Declaration for Inter Partes Review of USPN 8,922,628
`
`IN THE UNITED STATES PATENT AND TRADEMARK OFFICE
`
`
`
`
`
`In re Inter Partes Review of:
`U.S. Patent No. 8,922,628
`Issued: December 30, 2014
`Application No.: 12/874,190
`Filing Date: September 1, 2010
`
`For: System and Process for Transforming Two-Dimensional Images Into
`Three-Dimensional Images
`
`FILED VIA PRPS
`
`)
`)
`)
`)
`)
`
`
`
`DECLARATION OF CARLOS VAZQUEZ, Ph.D. IN SUPPORT OF
`PETITION FOR INTER PARTES REVIEW
`OF U.S. PATENT NO. 8,922,628
`
`
`
`
`
`
`
`Legend3D, Inc.
`Exhibit 1011-0001
`
`

`
`Declaration for Inter Partes Review of USPN 8,922,628
`
`TABLE OF CONTENTS
`
`
`
`
`I. INTRODUCTION ............................................................................................. 3
`II. QUALIFICATIONS ......................................................................................... 5
`III. UNDERSTANDING OF THE GOVERNING LAW ................................... 9
`A. LEVEL OF ORDINARY SKILL IN THE ART ......................................................... 9
`B. INVALIDITY BY ANTICIPATION OR OBVIOUSNESS ............................................ 11
`C. INTERPRETING CLAIMS BEFORE THE PATENT OFFICE ..................................... 12
`IV. OVERVIEW OF THE ‘628 PATENT ......................................................... 13
`A. FIELD OF THE PURPORTED INVENTION ............................................................ 13
`B. PURPORTED NOVELTY FROM THE USE OF IMAGE CHARACTERISTICS FOR
`CONVERSION ........................................................................................................ 16
`C. ADDITIONAL PURPORTED NOVELTY FROM THE USE OF A MULTIDIMENSIONAL
`VECTOR ................................................................................................................ 17
`V. ‘628 PATENT PROSECUTION HISTORY ................................................ 21
`VI. CLAIM CONSTRUCTION .......................................................................... 21
`A. “DEPTH SELECTION MASK” .............................................................................. 23
`B. “VECTOR FIELD” ............................................................................................. 24
`VII. TECHNOLOGY BACKGROUND ............................................................. 25
`VIII. OVERVIEW OF THE PRIOR ART RELIED UPON ............................ 39
`A. U.S. PATENT NO. 7,573,475 ........................................................................... 41
`B. U.S. PATENT NO. 8,488,868 ........................................................................... 44
`C. U.S. PATENT NO. 8,213,711 ........................................................................... 48
`IX. MOTIVATION TO COMBINE TAM ‘868 WITH TAM ‘711 AND
`SULLIVAN ‘475 ................................................................................................... 49
`A. ALL REFERENCES IN THE SAME FIELD AND ADDRESS THE SAME PROBLEM .... 49
`B. MOTIVATION TO COMBINE SULLIVAN ‘475 WITH THE DEPTH FROM IMAGE
`FEATURE FUNCTIONALITY OF TAM ‘868 AND TAM ‘711 ....................................... 50
`C. SULLIVAN IS COMPATIBLE WITH METHODS THAT USE DEPTH SELECTION MASKS
`BASED ON IMAGE FEATURES TO DERIVE DEPTH ..................................................... 52
`D. MOTIVATION TO COMBINE THE VECTOR FIELD OF SULLIVAN ‘475 WITH TAM
`‘868 TO SUPPORT TOED-IN CAMERA CONFIGURATIONS ........................................ 52
`E. TAM ‘868, TAM ‘711 AND SULLIVAN ‘475 ARE COMPATIBLE ........................ 53
`
`1
`
`Legend3D, Inc.
`Exhibit 1011-0002
`
`

`
`Declaration for Inter Partes Review of USPN 8,922,628
`
`2
`
`
`
`
`
`F. COMBINING SULLIVAN’S MULTIDIMENSIONAL VECTOR WITH TAM IS A SIMPLE
`SUBSTITUTION OF ONE KNOWN ELEMENT FOR ANOTHER THAT YIELDS
`PREDICTABLE RESULTS ........................................................................................ 56
`G. COMBINING SULLIVAN WITH TAM APPLIES A KNOWN TECHNIQUE TO A
`KNOWN DEVICE READY FOR IMPROVEMENT ........................................................ 56
`H. THE LIMITED NUMBER OF CAMERA CONFIGURATIONS SUGGEST THAT
`SULLIVAN IS OBVIOUS TO TRY IN COMBINATION WITH TAM ................................ 57
`I. SUMMARY OF MOTIVATIONS TO COMBINE ...................................................... 57
`X. SUMMARY OF OPINIONS .......................................................................... 58
`XI. OBVIOUSNESS ............................................................................................. 58
`A. GROUND 1 - CLAIMS 1-18 OF THE ‘628 PATENT ARE OBVIOUS OVER SULLIVAN
`‘475 IN VIEW OF TAM ‘868 AND TAM ‘711 UNDER 35 U.S.C. §103 ....................... 58
`XII. CONCLUSION ........................................................................................... 146
`
`Legend3D, Inc.
`Exhibit 1011-0003
`
`

`
`Declaration for Inter Partes Review of USPN 8,922,628
`
`INTRODUCTION
`
`
`
`I.  
`
`I, Carlos Vazquez, declare as follows:
`
`1.  
`
`I have been retained to provide expert testimony in support of
`
`Legend3D, Inc. (“Petitioner”) in its Petition for Inter Partes Review of U.S. Patent
`
`No. 8,922,628 (“Petition”).
`
`2.  
`
`I understand this proceeding involves U.S. Patent No. 8,922,628 (the
`
`‘628 Patent”), titled “System and Process for Transforming Two-Dimensional
`
`Images Into Three-Dimensional Images”. I further understand that the ‘628 Patent
`
`issued Dec. 30, 2014, based on U.S. utility patent application no. 12/874,190 filed
`
`Sep. 1, 2010, which claims priority to U.S. provisional patent application
`
`61/239,049, (the ‘049 provisional), filed Sep. 1, 2009. (“Earliest Potential Filing
`
`Date”). A copy of the ‘628 Patent is attached as Ex. 1001, its file history is
`
`attached as Ex. 1002 and the ‘049 provisional is attached as Ex. 1003. I have
`
`reviewed and am familiar with the ‘628 Patent as well as its file history along with
`
`the provisional to which the ‘628 Patent claims priority.
`
`3.  
`
`I understand that Petitioner challenges in its Petition the validity of
`
`claims 1-18 of the ‘628 Patent (“challenged claims”). The challenged claims
`
`include two independent claims, claims 1 and 12; and 16 dependent claims, claims
`
`2-11 and 13-18.
`
`3
`
`Legend3D, Inc.
`Exhibit 1011-0004
`
`

`
`Declaration for Inter Partes Review of USPN 8,922,628
`
`4.  
`
`I have reviewed and am familiar with, and am one of the two
`
`
`
`inventors of U.S. Patent No. 8,488,868, (“Tam ‘868”) (Ex. 1004) and U.S. Patent
`
`No. 8,213,711, (“Tam ‘711”) (Ex. 1006), the primary prior art references cited in
`
`the Petition, as well as Provisional Appl. Nos. 60/907,475, (“ ‘475 Provisional”),
`
`(Ex. 10005), filed 4/3/2007 as well as 61/129,869, (“ ‘869 Provisional”), (Ex.
`
`1007), filed 7/25/2008 to which Tam ‘868 and Tam ‘711 claim priority. I have
`
`also reviewed and am familiar with Zhang 2005 (Ex. 1015) that is incorporated by
`
`reference into Tam ‘868.
`
`5.  
`
`I have also reviewed and am familiar with U.S. Patent No. 7,573,475,
`
`(“Sullivan ‘475”). (Ex. 1008) and U.S. Patent No. 4,925,294, (“Geshwind ‘294”).
`
`(Ex. 1017)
`
`6.  
`
`I have also reviewed and am familiar with “Analysis of the Influence
`
`of Vertical Disparities Arising in Toed-in Stereoscopic Cameras” by Allison as
`
`published 3/7/2007, (“Allison 2007”), (Ex. 1012), 3D Revolution Webpages “How
`
`a 3-D movie is made – from Home to Hollywood 3-D” published 10/2007, (Ex.
`
`1013), and “2-D to 3-D Conversion” published 10/2007, (Ex. 1014) and
`
`“Stereoscopic Image Generation Based on Depth Images for 3D TV”, to Zhang
`
`and Tam, published 6/2005 (“Zhang 2005”) (Ex. 1015).
`
`7.  
`
`As set forth below, I am familiar with the technology at issue as of the
`
`Earliest Potential Filing Date.
`
`4
`
`Legend3D, Inc.
`Exhibit 1011-0005
`
`

`
`Declaration for Inter Partes Review of USPN 8,922,628
`
`8.  
`
`I have been asked to provide my technical analysis and opinions
`
`
`
`regarding the ‘628 Patent which form the basis for the grounds of invalidity set
`
`forth in the Petition. In forming my opinions, I relied on my own experience and
`
`knowledge as well as the ‘628 Patent and its file history, the Tam ‘868, Tam ‘711,
`
`‘869 Provisional and ‘475 Provisional, Sullivan ‘475 patents and the Allison 2007,
`
`Zhang 2005 publications.
`
`9.  
`
`I have also relied on additional data and facts of the type that experts
`
`in the field of invention would reasonably rely upon in forming an opinion on the
`
`subject. I have relied on such information in order to support and explain my
`
`opinion as to how a POSITA in the art at the time of the alleged invention of the
`
`‘628 Patent would have interpreted the disclosures of the prior art references set
`
`forth in the Petition. Where relevant, I cite such references in this declaration.
`
`II.   QUALIFICATIONS
`
`10.  
`
`I have studied, taught, and practiced electrical engineering and
`
`software engineering for over 20 years. I attended the Higher Polytechnic Institute
`
`José Antonio Echeverría (ISPJAE) in Havana, Cuba and earned a Bachelors (Eng)
`
`Degree in Electronics (Electronic Components and Devices) in 1992 where I
`
`graduated Summa cum laude, and a Masters (M.Sc.) of Science Degrees in
`
`Applied Informatics (Signal and Image Processing) in 1997. I earned my Ph.D. in
`
`Telecommunications from the Institut National de la Recherche Scientifique
`
`5
`
`Legend3D, Inc.
`Exhibit 1011-0006
`
`

`
`Declaration for Inter Partes Review of USPN 8,922,628
`
`
`
`(INRS), in Montréal, Canada in 2003, where I was also a post-doctoral fellow from
`
`2002-2004 conducting research on a project related to the application of Level Set
`
`Methods to image segmentation, motion estimation and video super resolution.
`
`11.  
`
`I am currently an Associate Professor of Software Engineering and
`
`Information Technology at the École de Technologie Supérieure (ETS) in
`
`Montréal, Canada. I have held this position since 2013, teaching courses on
`
`Parallel programming, multimedia systems, digital
`
`imaging and computer
`
`graphics; 3D Vision Systems, and operating systems, and conducting research in
`
`3D video systems, 3D object extraction and reconstruction; medical imaging
`
`applications; 3D cinema and TV applications.
`
`12.   Prior to my employment at ETS, I was a research Scientist with the
`
`Communications Research Center (CRC) in Ottawa, Canada in the Advanced
`
`Video Systems, where from 2005 through 2013 I worked in 3D-TV systems,
`
`namely developing algorithms and applications to help advance the 3D-TV
`
`industry, 2D-to-3D conversion, novel displays, S3D video processing; Multi-View
`
`Video Coding, namely, 4D wavelet-based multi-view video coding, 2D + Depth +
`
`Occlusion multi-view coding and representation, and view synthesis and
`
`occlusions removal; Depth-Image Based Rendering, namely, real time image based
`
`rendering techniques, parallel programming, GP-GPU, among other topics.
`
`6
`
`Legend3D, Inc.
`Exhibit 1011-0007
`
`

`
`Declaration for Inter Partes Review of USPN 8,922,628
`
`13.  
`
`I have also worked as a Research Associate in the Electrical and
`
`
`
`Computer Engineering Department of Concordia University, Montréal, Canada
`
`from 2004 to 2005, and as a Research Assistant at the University of Quebec,
`
`Montréal, Canada from 1998 to 2004 in the areas of software development and
`
`hardware installation and administration areas, namely, video equipment and
`
`stereoscopic video equipment.
`
`14.   Upon graduation with my Bachelors degree and while obtaining my
`
`Masters Degree, I was an Assistant Professor in the Telecommunications
`
`Department of the ISPJAE in Havana, Cuba where I taught courses on
`
`programming languages, embedded systems, signal acquisition and processing
`
`from 1992 through 1997.
`
`15.   My twenty-plus years of industry experience includes authoring
`
`numerous peer-reviewed journal and conference articles, including attaining as a
`
`co-recipient the Scott Helt best paper award from IEEE Broadcast Technology
`
`Society in 2011. My research interests are in the area of 3D-TV, 2D-to-3D video
`
`conversion; stereo and multi-view vision systems; 3D object extraction and
`
`reconstruction; image/video representation, sampling and interpolation; image and
`
`video coding; and motion/disparity/depth estimation and compensation.
`
`16.  
`
`I have authored over 30 papers at various conferences and symposia
`
`over the past ten-plus years, such as the IEEE International Conference on Image
`
`7
`
`Legend3D, Inc.
`Exhibit 1011-0008
`
`

`
`Declaration for Inter Partes Review of USPN 8,922,628
`
`
`
`Processing, the International Workshop on Computer Vision and Its Application to
`
`Image Media Processing, The SPIE Visual Communications and Image
`
`Processing, and SPIE Stereoscopic Displays and Applications. I also am co-
`
`inventor on three U.S. Patents disclosing Generation of a depth map from a
`
`monoscopic color image for rendering stereoscopic still and video images; Method
`
`and graphical interface for modifying depth maps; and Enhancing virtual presence
`
`using multiview and stereoscopic 3D displays with a grounded curvilinear screen.
`
`17.  
`
`I became a Senior Member of the IEEE in 2012, and obtained the
`
`2012 Journal Certificate of Merit from SMPTE Motion Imaging Journal. In 2011 I
`
`received the NAB Technology Innovation Award as the co-inventor of one of the
`
`technologies supporting the award: 2D-to-3D video conversion.
`
`18.  
`
`I have also authored book chapters on “2D to 3D video conversion-
`
`overview and perspectives” in the book Emerging Technologies for 3D Video:
`
`Creation, Coding, Transmission, and Rendering; and “DIBR-based conversion
`
`from monoscopic to stereoscopic and multiview video” in 3D-TV System with
`
`Depth-Image Based Rendering: Architectures, Techniques and Challenges.
`
`19.   My Curriculum Vitae is provided as Ex. 1010.
`
`20.  
`
`I am being compensated for my work in this matter at a rate of $200
`
`per hour, with reimbursement for necessary and reasonable expenses. This
`
`compensation is not in any way contingent upon the outcome of this Inter Partes
`
`8
`
`Legend3D, Inc.
`Exhibit 1011-0009
`
`

`
`Declaration for Inter Partes Review of USPN 8,922,628
`
`
`
`Review. I have no interest in the outcome of this proceeding or any related
`
`litigation.
`
`III.   UNDERSTANDING OF THE GOVERNING LAW
`
`21.  
`
`I am not a patent attorney, nor have I independently researched the
`
`law on patent validity. Attorneys for the Petitioner have explained certain legal
`
`principles to me that I have relied upon in forming my opinions set forth herein.
`
`Level Of Ordinary Skill In The Art
`A.  
`22.   The ‘628 Patent describes the field of the invention as follows: “The
`
`field of the invention is generally related to three-dimensional film post-production
`
`processes. In particular, the invention relates to a system and process for
`
`converting two-dimensional images into three-dimensional images.” ‘628 Patent at
`
`Col. 1, ll. 15-18. Based on this description as well as the ‘628 Patent’s description,
`
`it is my opinion that the relevant art for the ‘628 Patent is two-dimensional to
`
`three-dimensional image conversion including use of information contained in
`
`images to generate depth maps for rendering stereoscopic images.
`
`23.  
`
`I believe that a POSITA in the art (“POSITA”) in the field of the ‘628
`
`Patent is a person with an undergraduate degree in electrical engineering or
`
`computer science (or an equivalent subject) and would have been someone with a
`
`good working knowledge of computer programming, data structures, and image
`
`processing. The person would have gained this knowledge either through an
`
`9
`
`Legend3D, Inc.
`Exhibit 1011-0010
`
`

`
`Declaration for Inter Partes Review of USPN 8,922,628
`
`
`
`undergraduate education in computer science or comparable field, in combination
`
`with training or several years of practical working experience. This description is
`
`approximate, and a higher level of education or skill might make up for less
`
`experience, and vice-versa.
`
`24.  
`
`I believe that I would qualify as at least a POSITA in the art at that
`
`time, and that I have a sufficient level of knowledge, experience and education to
`
`provide an opinion in the field of the ‘628 Patent. Specifically, I work with many
`
`people including students, researchers that I supervise and colleagues in the
`
`industry who are persons of ordinary skill in the art, so I have an understanding of
`
`what a POSITA in the art is and would have had as of the filing date of the ‘628
`
`Patent and in light of the specification and file history of the ‘628 Patent. It is
`
`through this viewpoint that I make my conclusions in this matter.
`
`25.  
`
`I understand that a POSITA in the art is not a specific, real individual,
`
`but rather a hypothetical individual presumed to have knowledge of the relevant art
`
`at the time of the invention.
`
`26.   My opinion of the level of ordinary skill in the art is based on my
`
`extensive personal experience working in the field of image processing; my
`
`knowledge of the level of education and experience of colleagues and others
`
`actively working in the field as of and for several years prior to the Earliest
`
`Potential Filing Date; the types of problems encountered in the art at the time the
`
`10
`
`Legend3D, Inc.
`Exhibit 1011-0011
`
`

`
`Declaration for Inter Partes Review of USPN 8,922,628
`
`
`
`subject matter was developed, and prior art solutions to those problems including,
`
`in particular, problems arising from creating
`
`three-dimensional (3D) or
`
`stereoscopic images; and the level of sophistication of the technology at issue in
`
`this proceeding.
`
`B.  
`27.  
`
`Invalidity by Anticipation or Obviousness
`
`I understand that a claim is invalid if it is anticipated or obvious. I
`
`understand that anticipation requires that every element of a claim is disclosed
`
`expressly or inherently in a single prior art reference, as arranged in the claim.
`
`28.  
`
`I understand that a claim is invalid as obvious if it would have been
`
`obvious from the perspective of a POSITA in the relevant art, at the time the
`
`invention was made. In analyzing obviousness, I understand that it is important to
`
`understand the scope of the claims, the level of skill in the relevant art, the scope
`
`and content of the prior art, the differences between the prior art and the claims,
`
`and any secondary considerations.
`
`29.  
`
`I also understand that if a technique has been used to improve one
`
`device, and a POSITA in the art would recognize that it would improve similar
`
`devices in the same way, using the technique is obvious unless its actual
`
`application is beyond his or her skill. There may also be a specific teaching,
`
`suggestion or motivation to combine any first prior art reference with a second
`
`prior art reference. Such teaching, suggestion, or motivation to combine can be
`
`11
`
`Legend3D, Inc.
`Exhibit 1011-0012
`
`

`
`Declaration for Inter Partes Review of USPN 8,922,628
`
`
`
`explicit or implicit in the first or second prior art references as per KSR Int’l Co. v.
`
`Teleflex, Inc. It is my understanding that KSR confirms that any motivation that
`
`would have been known to a person of skill in the art, including common sense, or
`
`derived from the nature of the problem to be solved, is sufficient to explain why
`
`references would have been combined. I also understand the combination of
`
`familiar elements according to known methods is likely to be obvious when it does
`
`no more than yield predictable results and that if a person of ordinary skill can
`
`implement a predictable variation, that variation would have been considered
`
`obvious. In addition, I understand that the obviousness inquiry should not be done
`
`in hindsight, but must be done using the perspective of a person of ordinary skill in
`
`the relevant art as of the effective filing date of the patent claim, and that prior art
`
`may be considered non-obvious if one or more prior art references discourages or
`
`lead away from the line of inquiry disclosed in the reference(s).
`
`30.  
`
` I understand that the earliest filed patent application to which the
`
`‘628 Patent claims priority was filed September 1, 2009. I have therefore analyzed
`
`the prior art references discussed herein from the perspective of a POSITA in the
`
`art in the 2009 timeframe, understanding that as time passes, the knowledge of a
`
`POSITA in the art will increase.
`
`C.  
`
`Interpreting Claims Before the Patent Office
`
`12
`
`Legend3D, Inc.
`Exhibit 1011-0013
`
`

`
`Declaration for Inter Partes Review of USPN 8,922,628
`
`31.  
`
`I understand that in an Inter Partes Review proceeding, claims are
`
`
`
`given their broadest reasonable interpretation that is consistent with the patent
`
`specification. I understand that a patent’s “specification” includes all the figures,
`
`discussion, and claims within the patent document. I understand that the USPTO
`
`will look to the specification to see if there is a definition for a claim term, and if
`
`not, will apply the broadest reasonable interpretation from the perspective of a
`
`POSITA. I have reviewed the ‘628 file history in making my conclusions. I
`
`present a more detailed explanation of the interpretation of certain terms in the
`
`‘628 Patent in section IX below.
`
`IV.   OVERVIEW OF THE ‘628 PATENT
`
`A.  
`
`Field of the Purported Invention
`
`
`
`32.   According to its specification, the ‘628 Patent relates to a system and
`
`method for the conversion of two-dimensional (2D) images into left and right
`
`image pairs providing the perception of three-dimensional (3D) images as per the
`
`Abstract and Field of Invention at Col. 1, ll. 15-20. ‘628 Fig. 3 shows the 2D to
`
`3D conversion system.
`
`13
`
`Legend3D, Inc.
`Exhibit 1011-0014
`
`

`
`Declaration for Inter Partes Review of USPN 8,922,628
`
`
`
`
`
`33.   Converting 2D images to left and right image pairs for a 3D effect
`
`digitally has been known in the art for decades. For example, Geshwind ‘294,
`
`in 1986, taught a computer-based system to convert 2D images to left and right
`
`image pairs, as well as claiming and suggesting the other points of novelty of the
`
`14
`
`Legend3D, Inc.
`Exhibit 1011-0015
`
`

`
`Declaration for Inter Partes Review of USPN 8,922,628
`
`
`
`‘628 Patent as detailed below in the Technology Background section. Geshwind
`
`‘294 Figure 1 shows the overall 2D to 3D conversion process:
`
`
`
`
`
`34.   The ‘628 specification acknowledges that 2D to 3D conversion is not
`
`novel; however, it asserts that existing methods are costly:
`
`Various methods have been attempted to convert these 2D image
`streams
`into
`three-dimensional
`image streams, most providing
`
`15
`
`Legend3D, Inc.
`Exhibit 1011-0016
`
`

`
`Declaration for Inter Partes Review of USPN 8,922,628
`
`reasonable results only after expending significant effort and cost.
`[Col. 2, ll. 25-28]
`
`
`
`Purported Novelty from the Use of Image Characteristics for
`B.  
`Conversion
`35.   One purported point of novelty of the method taught in ‘628 is to
`
`convert 2D images to 3D images using the image’s own intrinsic qualities or
`
`“characteristics” (e.g., how bright a pixel is). The ‘628 specification asserts that
`
`typical industry practice is to manually outline (“rotoscope” or “mask”) objects in
`
`a 2D image and manually assign depth (apparent distance from the viewer) to the
`
`objects for a 3D effect. In contrast, ‘628 asserts that its method automatically
`
`selects objects and assigns depth to them based on the characteristics of the pixels
`
`in the image. ‘628 refers to object selection as “segmenting,” and it refers to
`
`assigning a depth effect as “remapping.” ‘628 teaches that characteristics may
`
`include hue (a color irrespective of brightness), luminance (how bright), saturation
`
`(how pure or gray a color is), and color (the combination of hue, luminance, and
`
`saturation) at Col. 5, ll. 64-67.
`
`36.  
`
` An example is provided of generating the depth of an actor’s face
`
`automatically based on the redness, contrast, gamma, or luminance of the
`
`individual pixels in the face:
`
`Consider next the common appearance of an actor's face in film
`making. The blood vessels of the face generally radiate out towards or
`
`16
`
`Legend3D, Inc.
`Exhibit 1011-0017
`
`

`
`Declaration for Inter Partes Review of USPN 8,922,628
`
`away from the actors nose, depending on their current performance.
`Using the Red channel, and some simple modifiers whether they be
`contrast, gamma, luminance of that channel, or operators such as
`addition and subtraction for other regions, the artist can obtain a
`gradient across the actors face, following the contours of the actual
`image which, when applied at Step 5, allow
`the weighted
`displacement to create a depth of considerable volume and detail, or a
`very realistic three-dimensional face. Because the artist does not need
`to build 3D geometry, the face does not need to be modeled, or
`tracked throughout the shot… [Col 33, ll. 21-33]
`
`
`
`37.   Thus, the gist of the ‘628 Patent is to use the actual values of an
`
`image’s characteristics, for example each pixel’s hue, luminance, saturation or
`
`color, to select or segment objects, and to assign depth to the pixels within an
`
`object. Geshwind ‘294 claims this in 1986 as detailed in Section VII below.
`
`Additional Purported Novelty from the Use of a Multidimensional
`C.  
`Vector
`38.   All 2D to 3D conversion processes generate a pair of viewpoints (left
`
`and right views) from an original 2D image. The illusion of depth results from the
`
`different positions of objects in the left view compared to the right view. This
`
`basic fact of 3D vision (referred to in the art as “parallax”) has been known for
`
`more than a century. See Geshwind ‘294 Figure 1 above for an example of the
`
`shifts in object positions between left and right views.
`
`17
`
`Legend3D, Inc.
`Exhibit 1011-0018
`
`

`
`Declaration for Inter Partes Review of USPN 8,922,628
`
`39.   The ‘628 Patent discloses shifting objects between left and right views
`
`
`
`to create the 3D effect. While shifting between left and right views is not novel
`
`(again, see Geshwind ‘294 Figure 1), the Patent Owner asserted in its Patent
`
`Owner Preliminary Response (Ex. 1019) to the previously filed petition IPR2015-
`
`01350 that the specific type of shifting described in ‘628 is novel. Specifically,
`
`‘628 describes shifting pixels with a “multidimensional vector”, which Patent
`
`Owner defines as a vector with both a horizontal and a vertical component. Patent
`
`Owner has asserted that this multidimensional vector is a point of novelty of ‘628,
`
`since (according to the Patent Owner) other 2D to 3D conversion processes only
`
`shift pixels horizontally.
`
`The Bond Patent envisions a multidimensional vector field for each
`pixel because otherwise displacement could by default only ever
`occur in the horizontal direction, precluding the potential for vertical
`displacement. [Ex. 1019, P. 16, last full paragraph]
`
`40.  
`
`In contradiction to this definition of “multidimensional vector” (which
`
`requires a vertical component as well as a horizontal component) asserted by the
`
`Patent Owner, the ‘628 specification describes several cases of horizontal-only
`
`shifting and does not describe a single embodiment showing how “determining a
`
`vector field” could be accomplished, i.e., when would the vector field have
`
`multidimensional vectors of 5 degrees off of horizontal or 10 degrees off of
`
`horizontal, etc. For example:
`
`18
`
`Legend3D, Inc.
`Exhibit 1011-0019
`
`

`
`Declaration for Inter Partes Review of USPN 8,922,628
`
`Third, the elements of the scene are then shifted or moved horizontally
`and placed in the complimentary image of the stereo pair. [Ex. 1001,
`Col. 4, ll. 60-62]
`
`This depth selection mask is then used with a horizontal displacement to
`the right to produce FIG. 9B, which would be used as a left eye (LE)
`image of the stereo pair. [Ex. 1001, Col. 29, ll. 27-30]
`
`FIG. 9D depicts the degree of horizontal displacement of the pixels of
`the image. [Ex. 1001, Col. 29, ll. 33-34]
`
`When the image is recombined in a stereo pair, this divergence
`corresponds directly with horizontal parallax and occlusion revelation,
`two key factors in stereoscopic perception. [Ex. 1001, Col. 29, ll. 45-48]
`
`
`
`41.   Although the ‘628 Patent teaches horizontal shifting, it claims a
`
`vector field with multidimensional vectors since horizontal shifting was well-
`
`known in the art long before the filing date of the ‘628 Patent (see Geshwind ‘294
`
`Figure 1 above and Section VII below). As such, the ‘628 Patent claims shifting
`
`objects between left and right images using a multidimensional vector that must
`
`have both horizontal and vertical components, even if the vertical y component is
`
`set to zero as asserted by the Patent Owner in the Preliminary Response in
`
`IPR2015-01350 (Ex. 1019, P. 16). However, the ‘628 Patent does not claim that at
`
`least one vector in the vector field has a non-zero vertical component. Thus, in a
`
`parallel camera configuration, since all vertical y components are zero, the ‘628
`
`19
`
`Legend3D, Inc.
`Exhibit 1011-0020
`
`

`
`Declaration for Inter Partes Review of USPN 8,922,628
`
`
`
`multidimensional vector field contains only vectors with a horizontal direction
`
`without any non-zero vertical y component, regardless of how many dimensions
`
`the vectors have, i.e., V(x) = V(x,0,0,…0), e.g., ß = ß. Thus, in the parallel
`
`camera configuration, the vector field of the ‘628 Patent has the same magnitude
`
`and direction for all vectors in the Geshwind ‘294 vector field and therefore, the
`
`Geshwind ‘294 species of vector anticipates the ‘628 Patent multidimensional
`
`vector since the ranges overlap, i.e., ß = ß, i.e., the vectors themselves have the
`
`same magnitude and direction. Even if the ‘628 Patent claimed at least one of the
`
`vectors in the vector field has a non-zero vertical y component to assert novelty
`
`over Geshwind ‘294, (as well as Tam ‘868), Geshwind ‘294 suggests that the
`
`vertical y components may be non-zero and thus would render the ‘628 Patent
`
`obvious. Specifically, Geshwind ‘294 in 1986 suggests that 2D shifting could be
`
`used instead of simple left and right shifting. To be clear, use of horizontal and
`
`vertical offset differences or “disparities” together to shift pixels is a suggestion to
`
`use multidimensional vectors, which constitutes a vector field as defined in the
`
`‘628 Patent. Geshwind ‘294 thus shows that this purported novelty in the ‘628
`
`Patent was suggested in 1986, well before the ‘628 Patent priority date.
`
`It should be noted that horizontal parallax offsets are, by far,
`more obvious, due to the fact that our eyes are separated in the
`horizontal direction. However, vertical offset differences between
`left- and right-eye views may also be appropriate. [Col 5, ll. 48-52]
`
`20
`
`Legend3D, Inc.
`Exhibit 1011-0021
`
`

`
`Declaration for Inter Partes Review of USPN 8,922,628
`
`42.  
`
`In summary, in view of Geshwind ‘294, the purported points of
`
`
`
`novelty of the ‘628 Patent are all old in the art and not novel alone or in
`
`combination. See Section VII below for a detailed description of Geshwind ‘294
`
`and other prior art.
`
`V.  
`
` ‘628 PATENT PROSECUTION HISTORY
`
`43.   During prosecution of the ‘628 Patent, the Examiner did not rely on or
`
`consider Sullivan ‘475, Tam ‘868, or Tam ‘711, the prior art that are the subject of
`
`this inter partes review. The limited number of prior art references that were cited
`
`and relied upon during prosecution do not contain the same or equivalent
`
`disclosures of Sullivan ‘475, Tam ‘868, Tam ‘711, (or Geshwind ‘294, not relied
`
`upon, but which claims and suggests the purported points of novelty of the ‘628
`
`Patent as claimed in 1986, as detailed below).
`
`44.   The primary references cited by the examiner during prosecution of
`
`the ‘628 Patent were Chen (US 20030091225), O'Neill (US 5691843), and Kaye
`
`(US 6686926). None of these references teach use of image features to determine
`
`depth, such as Tam ‘868 and Tam ‘711 disclose, and none of these references teach
`
`use of a vector field to shift horizontally and vertically, using a vector field as
`
`Sullivan ‘475 discloses.
`
`
`VI.   CLAIM CONSTRUCTION
`
`21
`
`Legend3D, Inc.
`Exhibit 1011-0022
`
`

`
`Declaration for Inter Partes Review of USPN 8,922,628
`
`45.   For ease of reference, Claim 1 of the ‘628 Patent is listed below with
`
`
`
`identifiers in the leftmost column used to refer to the specific limitations herein,
`
`wherein bold claim limitations deserve claim constructions:
`
`No.
`1a
`
`1b
`
`1c
`
`1d
`
`1e
`
`1f
`
`1g
`
`1h
`
`1i
`
`‘628 Limitation
`A process for creating a three dimensional media projection from a two-
`dimensional image, comprising:
`copying a two-dimensional image to create a working image of the