`571-272-7822
`
`
`
`
`
`
`IPR2014-00631, Paper No. 49
`IRP2014-00632, Paper No. 15
`June 4, 2015
`
`RECORD OF ORAL HEARING
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`- - - - - -
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`- - - - - -
`
`
`
`CANON, INC.,
`
`Petitioner,
`
`v.
`
`INTELLECTUAL VENTURES II LLC,
`
`Patent Owner.
`
`- - - - - - -
`
`
`Case Nos. IPR2014-00631 and IPR2014-00632
`U.S. Patent No. 7,817,914
`Technology Center 2800
`Oral Hearing Held on Thursday, May 14, 2015
`
`
`
`Before: RICHARD E. RICE, JAMES B. ARPIN (via video link) and
`
`PETER P. CHEN (via video link), Administrative Patent Judges.
`
`The above-entitled matter came on for hearing on Thursday, May 14,
`
`
`
`2015, at 1:06 p.m., in Hearing Room A, taken at the U.S. Patent and
`
`Trademark Office, 600 Dulany Street, Alexandria, Virginia.
`
`
`
`APPEARANCES:
`
`ON BEHALF OF THE PETITIONER:
`
`
`
`
`
`
`
`
`
`
`
`JUSTIN J. OLIVER, ESQ.
`
`STEPHEN YAM, ESQ.
`
`Fitzpatrick, Cella, Harper & Scinto
`
`975 F Street, N.W.
`
`Washington, D.C. 20004-1462
`
`ON BEHALF OF THE PATENT OWNER:
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`HERBERT D. HART, III, ESQ.
`
`STEPHEN J. HAMPTON, Ph.D., ESQ.
`
`DAVID Z. PETTY, ESQ.
`
`McAndrews Held & Malloy Ltd.
`
`500 West Madison Street
`
`34th Floor
`
`Chicago, Illinois 60661
`
`312-775-8000
`
`JAMES R. HIETALA
`
`Director, Post Grant Proceedings
`
`Intellectual Ventures LLC
`
`2
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`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 Nos. IPR2014-00631 and IPR2014-00632
`U.S. Patent No. 7,817,914
`
`
`P R O C E E D I N G S
`
`JUDGE R IC E: Pl ease be seated. Good afternoon.
`
`(1:06 p. m.)
`
`This is the h e aring in IPR2014 -00631. It has be en
`
`consolidated with IPR2014 -00632.
`
`I a m Judge Rice . I a m the onl y jud ge in this Panel
`
`, who is in Alexa ndria toda y. Two of the other judges on the
`
`Panel ar e appea ri ng re motel y: Jud ge Arpin fro m De nver and
`
`Judge Chen fro m Menlo Park, Calif ornia .
`
`You must, becaus e of the re mote p articipation of
`
`the other two judges, speak into th e microphone at the podiu m
`
`for the m to hear you. Please re me mber that.
`
`With that caution, will counsel for Petitioner
`
`introduce the msel ves.
`
`MR. OLI VER: Your Honor, Justin Oliver of
`
`Fitzpatrick, Cella , counsel for P etitioner, Canon.
`
`JUDGE R IC E: And who is with yo u?
`
`MR. OLI VER: St ephen Ya m, also with Fitzpatrick,
`
`Cella.
`
`JUDGE R IC E: Thank you. I f you would please be
`
`seated. Welco me to the P TAB . We're happ y to have you .
`
`Would counsel fo r Patent Owner please introduce
`
`the mselves?
`
`MR. HAR T: Goo d afternoon, Your Honors. I' m
`
`Herbert Hart he re on behalf of Pat e nt Owner as lead counsel.
`
`
`
`3
`
`
`
`Case Nos. IPR2014-00631 and IPR2014-00632
`U.S. Patent No. 7,817,914
`
`
`With me toda y is Steven Ha mpt on, who will be
`
`presenting our ar gu ment. And als o with me toda y a re backup
`
`counsel Ja mes Hi etala and David P ett y.
`
`JUDGE R IC E: Welco me . We' re gl ad to have you.
`
`MR. HAR T: You r Honor, we have a printed cop y
`
`of our de monstrat ives. Would it be helpful f or you t o have
`
`that in front of yo u?
`
`JUDGE R IC E: Wh y don't you prov ide the m a fter
`
`the hearing.
`
`MR. HAR T: Ver y well . Thank yo u ver y much.
`
`JUDGE R IC E: Just a f ew preli min aries before we
`
`start. As set forth in the hearing o rder, each part y will have
`
`45 minutes for ar gu ment.
`
`Please keep in mi nd that whatever you project on
`
`the screen will no t be viewable b y Judges Arpin and Chen. So
`
`it is i mportant wh en you refe r to an exhibit fro m the record to
`
`state the re cord e xhibit number an d the page nu mbe r. An d, for
`
`de monstratives , p lease re fer to the slide or screen number
`
`Petitioner ma y re serve ti me for reb uttal of Patent
`
`Owne r's argu ment, and Patent Owner ma y also r eser ve ti me ,
`
`but only in case P etitioner refers to the motion t o ex clude.
`
`Just one further r e minder , no interruptions. Under
`
`no circu mstanc es is a part y to inter rupt the other par t y while
`
`that part y is pres enting its argu me nts. And if a pa rt y believes
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`
`
`4
`
`
`
`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 Nos. IPR2014-00631 and IPR2014-00632
`U.S. Patent No. 7,817,914
`
`that a de monstrat ive, for exa mple , is objectionable, that
`
`objection ma y be raised onl y durin g the part y's argument ti me.
`
`So with that , let 's hear fro m counsel for Petitioner.
`
`First, do you wish to reserve so me rebuttal ti me?
`
`MR. OLI VER: Yes, Your Honor. 12 minutes,
`
`please.
`
`JUDGE R IC E: Pl ease begin.
`
`MR. OLI V ER: Good afternoon. M a y it please the
`
`Board. I a m Justin Oliver , counsel for Petitioner, Ca non.
`
`The patent at issue toda y in this IP R is the '914
`
`patent. It is dire c ted to a method of auto maticall y ca pturing
`
`an i mage. It has one independent clai m, a method c lai m for
`
`that auto matic ca pture.
`
`The cl ai m r ecites steps including selecting an
`
`auto matic trigger condition. A trigger condition ma y be , for
`
`instance, a sound or i mage content , such as a s mile . It also
`
`includes a step of entering a threshold l evel corresponding to
`
`that trigger condition. For instance, when the trigger
`
`condition is a sound, the threshold level ma y be a de cibel
`
`level.
`
`In addition, the method also recites the step of
`
`auto maticall y ope rating the device to capture an i ma ge upon
`
`detection of meet ing the threshold level. Ho wever , t hese
`
`features wer e all known in the prio r art as established b y the
`
`three pri mar y re f erences applied a gainst independent clai m 1
`
`
`
`5
`
`
`
`Case Nos. IPR2014-00631 and IPR2014-00632
`U.S. Patent No. 7,817,914
`
`and three diff eren t grounds of unpatentability. In f a ct,
`
`ca meras that use auto matic i mage capture were well known
`
`throughout the field.
`
`In response to this fact , I V, the Pat ent Owner , also
`
`known as Intellec tual Ventures, pr oposes a narrow c lai m
`
`construction for both the ter ms , “en tering a threshold level ”
`
`and for “auto mati call y oper ating. ”
`
`IV argued for the se constructions in its preli mina r y
`
`Patent Owner res ponse. This Boar d's decision on institution
`
`denied both constructions. Yet IV' s argu ments ar e s till
`
`predicated on both constructions being adopted. If either
`
`constructi on is not adopted, I V's unpatentability arguments do
`
`not work.
`
`Even still, i f both constructions are adopted, there
`
`are man y issues with the a rgu ments that relate to th e
`
`mischar acterizati on of the a ctual d isclosure of the r eferences .
`
`For instance, I V a cknowledges that the Halpern
`
`referenc e te aches all of the features of independent c lai m 1 ,
`
`but argues that these fe atures appe ar in diff erent e mbodiments
`
`and cannot be use d together.
`
`To find in IV's f a vor would require not onl y
`
`overturning the initial de cision on both ter ms fo r cl ai m
`
`construction, but then buying the c haracterizations o f prior art.
`
`For the reasons we will discuss here toda y, that will not work.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`
`
`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
`
`Case Nos. IPR2014-00631 and IPR2014-00632
`U.S. Patent No. 7,817,914
`
`
`Now, I would like to discuss clai m construction
`
`first. Be fore I do there are two pri ma r y co ncepts I would like
`
`to introduce fro m a te chnological point of view, the first of
`
`which is the t rigger conditions.
`
`As I mentioned with respect to the independent
`
`clai ms , trigger co nditions are cond itions that cause auto matic
`
`i mage capture in a ca me ra .
`
`For instance, the ' 914 patent describes ele ments
`
`such as sounds or changes in sound levels as t rigger
`
`conditions. The ' 914 patent also re fers to i mag e con tent as
`
`being trigger conditions, such as the nu mbe r of fac e s, a
`
`particular fa ce or a s mile appearin g in the field o f v iew of the
`
`ca mera .
`
`Now, coupled wit h the trigger conditions are what
`
`are called threshold levels. Those are levels that indicate that
`
`the trigger condition has actuall y been met . For instance, as
`
`shown in Figure 5 of the '914 paten t, for the exa mpl e trigger
`
`condition of a ch ange in a sound level, the docu me nt refers to
`
`a decibel level be ing entered.
`
`Now, the '914 pat ent does not discuss what
`
`threshold levels are used or how the threshold levels would be
`
`entered when you have mor e co mpl ex trigger conditions, for
`
`instance, a partic ular person's f ace or a s mile .
`
`JUDGE ARP IN: Counselor, turning to Figure 5 of
`
`the '914 patent, it sa ys "set pre fer r ed change threshold."
`
`
`
`7
`
`
`
`Case Nos. IPR2014-00631 and IPR2014-00632
`U.S. Patent No. 7,817,914
`
`
`Are you construing or ar e you equa ting “set ” with
`
`“ente r”?
`
`MR. OLI VER: No, Your Honor, n ot necessaril y. I
`
`think as required b y the clai m construction set forth -- well,
`
`setting could be entering. The clai m construction set forth b y
`
`IV is that a user s pecificall y enters a threshold level . And as
`
`argued b y that side , entering would be entering a spe cific level
`
`itself.
`
`Howeve r, the clai ms do not see m t o be so li mited.
`
`For instance, a ca me ra could auto maticall y put in a t hreshold
`
`level or could be set otherwise. And in that context the clai ms
`
`would be invalidated b y prior art th at actuall y descri bes
`
`selection of a threshold level or en tr y b y the ca mera or even ,
`
`as described in th e '914 patent, the threshold level ca n be
`
`learned b y a ca me ra , so that all a u ser does is a ctuall y set a
`
`mode in which the ca me ra itself de ter mines the thres hold
`
`level.
`
`JUDGE R IC E: Is n't Patent Owner' s position a bit
`
`broader? In the Response on page 12, I understand the Patent
`
`Owne r to argue th at the user enters the threshold level b y
`
`placing the ca me r a in the sel f -learn ing mode with r esp ect to
`
`the self -learning e mbodi ment and, in addition, providing
`
`training that can i nclude the preset threshold levels.
`
`Do you agre e with that?
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`
`
`8
`
`
`
`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 Nos. IPR2014-00631 and IPR2014-00632
`U.S. Patent No. 7,817,914
`
`
`MR. OLI VER: Well, if that is the case, then
`
`docu ments such a s Mishi ma , which I will talk about, would
`
`also b e entering a threshold level.
`
`JUDGE R IC E: Let's focus on the clai m
`
`construction issue.
`
`MR. OLI VER: C ertainl y, Your Honor. Patent
`
`Owne r in its Response actuall y goe s back and forth a s to
`
`whether that self - learning mode is part or not part of the
`
`clai ms . I believe Your Honor refe r red to page 12, was it?
`
`JUDGE R IC E: Yes.
`
`MR. OLI VER: On page 10, which is actuall y
`
`shown in our de monstrative, IV a rgues, or Patent Owner
`
`argues, with resp ect to the s econd or third or other
`
`e mbodi ments do not require a u ser to enter a thr eshold value
`
`corresponding to an auto matic trig ger condition. The y are
`
`outside the scope of clai m 1 . The t hird e mbodi ment being
`
`refer red to the Pa tent Owner being the learning mode .
`
`So in one c ase th e Patent Owner sa ys, well, i f that
`
`doesn't require us er entr y, si mpl y e xclude it fro m the clai ms.
`
`You are cor rect, on page 12, the P atent Owner see ms to
`
`indicate that the l earning mode is p art of the user entr y.
`
`But then if you lo ok again at page 13 of the
`
`Response, about midwa y down be f ore the par agraph above
`
`Section B, it sa ys : Therefor e, this e mbodi ment is no t within
`
`the scope of clai m 1.
`
`
`
`9
`
`
`
`Case Nos. IPR2014-00631 and IPR2014-00632
`U.S. Patent No. 7,817,914
`
`
`So I don't think Patent Owner has q uite decided
`
`whether or not the self -le arning mode is or is not p art of the
`
`scope of clai m 1. What it see ms t o be a rguing is it would like
`
`a clai m construction where onl y user entr y is per mit ted for
`
`entr y of the thr eshold level and, to the extent that the
`
`self-learning mode is part of that , t hen it would be in the
`
`scope of clai m 1.
`
`To the extent this Board does n't find that wa y, that
`
`or an y other e mb odi ment in the specification would si mpl y be
`
`excluded fro m th e clai ms , which it self is a rather te nuous
`
`argu ment for clai m construction.
`
`JUDGE ARP IN: Counselor, just for clarit y, it is
`
`your position that all the e mbodi me nts should be covered b y
`
`the clai m. Is that corre ct?
`
`MR. OLI VER: That's corre ct, You r Honor.
`
`Actuall y turning specificall y to the definition of en tering a
`
`threshold level, starting at page 2 of our de monstrat ives, there
`
`were three constr uctions set forth, one b y Canon at t he outset
`
`of this Petition, which said that th e entr y of a thres hold level
`
`didn't need to be perfor med b y a us er. It could also be set at
`
`the manufacturin g stage.
`
`IV argued for its preli minar y Paten t Owner
`
`response, it has t o be li mited to a user entr y of the t hreshold
`
`level.
`
`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
`
`Case Nos. IPR2014-00631 and IPR2014-00632
`U.S. Patent No. 7,817,914
`
`
`As noted on page 2 of the de monstr atives on the
`
`far le ft, the Boa r d denied that construction , sa ying that the
`
`ter m “ente ring ” c arries its ordinar y and custo mar y meaning
`
`and is not li mited to acts of e ntr y p erfor med b y a use r.
`
`We would cert ainl y agr ee that beca use the
`
`specification describes nu me rous e mbodi ments that don't
`
`require user entr y, that li miting the clai ms to user en tr y would
`
`be i mp roper fo r a t least the reason that the specificat ion
`
`includ es e mbodiments wher e a user does not enter the
`
`threshold level.
`
`But it is even mo re co mplicated than that.
`
`Specificall y, during deposition , Dr . Bovik indicated that there
`
`are certain thresh old levels not only desc ribed in the
`
`specification, but actuall y clai med that could not possibly be
`
`entered b y a user.
`
`Again, looking at page 2 of the de monstratives, Dr .
`
`Bovik during cross -exa mination wa s asked about var ious
`
`trigger conditions, including at a fa ce detection.
`
`And as he stated here: "Face detec tion is used in
`
`the '914 patent. But there's no tea ching of a user pl acing a
`
`threshold on the face dete ction process. I believe th at in ter ms
`
`of fac e detection, a user would not be able to plac e a threshold
`
`upon face detecti on."
`
`So I V's ve r y own -- the Pat ent Owner's ve r y own
`
`declarant has indi cated that for ce rt ain things in the
`
`
`
`11
`
`
`
`Case Nos. IPR2014-00631 and IPR2014-00632
`U.S. Patent No. 7,817,914
`
`specification user entr y is not possible, yet at the s a me ti me
`
`arguing for the cl ai m as being li mi ted to user entr y.
`
`And it actuall y ge ts worse . If you t urn to page 3
`
`of the de monstrat ives, even things recited in the clai ms the
`
`Patent Owner 's de clarant indicated could not be perf or med b y
`
`a user with r espect to user entr y.
`
`As shown on pag e 3 of the de mons tratives we have
`
`an exa mple fro m clai m 5, which de pends fro m independent
`
`clai m 1 through c lai ms 3 and 2 . There with respect to possible
`
`trigger conditions there is a group listed that includes an
`
`object, an ani mal and a person.
`
`The Patent Owne r 's declar ant was a sked about the
`
`ani mal , and the fi rst -- Dr. Bovik i ndicated: "I d on't believe
`
`that the writer of the '914 patent ha s explained an an i mal to be
`
`so mething that is an auto matic trig ger condition."
`
`He went on to sta te: "In fact , I do n't believe that a
`
`user would be able to enter a thres hold level on that."
`
`Thus, in ord e r to adopt the Patent Owne r's
`
`proposed clai m c onstruction of threshold levels only being
`
`entered b y users, that construction would be so nar row that it
`
`would exclude cl ai m 5 as ad mitted to b y Dr. Bovik, which
`
`obviously would be a nonsensical clai m constru ction,
`
`particularl y under the broadest r eas onable interpretation to be
`
`applied b y this B oard.
`
`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
`
`
`
`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 Nos. IPR2014-00631 and IPR2014-00632
`U.S. Patent No. 7,817,914
`
`
`A si milar problem happened with respect to clai m
`
`13, which is shown on page 4 of th e de monstratives. There the
`
`auto matic trigger condition claimed is a prefe rr ed su bject.
`
`Wh en asked abou t that, Dr . Bovik testified:
`
`"Prefe rred sub jec t, to me , would b e a fac e that is re cognized."
`
`And then he went on to sa y: "It would be too co mpl ex for a
`
`person to understand a - - a level of -- to be entered a s a
`
`threshold on that."
`
`So to adopt Paten t Owner's clai m c onstruction on
`
`threshold levels entr y being li mite d to a user , that would be so
`
`narrow that it would exclude dependent clai ms including 5 and
`
`13.
`
`And obviously un der a clai m construction standard
`
`not only would so mething that incl udes e mbodi ment s in the
`
`specification be i mproper , but it would be even furt her
`
`i mproper to introduce a clai m cons truction that actuall y
`
`excludes the depe ndent clai ms .
`
`For these reasons at le ast that clai m construction is
`
`i mproper .
`
`JUDGE C HEN: Counsel, I have a couple questions
`
`for you on the te r m autono mous which appears in the title of
`
`the patent, " Ca me ra Configurable f or Autono mous Operation."
`
`And you cite in your Repl y at page 2 that the '914 p atent
`
`describes a ca mer a auto matic all y e ntering an auto ma tic
`
`response mode an d citing to the De cision to Institute which, in
`
`
`
`13
`
`
`
`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 Nos. IPR2014-00631 and IPR2014-00632
`U.S. Patent No. 7,817,914
`
`turn, re ferenced a spects of colu mn 12 -- I ' m so rr y, c olu mn 9
`
`and colu mn 10 .
`
`Do you think we need to construe t he ter m
`
`autonomous give n its usage in col u mns 9 or 10? And, for
`
`instance, looking at colu mn 9 in th e learning mode
`
`e mbodi ment, column 9, line 29, " af ter so me l earning ti me
`
`period, digital i mage capture device 10 is then read y for an
`
`autonomous oper ation step 220."
`
`Does that suggest that there is so me us er
`
`involvement even in that mode bef ore autono mous o peration
`
`occurs?
`
`MR. OLI VER: Well, the user does certainl y place
`
`the ca me ra -- wel l, there ar e t wo o ptions, one that the user c an
`
`place the c a mer a in an autono mous operation mode where it
`
`learns a thre shold level, but the use r in that case is still not
`
`actuall y entering the threshold level. It's ulti matel y learned
`
`b y the ca me ra.
`
`Further mo re, ther e is the option of actuall y just
`
`turning the ca mer a on and it auto maticall y going into the
`
`auto matic captu re mode.
`
`Our position is simpl y that under t he e mbodi ments
`
`described in the s pecification and under the fe atures recited in
`
`the clai ms, entr y does not have to be b y a use r and, therefore,
`
`with respect to the learning mode , a learning mode is not an
`
`entr y of the thr eshold level b y a us er. The user si mply sets up
`
`
`
`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
`
`Case Nos. IPR2014-00631 and IPR2014-00632
`U.S. Patent No. 7,817,914
`
`the ca me ra so tha t the ca me ra can deter mine the thr eshold
`
`itself.
`
`Si mpl y, the thres hold level does not exist when the
`
`ca mera -- when th e user sets that mode. The ca me ra must
`
`learn that level be fore it can be us ed and, the refore , the user is
`
`not entering it in that situation.
`
`JUDGE C HEN: I s autono mous s yn onymous with
`
`auto matic?
`
`MR. OLI VER: I think autonomous and auto matic
`
`are the sa me , but we' re talking abo ut two diffe rent c lai m ter ms
`
`here. We have en tr y of a threshold level, which is not
`
`required to be auto matic or user en tered, but we sep aratel y
`
`have the clai m ter m auto maticall y o perating, which t alks about
`
`the actual i mage capture.
`
`So there is first b efore i mag e c apture an entr y of a
`
`threshold level, whether that entr y is perfor med b y the ca me ra
`
`itself or b y a user . After that, once the threshold level is set,
`
`the ca me ra works auto maticall y or autonomousl y to deter mine
`
`the precise ti min g of i mage captur e.
`
`And that actuall y brings me t o the s econd clai m
`
`ter m that would b e construed b y I V to make its argu ment , and
`
`that is the definition of auto matica lly operating. Th is is
`
`shown on page 1 of our de monstrat ives.
`
`Now, auto matical l y operating, unlike entering a
`
`threshold level, C anon did not provide a construction in the
`
`
`
`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
`
`Case Nos. IPR2014-00631 and IPR2014-00632
`U.S. Patent No. 7,817,914
`
`Petition inas much as we understand that the plain and
`
`ordinar y - - or cus to mar y and ordinar y meaning would be used.
`
`Howeve r, in the p reli minar y Patent Owner response
`
`and, again, during -- after institution of the trial , Pa tent
`
`Owne r, I V, argue d that the auto ma ticall y operating, which,
`
`again, could be a utonomous operat ion, should be construed to
`
`me an "auto matica lly operating the i mage capture de vice" --
`
`and this is on page 1 of the de mons tratives -- "without
`
`requiri ng further direct operation o f the i mag e c apture device
`
`b y a user ," with t he italicized lang uage being the la nguage
`
`fro m the speci fic ation the Patent Owne r proposes t o read into
`
`the clai ms.
`
`Now, first and for e most under the b roadest
`
`reasonable interpreta tion, it is not appropriate to rea d in
`
`language fro m th e specification to the clai ms .
`
`Second of all , of note, it is interest ing that the
`
`language proposed to be r ead in fr o m the specificati on actuall y
`
`appears in the fie ld of the invention, not the discus sion of the
`
`su mmar y of the al leged invention, but the field in which the
`
`invention lies.
`
`Neither of those a re appropriate things to read in
`
`the clai m, specifi cation alone, let a lone the general f ield of the
`
`invention.
`
`Further mo re --
`
`
`
`16
`
`
`
`Case Nos. IPR2014-00631 and IPR2014-00632
`U.S. Patent No. 7,817,914
`
`
`JUDGE ARP IN: Coun selor, even t hough we ma y
`
`not read the lang uage directl y into the clai ms , don't we have to
`
`consider the language of the clai ms in view of the
`
`specification?
`
`MR. OLI VER: C ertainl y. And I t hink you should
`
`consider both the specification and the prosecuti on history,
`
`exa mples of whic h are also provided on page 1 of our
`
`de monstratives.
`
`For instance, the prosecution histor y, what's shown
`
`in the middle sect ion of page 1 of t he de monstrative , is the
`
`basis that the applicant argued for patentability be fore
`
`allowance of the application, and specificall y argue d: "In
`
`contrast, a mende d clai m 1 provides a method for th e auto matic
`
`capture of i mages wher e the ti ming of i mage capture is
`
`deter mined auto maticall y b y the i mage capture device without
`
`the involve ment o f th e user ."
`
`In that context all that is r equired t o be auto matic
`
`is the precise triggering of i mage c apture, which is n ot done
`
`b y the user but b y the ca me ra whe n it detects that t he trigger
`
`condition is present.
`
`That is sepa rate f ro m sa ying the us er c an't directl y
`
`be involved with the operation of t he ca me ra at all . First, it is
`
`not even clear what direct oper ation would mean an d what the
`
`metes and bounds of that would be.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`
`
`17
`
`
`
`Case Nos. IPR2014-00631 and IPR2014-00632
`U.S. Patent No. 7,817,914
`
`
`Also, if you look at the quote f ro m the
`
`specification that I provided on pa ge 1 o f the de mon strative,
`
`the '914 patent ac counts for a user operating or mani pulating
`
`the ca me ra during the auto matic i mage capture mode .
`
`Specificall y the s pecification sa ys: " Digital i mage
`
`capture device 10 could be hand -he ld or attached to t he bod y
`
`or ot her method wherein the digital i mage capture device 10 is
`
`vaguely pointed in the direction where an i mage cap ture is
`
`desired."
`
`Thus, the '914 ac counts for the use r might be using
`
`the ca me ra, holding the ca me ra, ai ming the ca me ra, but when
`
`it co mes to the ac tual precise ti min g of the triggerin g of i mage
`
`capture, that is d one auto maticall y. And that is actu all y the
`
`true scope of the clai ms .
`
`Reading in without an y further dir ect operation of
`
`the i mage c apture device goes be yo nd that, particula rl y when
`
`it is onl y read in for the purpose of tr ying to distinguish prior
`
`art which clea rl y shows that the ti ming of i mage ca pture is
`
`done auto maticall y.
`
`JUDGE ARP IN: Counselor, turning to the clai ms
`
`for a mo ment and these issues of a uto matic oper ation and the
`
`threshold level.
`
`If we look at clai m 1, the fi rst step is selecting an
`
`auto matic i mage t rigger condition. And the second s tep is
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`
`
`18
`
`
`
`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 Nos. IPR2014-00631 and IPR2014-00632
`U.S. Patent No. 7,817,914
`
`entering a threshold level corresponding to the auto matic
`
`i mage trigger condition.
`
`Clai m 22 depends fro m clai m 1 and sa ys that
`
`"wherein sele cting the auto matic i mage trigger condition
`
`further co mprises selecting more than one auto matic i mage
`
`trigger condition, with corresponding thresholds."
`
`How do we r ead t he two steps of cl ai m 1, selecting
`
`and entering, and the single des cription of the selecti ng step in
`
`clai m 22 together , what is a corres ponding threshold level or a
`
`corresponding threshold as used in clai m 22?
`
`MR. OLI VER: Well, I think the an swer to that is
`
`both in clai ms 1 a nd clai ms 22. Th ere is going to be a
`
`select ion of a trigger condition. Fo r instance, it coul d be a
`
`s mile .
`
`Now, obviously t he ca me ra doesn't know to look
`
`for a s mile. It ha s to be defined in so me algorith mi c wa y or, if
`
`it is a change in s ound, it has to be defined in so me level of,
`
`for instance, deci bels. So the thr e sholds are the cor responding
`
`me asure ments or values b y which whether a s mile a ppears or
`
`whether a sound c hange appears a re measured .
`
`So if it is, for ins tance, a sound ch ange, that's the
`
`trigger condition. The associated t hreshold ma y be a change
`
`of five decibels . For instance, as d escribed in so me of the
`
`prior art refe renc es, when the trigger condition is a s mile , the
`
`actual threshold level ma y be the width of the use r' s -- the
`
`
`
`19
`
`
`
`Case Nos. IPR2014-00631 and IPR2014-00632
`U.S. Patent No. 7,817,914
`
`width of the t arge t's mouth. As a person s miles the ir mouth
`
`gets wider .
`
`Now, with respec t to using multiple trigger
`
`conditions, I beli eve all that mean s is that when mu ltiple
`
`conditions are ch osen, for instance, I ' m not looking for just a
`
`s mile , but a s mile plus a non -blinking status, there h as to be a
`
`corresponding threshold for each .
`
`That doesn't nec e ssaril y r equire th at a use r enters
`
`those threshold levels of what it means to be a s mile or what it
`
`me ans to be not blinking. In fa ct, as Dr. Bovik testified, those
`
`would have to be deter mined in the fac tor y o r in the laborator y
`
`and then provided as cor responding.
`
`So for that reason I think - - sorr y, go ahead. I
`
`think you we re go ing to ask so meth ing else, Your Ho nor?
`
`JUDGE ARP IN: I was just going to ask, with
`
`respect to colu mn 12 of the '914 pa tent, lines 23 thro ugh 30,
`
`here we are discussing the auto mat ic response mode, and we'r e
`
`talking about a corresponding threshold.
`
`Is the cor responding threshold, wit h respect to the
`
`auto matic response mode in colu mn 12, the corresponding
`
`threshold referred to in clai m 22?
`
`MR. OLI VER: And I' m sor r y, You r Honor, which
`
`specific lines aga in wer e the y?
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`
`
`20
`
`
`
`Case Nos. IPR2014-00631 and IPR2014-00632
`U.S. Patent No. 7,817,914
`
`
`JUDGE ARP IN: The lines are column 12 of the
`
`'914 patent, lines 23 to 30, and , in particular , in line 30 whe re
`
`it sa ys " and a cor responding threshold has been sat isfied."
`
`MR. OLI VER: And the question is what is that
`
`corresponding threshold?
`
`JUDGE ARP IN: M y question is do es that
`
`corresponding threshold correspond to the language of clai m
`
`22, or is it the en tered corr esponding threshold, or t he
`
`corresponding thr eshold described in the entering st ep of clai m
`
`1, and is the re a d ifference?
`
`MR. OLI VER: I don't think there's a diff erence . I
`
`think certainl y th at clai m 22 was p erhaps awk wardl y drafted .
`
`But I think in both cases cor responding thresholds refer to the
`
`threshold levels for the associated triggers. And I b elieve
`
`what clai m 22 is t r ying to a chieve i s that instead of u sing a
`
`single trigger wit h a single corr esponding threshold, you have
`
`multiple triggers with multiple cor responding.
`
`But my understanding , although the language of
`
`clai m 22 could be clear er, that we are si mpl y r eciting -- or
`
`what is being reci ted there is si mpl y that there is an entr y of
`
`multiple threshold levels, still not necessaril y b y a u ser, but
`
`you would have t o have corr esponding thr esholds for the
`
`different trigger c onditions, if I understand the question
`
`correctl y.
`
`1
`
`2
`
`3
`
`4
`
`5
`
`6
`
`7
`
`8
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`21
`
`22
`
`23
`
`24
`
`
`
`21
`
`
`
`Case Nos. IPR2014-00631 and IPR2014-00632
`U.S. Patent No. 7,817,914
`
`
`JUDGE ARP IN: So you read clai m 22 as
`
`enco mpassing the entering step eve n though it doesn't
`
`specificall y sa y e ntering?
`
`MR. OLI VER: Well, I think it is a wkwa rdl y
`
`writ ten because it sa ys selecting tri gger conditions with
`
`corresponding thresholds. I think, in the wa y these c a meras
`
`actuall y oper ate, you select a t rigger condition , and there is a
`
`corresponding threshold that is entered.
`
`The cl ai m 1 is cer tainly more pre ci se, that the re is
`
`a selection and an entr y. Cl ai m 22 certainl y melds t hat
`
`language and mak es it see m like it is part of the sa me first
`
`step, but I think that is just awk wa rd drafting and th at the
`
`threshold levels of clai m 22 would be entered in the sa me wa y
`
`as clai m