`
`
`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`SAMSUNG ELECTRONICS CO., LTD; AND SAMSUNG
`ELECTRONICS AMERICA, INC.,
`Petitioners,
`
`v.
`
`IMAGE PROCESSING TECHNOLOGIES, LLC,
`Patent Owner.
`____________
`
`IPR2017-01190 (Patent 6,717,518 B1)
`IPR2017-01218 (Patent 8,983,134 B2)
`____________
`
`Record of Oral Hearing
`Held: June 29, 2018
`____________
`
`
`
`
`Before JONI Y. CHANG, MIRIAM L. QUINN, and SHEILA F.
`McSHANE, Administrative Patent Judges.
`
`
`
`
`
`
`
`
`
`
`
`IPR2017-01190 (Patent 6,717,518 B1)
`IPR2017-01218 (Patent 8,983,134 B2)
`
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONER:
`
`
`NICHOLAS WHILT, ESQUIRE
`BRIAN COOK, ESQUIRE
`O'Melveny & Myers, LLP
`400 South Hope Street
`Los Angeles, California 90071
`
`ON BEHALF OF PATENT OWNER:
`
`
`CHRIS J. COULSON, ESQUIRE
`MICHAEL N. ZACHARY, ESQUIRE
`Bunsow DeMory, LLP
`101 Brambach Road
`Scarsdale, New York 10883
`
`
`
`
`The above-entitled matter came on for hearing on Friday, June
`
`
`29, 2018, commencing at 10:00 a.m., at the U.S. Patent and Trademark
`Office, 600 Dulany Street, Alexandria, Virginia.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
` 2
`
`
`
`IPR2017-01190 (Patent 6,717,518 B1)
`IPR2017-01218 (Patent 8,983,134 B2)
`
`
`P R O C E E D I N G S
`- - - - -
`JUDGE McSHANE: Good morning, everyone. We are here to
`conduct the hearings in IPR2017-01190 and IPR2017-01218. If we could
`have appearances, please, who is here from petitioner?
`MR. WHILT: Nick Whilt for petitioner, Samsung. And I'm also
`joined by my colleague, Brian Cook.
`JUDGE McSHANE: And who do we have from patent owner?
`MR. COULSON: Good morning, Your Honors. My name is
`Chris Coulson with Bunsow DeMory on behalf of patent owner, Image
`Processing Technologies. With me, also from Bunsow DeMory, is Michael
`Zachary.
`JUDGE McSHANE: So we sent out an order on this oral hearing
`or these oral hearings, and in that we included the general guidance for
`today. And what we are going to do is we are going to have the arguments
`for the 1190 case first, and then we are going to follow it with the 1218 case.
`The way that we are going to handle each case is the petitioner is going to
`present its arguments. You are going to have 30 minutes for that. You can
`reserve rebuttal time. Patent owner is going to provide its response, and then
`petitioner can use any rebuttal time it has.
`A couple of reminders here, I have a feeling that you folks have
`been here before PTAB before, but just a few reminders. If you are using
`demonstratives, please call out the demonstrative numbers. It helps the
`record, of course, and we do have Judge Quinn attending remotely, and it
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`
`
` 3
`
`
`
`IPR2017-01190 (Patent 6,717,518 B1)
`IPR2017-01218 (Patent 8,983,134 B2)
`
`helps her even more. And if you could, please, use the microphone as well,
`it helps the court reporter as well as Judge Quinn and us to get a clear record.
`And please, if you have objections, please don't interrupt the other
`party while they are speaking. Wait until you have an opportunity to speak
`and then present your objections.
`Any questions on any of that?
`MR. WHILT: No, Your Honor.
`MR. COULSON: Your Honor, I do have a question. Do we have
`a total for both IPRs of 30 minutes? That was my understanding of the
`order. Or is it a total of hour that's broken into 30 minutes for each IPR?
`JUDGE McSHANE: It was per case, right. So I believe that was
`clear in the order.
`MR. COULSON: That's fine, Your Honor. I just wanted to
`
`clarify.
`
`MR. WHILT: My understanding, Your Honor, was that it was an
`hour for each case and 30 minutes for each side per case.
`JUDGE McSHANE: Yeah, it says each party will be permitted
`30 minutes per proceeding.
`MR. COULSON: Thank you, Your Honor.
`JUDGE McSHANE: So that's the deal so that you get 30 minutes
`per proceeding. He gets 30 minutes per proceeding, okay. And they are two
`separate proceedings.
`MR. COULSON: Thank you, Your Honor.
`JUDGE McSHANE: So petitioner, if you would like to proceed,
`please, and do you want to reserve rebuttal time?
`
`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
`
`
`
`IPR2017-01190 (Patent 6,717,518 B1)
`IPR2017-01218 (Patent 8,983,134 B2)
`
`
`MR. WHILT: Yes, Your Honor. May I reserve 8 minutes for
`rebuttal.
`JUDGE McSHANE: What I'm going to do is put 30 minutes on
`the clock and then we'll see how you make your progress and we'll work
`from there.
`MR. WHILT: Good morning, Your Honors. My name is Nick
`Whilt, and I'm representing petitioner, Samsung, in IPR2017-1190. The
`only claim at issue in this IPR is claim 39 of the '518 patent. So we are only
`going to be focusing on that claim today.
`JUDGE CHANG: May I interrupt you. Do you have any
`demonstratives for the court reporter?
`MR. WHILT: I apologize, Your Honor.
`JUDGE CHANG: Oh, okay. I thought all that was for the court
`reporter. Thank you very much.
`MR. WHILT: So I'm going to jump to slide 18. I would like to
`begin by addressing claim construction. As Samsung addressed in --
`explained in its briefs, claim construction is largely unnecessary to deciding
`this IPR. And therefore, unless the Board has questions regarding other
`claim construction issues, I only plan to address the patent owner's proposed
`construction for element 39D which is quoted on slide 18.
`The reason I would like to address this proposed -- the patent
`owner's construction of this term is because the Board rejected the proposed
`construction in the institution decision, and if the Board maintains its
`interpretation, then there's no dispute that each one of the prior art references
`in this IPR discloses element 39D. And the reason for that is because the
`
`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
`
`
`
`
`
` 5
`
`
`
`IPR2017-01190 (Patent 6,717,518 B1)
`IPR2017-01218 (Patent 8,983,134 B2)
`
`patent owner's arguments relating to this element are purely based on its
`incorrect claim construction.
`JUDGE QUINN: Counsel, can I ask you a question about this
`claim. It seems to me that there are multiple terms here that use the word
`"characteristic," but it's not consistent within the claim what characteristic
`we are referring to. And perhaps the confusion is that you have the
`characteristic of the face and then a different type of characteristic which are
`the characteristics corresponding to the feature to be detected. So is it your
`proposal that the characteristics all refer to the same characteristic as the
`construction or that there may be two different characteristics that we are
`looking at in this claim?
`MR. WHILT: Your Honor, let me go to the claim language and
`walk through this. I think it's actually two different characteristics. If you
`go to slide -- so if you look at, I have got on this slide 5, the claim broken up
`by elements. If you look at claim element 39B --
`JUDGE McSHANE: What number is that?
`MR. WHILT: This is slide 5. If you look at element 39B, what it
`identifies is it's got an identifying characteristic of a face other than the
`feature to be detected. So first in this claim you are identifying a
`characteristic of the face. And then after that, if you go down to element C,
`what it shows is you are going to find a portion of the image of the face
`based on using that first characteristic that you identified. So you have the
`first characteristic you are identifying in B use what's called an
`anthropomorphic model in element C to identify a different portion of the
`image. And then after you have identified the portion, in element D you
`
`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
`
`
`
`
`
` 6
`
`
`
`IPR2017-01190 (Patent 6,717,518 B1)
`IPR2017-01218 (Patent 8,983,134 B2)
`
`select pixels of the portion of the image having characteristics corresponding
`to the feature to be detected.
`So there's two features that are discussed in this claim, but it's
`different characteristics because it's two different things you are looking at.
`JUDGE QUINN: Right. So you may have characteristics of the
`face, but that may not necessarily be the same characteristics you are looking
`at for selecting the pixels. And I think that's the confusion here, that you
`may be selecting pixels based on their luminance, hue, saturation, whatever
`that may be, but that may correspond to a feature of the face, but the
`characteristics of the face are separate and distinct from these characteristics.
`MR. WHILT: Right. So for example, for element B you may have
`a characteristic of the face, like the center of the face where you have your
`nose. But when you get to element D where you are looking for pixels
`corresponding to the feature to be detected, you may be looking for pixels
`closest to the center of your face or dark pixels near the center of your face
`that represent the characteristics of the pupil or iris.
`JUDGE QUINN: The way I read this limitation, selecting pixels
`of the portions of the image, so you are going to analyze the pixels of that
`portion that you have already selected out of the whole image, right?
`MR. WHILT: That's right.
`JUDGE QUINN: And that portion of the image is already
`identifying some facial characteristics where it's related to that facial
`characteristic, right?
`MR. WHILT: That's right, Your Honor. The identifying a portion
`of the image, the portion of the image comprises, so it includes the feature to
`
`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
`
`
`
`
`
` 7
`
`
`
`IPR2017-01190 (Patent 6,717,518 B1)
`IPR2017-01218 (Patent 8,983,134 B2)
`
`be detected. And then within that portion you are identifying pixels having a
`characteristic of the feature to be detected.
`JUDGE QUINN: So break it down for me. You have the picture,
`then you have the portion of the image, and then you are selecting pixels in
`that portion of the image.
`MR. WHILT: That's right.
`JUDGE QUINN: And for me the characteristics corresponding to
`the feature to be detected really refer to the pixels themselves, whether they
`have a certain luminance, a certain hue, a certain intensity, whatever that
`may be that you are looking for; isn't that right?
`MR. WHILT: That's right. There's a lot of different characteristics
`other than those that could correspond to the pixels, but those are examples,
`that's correct.
`JUDGE QUINN: Okay.
`MR. WHILT: But it could include other pixels as well. One of the
`things I just want to point out is in this selecting step -- actually, if we could
`go to slide 20, in this selecting step, it's not only limited to -- it doesn't say
`that you have to only select pixels having characteristics of the feature to be
`detected.
`As the Board found in the institution decision in interpreting this
`limitation -- actually, let me go to what the Board found just as a reminder.
`In the institution decision, the Board explained that we do not agree that the
`claim limitation, element 39D, and I'm on slide 19, precludes selection of
`pixels that are not of the feature itself. So in the institution decision, the
`Board rejected the patent owner's argument that this element is limited to
`
`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
`
`
`
`IPR2017-01190 (Patent 6,717,518 B1)
`IPR2017-01218 (Patent 8,983,134 B2)
`
`selecting only iris, pupil or cornea pixels. Instead, what the Board
`explained, and I have got this quoted on slide 19 --
`JUDGE QUINN: Well, I understand what we said in the
`institution, but now we have briefings. So we can alter our decision
`depending on how we see the claims based on what the patent owner has
`argued to us. But if you are saying that you take an image, you find the eye,
`because that's the area, the characteristic of the area you want, right, and
`then you are selecting pixels within that portion, you are saying that if I'm
`interested in not just the eye but the pupil and I catch pixels that refer to the
`eyelid or the cornea that you are still meeting the claim because the pixels
`are being selected based -- because they are in that window.
`MR. WHILT: That's right, Your Honor. And I think the patent
`makes it clear that when you are selecting the pixels, you have to select
`pixels having characteristics corresponding to what you are trying to detect,
`like you are saying, like you are looking for the pupil, you might pick up
`dark pixels, but you might pick up other pixels as well as part of that
`process.
`
`JUDGE QUINN: Okay. But you have to have -- but the pixels
`you are selecting, even if they are not of the pupil, have to have the
`characteristics that you selected that correspond to a pupil?
`MR. WHILT: Your Honor, the way that I interpret this element
`and I believe the way the Board interpreted it in the institution decision was
`that you do have to select pixels having properties corresponding to the
`feature to be detected such as the pupil, but you could select other pixels as
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`
`
` 9
`
`
`
`IPR2017-01190 (Patent 6,717,518 B1)
`IPR2017-01218 (Patent 8,983,134 B2)
`
`well. The claim doesn't say just because you select like one additional pixel
`that might not have a characteristic that somehow that doesn't qualify.
`JUDGE QUINN: So are you saying that because you select the
`eye, and the characteristic you are interested in is within the eye, then
`selecting the pixels, this limitation will be met because somewhere in the eye
`is the pupil?
`MR. WHILT: Your Honor, yeah, I believe it would be met under
`
`that.
`
`JUDGE QUINN: But if the portion of the image is the eye and the
`claim says selecting pixels of the portion of the image that have the
`characteristics and the actual feature you want is the pupil, then you must be
`selecting pixels within the eye that match characteristics you are looking for
`that correspond to only the pupil.
`MR. WHILT: Well, Your Honor, I think this is illustrated in, for
`example, like Figure 36 of the '518 patent that I have on slide 20. So for
`example, this is an embodiment from the '518 patent, and on the left you see
`Figure 36 where you have the pupil that's labeled 432. So this would be
`with within what you are talking about, within a portion of the image that
`was selected.
`Now, in the patent what it provides as an example is you can see
`on the left on the X and Y axes you have the histograms, right, and you can
`see that those histograms include the pixels of the pupil. But you can also
`see that the histograms include pixels other than pupil pixels, because you
`can see along the Y axis, the vertical extent of the histogram exceeds the
`pupil and similarly along the horizontal axis. So you are selecting -- you do
`
`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
`
`
`
`
`
` 10
`
`
`
`IPR2017-01190 (Patent 6,717,518 B1)
`IPR2017-01218 (Patent 8,983,134 B2)
`
`have to select pixels having characteristics of the feature to be detected, but
`that doesn't mean that you couldn't have some additional pixels in included
`in the selection because the claim doesn't say only in it.
`JUDGE QUINN: Right. I agree that the claim doesn't say only,
`but it really -- it defines the claim as to which pixels you select to perform
`the histogram. And I think what you are saying is that the specification
`supports the broader construction of this to mean it can have -- the histogram
`will show pixels that are outside of the window, therefore, your selection
`must not be only of the pixels that have those characteristics.
`MR. WHILT: That's right. It can include other pixels as well. So
`like I say, we believe this is supported by the specification which shows in
`not just Figure 36 but in Figure 27 where you have a feature to be detected
`such as the pupil, but it includes other pixels as well.
`If we go to slide 21, what the patent owner is really doing in the
`briefs, and they don't really hide this fact, is they are trying to argue that the
`claims essentially should be rewritten to include the word "only" such that
`the claim requires only selecting pixels of the pupil or only selecting pixels
`having characteristics of the feature to be detected. But that's just not what
`the claim says. It doesn't say only. It wasn't drafted that way. So that's one
`reason why we believe that the Board got it right in the institution decision
`when it was interpreting this limitation.
`If we could go to -- one thing I would also like to point out about
`this element is that even if the element, this element were interpreted more
`narrowly such as if it was interpreted to require forming a histogram only of
`pixels having characteristics of the feature to be detected, the prior art would
`
`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
`
`
`
`
`
` 11
`
`
`
`IPR2017-01190 (Patent 6,717,518 B1)
`IPR2017-01218 (Patent 8,983,134 B2)
`
`still disclose this element under that interpretation because in everyone prior
`art reference in this IPR, the prior art discloses forming a histogram only of
`pixels that are selected and have characteristics of the feature to be detected,
`in particular pupil in the case of the prior art references.
`So if we could go to slide 27, unless there's any other questions on
`claim construction, I was going to move to ground 8. So in ground 8, this
`involves the combination of the reference Eriksson in view of Stringa.
`There's only two disputes that the patent owner raises with respect to this
`ground. The first one is that they argue that these references don't disclose a
`histogram. And second, they argue that these references don't disclose
`performing a histogram of the selected pixels. This is the claim construction
`issue we were just discussing a second ago. And so I want to first start by
`addressing the patent owner's argument about these references supposedly
`not disclosing a histogram.
`If we can go to slide 29, now, in the petition, Samsung pointed out
`that Eriksson discloses what's called, quote, a horizontal histogram across
`the pupil. And it's called a histogram matches the -- it's the exact language
`that's used in the claim. And it's even shown in Figure 5, and it's labeled a
`histogram.
`Now, after we filed our petition, the patent owner admitted that
`this is a histogram. And I have got in the yellow box on this slide an excerpt
`from their preliminary response, page 30, where they admitted repeatedly
`that this is a histogram. Only recently have they changed their minds and
`now they are arguing that it's not a histogram.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`
`
` 12
`
`
`
`IPR2017-01190 (Patent 6,717,518 B1)
`IPR2017-01218 (Patent 8,983,134 B2)
`
`
`JUDGE QUINN: Well, I think they are quoting Eriksson here.
`They are not admitting anything. That's what it says, right?
`MR. WHILT: That's right. It specifically calls it a histogram.
`That's right.
`JUDGE QUINN: Okay. Just because they are quoting it doesn't
`mean they are admitting it is, because they vehemently dispute that it is. So
`I don't think we should say that it is an admission.
`MR. WHILT: Yeah, I just wanted to point out that they called it a
`histogram in the preliminary response. But I appreciate what you are saying.
`JUDGE QUINN: Okay.
`MR. WHILT: And if we look at Eriksson, it's -- there should be no
`question that it discloses a histogram. First of all, Figure 5 is labeled a
`histogram. And by the way, this is an IEEE peer-reviewed article, and
`there's no reason that the authors would have gotten this wrong when they
`were identifying this as a histogram.
`JUDGE McSHANE: I'm sorry, this slide 31, right?
`MR. WHILT: I'm sorry, slide 30 we are on. And in fact, if you
`look at the text of Eriksson, it repeatedly explains that it uses a histogram to
`identify pupils. For example, on this slide you can say that it says on
`page 317, quote, we need a robust way to determine if the eyes are open or
`closed, so we developed a method that looks at the horizontal histogram
`across the pupil. Then it has a whole section called the Horizontal
`Histogram Across the Pupil section where it describes using a histogram to
`detect the pupil.
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`
`
` 13
`
`
`
`IPR2017-01190 (Patent 6,717,518 B1)
`IPR2017-01218 (Patent 8,983,134 B2)
`
`
`If we can go to slide 32, please, now, despite it calling it a
`histogram, the patent owner now argues that Eriksson's Figure 5 does not
`disclose a histogram. And the patent owner argues that it simply shows the
`intensity values for a single line of pixels through the pupil.
`But I just want to point out, first of all, that even if the patent
`owner were correct, that it's essentially unrebutted that Figure 5 shows a
`histogram. Samsung's expert, Dr. Hart, explained that this is a histogram.
`The patent owner hasn't presented any expert testimony in the IPR, including
`any expert testimony rebutting Dr. Hart's opinions or supporting their
`opinion that it's not a histogram. And as I mentioned before, the authors of
`Eriksson, this IEEE article, called it a histogram. There's no reason that this
`would not be a histogram.
`JUDGE QUINN: Counsel, I think from what I read from the
`cross-examination of your expert that the Figure 5 to a person of skill in the
`art could instead of a histogram be a scan line or a plot of a scan line which
`would not be consistent with the customary meaning of histogram to a
`person of skill in the art. Can you explain that?
`MR. WHILT: Sure. A couple things. First of all, Dr. Hart
`explained that under his interpretation and the way that the patent owners
`explain it, that it would be considered a histogram. He explained in detail
`that that is one way that histograms are made, that it's one way you can make
`it is by determining the frequency of photons that are in a particular area in a
`particular amount of time. That's a way to create a histogram.
`But I just want to point out that the way that the patent owner is
`interpreting Eriksson is just not right. It's just not -- it's inconsistent with
`
`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
`
`
`
`
`
` 14
`
`
`
`IPR2017-01190 (Patent 6,717,518 B1)
`IPR2017-01218 (Patent 8,983,134 B2)
`
`what Figure 5 actually shows because if you look at Figure 5, what it shows
`is you can see on the left-hand side is that the histogram has this curved
`shape that tracks both the horizontal and vertical dimensions of the pupil and
`iris. You wouldn't get that if this was simply a plot of intensity values across
`the eye. Instead it would be completely flat underneath the pupil because it's
`all black. It would be completely flat on the iris, completely flat in the white
`part of the eye. You just wouldn't get this shape that tracks the
`dimensionality of the pupil and iris if it was what the patent owner is saying
`that it is.
`JUDGE McSHANE: Counsel, on that issue, do you have any
`expert testimony on that?
`MR. WHILT: Yeah. So a couple places --
`JUDGE McSHANE: You can give them to me -- if you have them
`right there, read them off.
`MR. WHILT: There is. You want me to give it to you in a little
`
`bit?
`
`JUDGE McSHANE: Yeah, you can.
`JUDGE QUINN: Also, if you have testimony of the interpretation
`of histogram to a person of skill in the art, I read in the examination of your
`expert that he was using that, quote/unquote, meaning of histogram to a
`person of skill in the art, and there was no statement of what that is. So do
`you have evidence of that?
`MR. WHILT: We do. Dr. Hart actually has extensive explanation
`in his declaration about what that means. That is in paragraphs 25 to 30 of
`his declaration. There's some other testimony I'll find later, but that's, for
`
`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
`
`
`
`
`
` 15
`
`
`
`IPR2017-01190 (Patent 6,717,518 B1)
`IPR2017-01218 (Patent 8,983,134 B2)
`
`example, where he explains in depth what histograms are and where he
`applies the prior art.
`So unless there's any other questions on this slide, we can move to
`slide 34. The patent owner also argues that Eriksson doesn't use histograms
`to detect the pupil. And their main argument in support of this is that it's
`used as something called a matching formula or matching function instead of
`histograms. This is just wrong for several reasons. First of all, I just want to
`point out that this discussion of this matching function -- I'm sorry, I'm on
`slide 34. This discussion of the matching function is actually in the section
`of Eriksson called the Horizontal Histogram Across the Pupil section. So it
`would be really odd for Eriksson to discuss this matching function in the
`histogram section yet for it to have nothing to do with histograms.
`In fact, as Samsung explained in its briefs, what Eriksson does, it
`doesn't use this matching function in place of the histograms. Instead, what
`it does is it uses this matching function to analyze the histogram. And you
`can see this, I have got quoted on slide 34 an excerpt from page 18 of
`Eriksson, this is in the histogram section, where it explains that it uses this
`matching function to analyze the curve and determine if there's a match or
`not a match, to determine if the pupil is there so it can determine if the eye is
`open or closed.
`Your Honor, unless there's some other questions, I would like to
`reserve the last 7, 8 minutes.
`JUDGE QUINN: Question on this matching issue. Do you read
`the '518 patent as using something similar to, you know, comparing the
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`18
`19
`20
`21
`22
`23
`24
`
`
`
`
`
` 16
`
`
`
`IPR2017-01190 (Patent 6,717,518 B1)
`IPR2017-01218 (Patent 8,983,134 B2)
`
`image with another image to detect areas? I thought I read somewhere that
`it does something like that.
`MR. WHILT: You mean whether it does something similar to like
`what we are discussing in Eriksson?
`JUDGE QUINN: Correct.
`MR. WHILT: Yeah, Your Honor, it does have to perform analysis
`to determine if the eye is open or closed in Eriksson as well. You can't just
`look at a histogram and just know whether it's open or closed. By definition
`you'd have to do that. And the claim, by the way, doesn't require -- there's
`nothing in the claim that precludes you from generating the histogram and
`then precludes any other processing of the histogram to determine the
`characteristic you are looking for. There's nothing that says you can't take a
`histogram and do nothing with it or precludes additional operations. That
`actually wouldn't make sense because the patent has to look at it.
`JUDGE QUINN: Well, I mean, in the patent once you detect
`peaks or the particular characteristic in the histogram that you are looking
`for, you have some thresholds you are going to be matching that with or
`comparing it to, to see if you are either within the range or there's a match
`between that threshold or not, right?
`MR. WHILT: That's correct, Your Honor. And it specifically
`does that in the patent to determine whether the eye is open or closed. That's
`how it does it. It looks at the shape or various -- and does a particular
`analysis on it to make that determination.
`JUDGE QUINN: Right. But it is expecting a certain value there if
`it is open, and if that's not there, then it's not open, right?
`
`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
`
`
`
`IPR2017-01190 (Patent 6,717,518 B1)
`IPR2017-01218 (Patent 8,983,134 B2)
`
`
`MR. WHILT: Right. That's right, Your Honor.
`JUDGE QUINN: Okay.
`MR. WHILT: Any other questions? Otherwise I'll reserve.
`JUDGE McSHANE: Thank you. You have 6 minutes for rebuttal.
`MR. COULSON: Your Honors, again, I'm Chris Coulson on
`behalf of patent owners. I would like to begin my presentation by
`responding to the point raised by Judge Quinn early in the examination of
`petitioner regarding the claim language. Then I'll go through our claim --
`JUDGE QUINN: I wasn't examining petitioner, by the way. I just
`want that on the record. I just ask questions.
`MR. COULSON: I apologize, Your Honor. The earlier
`discussion. The point you raised. I'm trying to be responsive to the point
`you raised, Your Honor. Then I'll discuss claim construction. And then I
`would also like to proceed to the discussion of the histogram, including
`responding specifically to the paragraphs that were just mentioned in the
`declaration of the expert for the other side and showing that the testimony
`that we have obtained and is on our demonstratives specifically rebuts each
`point in those paragraphs.
`I'll begin on slide 8 of patent owner's demonstratives which are
`Exhibit 2016 in this proceeding. And I'll also be referring later to
`Exhibit 1020 for one particular slide of petitioner's demonstratives.
`I'm showing the claim language on slide 8 of exhibit -- of our
`Exhibit 2016. I agree with the discussion earlier there are two separate
`characteristics here. And the characteristic that I would like to discuss is the
`characteristic of the feature to be detected that's mentioned in element C, D
`
`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
`
`
`
`
`
` 18
`
`
`
`IPR2017-01190 (Patent 6,717,518 B1)
`IPR2017-01218 (Patent 8,983,134 B2)
`
`and F. I have a blow-up of those elements -- most of those elements on slide
`14 of our demonstratives. I'll go to slide 14.
`In section C of the patent -- and let me back up here and say, so the
`claim, as we've shown in the briefing, the Board has made its initial decision
`in institution, and I'm going through what we've shown in the briefing that
`the claim language itself requires that the pixels that are selected have
`characteristics corresponding to the feature to be detected, which is the plain
`language of the claim. And the specification also supports this construction.
`So I'll be going through and explaining those points.
`Returning to slide 14, the claim elements C, D and E, claim
`element C requires identifying a portion of an image of a face comprising
`the feature to be detected. As mentioned by petitioner's counsel who relied
`on this word "comprising," that's a point we make as well in the briefing,
`Your Honor, that you are selecting an area that includes -- that's how we
`interpret this as well -- the feature to be detected.
`So step D is something different than just selecting a window that
`includes the feature to be detected. It is, step D is selecting pixels of the
`portion of the image that was identified in C having characteristics
`corresponding to the feature to be detected, for example, a dark pixel for
`pupils.
`
`Essentially there was an allegation, I believe, that patent owner
`admits or that the prior art meets the claim limitation even under patent
`owner's interpretation. The only way that would be possible is if selecting a
`window that includes the feature to be detected is D. In other words, if a
`feature of the feature to be detected is that it's in a window and that window
`
`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
`
`
`
`IPR2017-01190 (Patent 6,717,518 B1)
`IPR2017-01218 (Patent 8,983,134 B2)
`
`could be any size, that's the only way that the prior art would read on this
`claim.
`
`Continuing with the claim language --
`JUDGE QUINN: So in your interpretation, identifying a portion
`of the image of the face, so that in and of itself means the window cannot be
`the entire face. It has to be some portion within the face of that image,
`right?
`
`MR. COULSON: That sounds right, Your Honor, that you are
`identifying -- we are just working with the plain language here, Your Honor.
`You are identifying a portion of the image, part of the image of the face that
`comprises the feature to be detected. So if you are looking for a feature of
`the iris, pupil or cornea, which is
Accessing this document will incur an additional charge of $.
After purchase, you can access this document again without charge.
Accept $ Charge
Still Working On It
This document is taking longer than usual to download. This can happen if we need to contact the court directly to obtain the document and their servers are running slowly.
Give it another minute or two to complete, and then try the refresh button.
A few More Minutes ... Still Working
It can take up to 5 minutes for us to download a document if the court servers are running slowly.
Thank you for your continued patience.
This document could not be displayed.
We could not find this document within its docket. Please go back to the docket page and check the link. If that does not work, go back to the docket and refresh it to pull the newest information.
Your account does not support viewing this document.
You need a Paid Account to view this document. Click here to change your account type.
Your account does not support viewing this document.
Set your membership
status to view this document.
With a Docket Alarm membership, you'll
get a whole lot more, including:
- Up-to-date information for this case.
- Email alerts whenever there is an update.
- Full text search for other cases.
- Get email alerts whenever a new case matches your search.
One Moment Please
The filing “” is large (MB) and is being downloaded.
Please refresh this page in a few minutes to see if the filing has been downloaded. The filing will also be emailed to you when the download completes.
Your document is on its way!
If you do not receive the document in five minutes, contact support at support@docketalarm.com.
Sealed Document
We are unable to display this document, it may be under a court ordered seal.
If you have proper credentials to access the file, you may proceed directly to the court's system using your government issued username and password.
Access Government Site
