orientation tracking.
`
` Specifically, the '068 and '648
`
` Patents relate to methods and systems
`
` for tracking the position of a user's
`
` limb relative to the orientation of a
`
` user's head.
`
` Do you see that?
`
` A I do.
`
` Q And that's your opinion; correct?
`
` A Well, this is stating what the --
`
` directly what the patents say.
`
` Q And so the patent is directly saying
`
` that, um, they (Quote as read):
`
` relate to methods and systems for
`
` tracking the position of a user's
`
` limb relative to the orientation of a
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`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
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 18 of 54
`
`Page 70
`
` user's head?
`
` I guess what I'm trying to --
`
` A Yes.
`
` Q -- to under -- that's what you took
`
` away from the patent? That that's what the systems
`
` and methods relate to. Correct?
`
` A This is a portion of the
`
` specification that I used to convey what the patent
`
` was generally about.
`
` Q And you go on to say, (Quote as
`
` read):
`
` To this end, the '068 Patent -- that
`
` the '068 and '648 Patents discuss
`
` mounting an orientation tracker on a
`
` user's head and using another tracker
`
` to track the position of a first
`
` localized feature associated with a
`
` limb and tracking the feature
`
` relative to the user's head.
`
` Do you see that?
`
` A I do.
`
` Q And so the -- the purpose of using a
`
` tracker to track the position of a first localized
`
` feature associated with a limb is to track the
`
` position of a user's limb relative to the -- to the
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`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
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 19 of 54
`
`Page 71
`
` head; correct?
`
` MR. TURNER: Objection, misstates the
`
` document.
`
` A No, that's not what it says.
`
` Q (BY MR. HARBER) Well, what -- what
`
` do you -- what does that mean, if not that?
`
` A Well, that part of the -- as you see,
`
` following where it says, (Quote as read):
`
` the localized feature associated with
`
` the limb may include a hand-held
`
` object or a hand-mounted object or a
`
` point on the hand, but does not
`
` otherwise define.
`
` So, so it's -- it's not what you just
`
` said. I mean, it's -- these are potential examples
`
` of what it might be.
`
` Q You say here "To this end," which is
`
` referring to methods and systems for tracking the
`
` position of a user's limb relative to the
`
` orientation of a user's head; correct?
`
` A So that's the sentence that I wrote
`
` previously, yes.
`
` Q And the "To this end" refers to that
`
` sentence; correct?
`
` A Well, this says that (Quote as read):
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`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
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 20 of 54
`
`Page 77
`
` the systems and methods that are described in these
`
` patents, part of the systems and methods is to
`
` track where a user's limb is; correct?
`
` MR. TURNER: Objection, it's
`
` compound.
`
` A The patent says that it's going to
`
` track a localized feature associated with the limb.
`
` Q (BY MR. HARBER) And you don't have
`
` an opinion one way or the other about whether it's
`
` doing that in order to track the position of the
`
` limb?
`
` MR. TURNER: Objection, it's vague
`
` and compound.
`
` A Given the background statement, it's
`
` doing it to enable a manual interaction with a
`
` virtual world.
`
` Q (BY MR. HARBER) And is -- do you
`
` have an opinion about whether that manual
`
` interaction in these patents is achieved by
`
` tracking the position and orientation of a limb, or
`
` not?
`
` MR. TURNER: Objection, it's vague
`
` and compound.
`
` A My -- my opinion is, as I said, that
`
` it's doing this by tracking the feature that's
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`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
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 21 of 54
`
`Page 78
`
` somehow associated, you know, with the limb.
`
` Q (BY MR. HARBER) Have you heard the
`
` term "localized feature" used in tracking before?
`
` A So localized, we talk about, say, a
`
` localized point in tracking, localized could be
`
` located in an image, or could be located in a
`
` three-dimensional world. I'm not sure I've seen
`
` localized feature in tracking per se.
`
` Q Have you ever used the term
`
` "localized feature" in any of your publications?
`
` A I'd have to review my publications.
`
` Q As you sit here today, you're not
`
` familiar with that?
`
` A Um, as I sit here today, I can't
`
` point to a publication that would definitely have
`
` that phrase in it.
`
` Q And I want to go back to the first
`
` question I asked because I'm not sure whether you
`
` answered that or not. Have you heard of the term
`
` "localized feature" before you were involved in
`
` this case?
`
` A I can't say for sure that I have not,
`
` but it would be in a specific context where I would
`
` then know what it is we were talking about.
`
` Q I guess I still don't understand the
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`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
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 22 of 54
`
`Page 79
`
` answer to that question. Have you -- do you recall
`
` hearing the phrase "localized feature" used before
`
` you saw this patent, or do you not?
`
` A So I -- I can't come up with a
`
` specific reference where we talked about "localized
`
` feature" as the -- as the -- as the phrase. But
`
` since we would sometimes find features location in
`
` an image that would presumably be in -- when
`
` talking about finding something in an image, it
`
` would be the localized feature in the image, it
`
` could very well be that we would have used that
`
` phrase to talk about locating the feature in the
`
` image.
`
` Q What does a "localized feature in an
`
` image" mean?
`
` A Typically you have some image-based
`
` description of what this feature would look like,
`
` and then it's where is it in the image. What, you
`
` know, what ij location at which the feature has
`
` been located.
`
` Q Why is it called a localized feature
`
` in that context?
`
` MR. TURNER: Objection, incomplete
`
` hypothetical.
`
` A Well, that would be using the term
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`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
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 23 of 54
`
`Page 80
`
` "localized" to mean as you found its location.
`
` Q (BY MR. HARBER) So it's just a
`
` feature thats location you have found?
`
` A Well, that -- that was -- I -- I
`
` could imagine using that phrase to mean that.
`
` There are local features, and in fact in one of the
`
` papers discussed here, they use the phrase
`
` "localized feature," meaning the selection of a
`
` local feature, meaning it's just the property of
`
` some local pixels in an area, but it's not
`
` particularly robustly defined.
`
` Q So I want to go back to "localized
`
` feature" for a minute, then we can talk about local
`
` feature.
`
` It -- it's just -- your -- have you
`
` heard the phrase used as just the selection of a
`
` point that one is tracking?
`
` MR. TURNER: Objection, vague.
`
` A Sitting here today, I can't think of
`
` that. Doesn't mean it hasn't happened, but did you
`
` say a selection of a point being tracked?
`
` Q (BY MR. HARBER) You -- I think
`
` that's what you said. Is that not what you said
`
` earlier?
`
` A Well, why don't you read back what I
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`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
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 24 of 54
`
`Page 81
`
` said.
`
` Q Let me ask you again. You said you
`
` can imagine the term "localized feature" being used
`
` to mean something, and I missed what you said.
`
` A Right, so if I have a, say, a little
`
` window template of what a feature looks like and I
`
` found it in a particular location in the image, I
`
` could refer to that feature as having been
`
` localized.
`
` Q And what does "localized" mean in
`
` that context?
`
` A That I found its location.
`
` Q So it's just referring to the fact
`
` that something has been located?
`
` MR. TURNER: Objection, vague.
`
` A Well, in this case that it's located
`
` in an image.
`
` Q (BY MR. HARBER) And is there any
`
` reason it has to be limited to located in an image?
`
` MR. TURNER: Objection, vague.
`
` A Well, you asked me did I -- could I
`
` -- had I ever thought of it being used and when we
`
` think about features, we think about locating them.
`
` When we think about image features, we think about
`
` locating them in the image. So yes, it would be in
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`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
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 25 of 54
`
`Page 84
`
` testimony.
`
` A What I said was if someone told me
`
` that they had localized a feature in an image, I
`
` would know what that meant.
`
` Q (BY MR. HARBER) And why would you
`
` not understand the same thing if someone told you
`
` they had localized a feature in a 3D rendering?
`
` A First of all, it's not a phrase we
`
` typically would use. We don't localize -- it's not
`
` -- it's not the kind of thing that we would talk
`
` about localization of -- of that feature in that
`
` way, you know.
`
` We estimate 3D positions of things,
`
` but that's not -- you know, when we think about the
`
` localization, we think about finding it in the
`
` image and then figuring out something about it in
`
` the world.
`
` Q Is there a reason why it would be --
`
` why the meaning of the phrase would be different if
`
` you're finding something in an image -- a 2D image
`
` versus in 3D space?
`
` A Typically the notion of localized in
`
` the image is very well-defined. And, you know, so,
`
` you know, we can locate, determine the position of
`
` something in the 3D world. I just don't think
`
`Veritext Legal Solutions
`215-241-1000 ~ 610-434-8588 ~ 302-571-0510 ~ 202-803-8830
`
`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
`
`
`
`Case 6:21-cv-00755-ADA Document 46-1 Filed 03/21/22 Page 26 of 54
`
`Page 85
`
` we've used the phrase "localized feature" to talk
`
` about location -- the position of a 3D -- of
`
` something in the world.
`
` Q Can you not locate a feature in a 3D
`
` world?
`
` A Um, like I certainly can try to
`
` ascertain the position of a thing. You know,
`
` there's this question of what do you mean by
`
` feature. So when we talk about localizing a
`
` feature in a 2D image, the feature is well-defined
`
` -- we know what the feature is. It's this little
`
` patch of image that we're looking for.
`
` I don't really know what localize a
`
` feature in a 3D world means. You have to tell me
`
` what the feature is -- what you mean by feature.
`
` Q Wouldn't -- wouldn't it just be the
`
` feature you're trying to track?
`
` A The thing I'm trying to track?
`
` Q Yes.
`
` A Well, it's the t