`
`1
`
` 1 UNITED STATES PATENT AND TRADEMARK OFFICE
`
` 2 BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
` 3
`
` 4 Case IPR2017-01131
` US Patent No. 8,464,304
`
` 5
`
` 6 Case IPR2017-01133
` US Patent No. 8,601,506
`
` 7
`
` 8 *********************************
`
` 9 TWITTER, INC.,
`
` 10 Petitioner,
`
` 11 vs.
`
` 12 STI-ACQ, LLC,
`
` 13 Patent Owner.
`
` 14 *********************************
`
` 15
`
` 16
`
` 17 DEPOSITION OF HENRY H. HOUH, Ph.D., a witness
`
` 18 called on behalf of the Patent Owner, pursuant to the
`
` 19 Rules of Civil Procedure, before Karen D. Pomeroy,
`
` 20 Registered Diplomate Reporter and Notary Public in
`
` 21 and for the Commonwealth of Massachusetts, at the
`
` 22 offices of Dunn Reporting Services, 185 Devonshire
`
` 23 Street, Boston, Massachusetts, on Wednesday,
`
` 24 June 20th, 2018, commencing at 9:08 a.m.
`
` 25
`
`Lexitas
`
`Twitter, Inc. v. VidStream LLC
`IPR2017-01131
`VidStream LLC | Ex. 2008
`Page 1 of 190
`
`
`
`Henry H. Houh, PhD
`
`2
`
` 1 APPEARANCES:
`
` 2 TODD M. SIEGEL, ESQUIRE
`
` 3 Klarquist Sparkman, LLP
`
` 4 One World Trade Center
`
` 5 121 SW Salmon Street, Suite 1600
`
` 6 Portland, Oregon 97204
`
` 7 503-595-5300
`
` 8 todd.siegel@klarquist.com
`
` 9 For the Petitioner
`
` 10
`
` 11
`
` 12 EAGLE ROBINSON, ESQUIRE
`
` 13 Norton Rose Fulbright US LLP
`
` 14 98 San Jacinto Boulevard, Suite 1100
`
` 15 Austin, Texas 78701-4255
`
` 16 512-474-5201
`
` 17 eagle.robinson@nortonrosefulbright.com
`
` 18 For the Patent Owner
`
` 19
`
` 20
`
` 21
`
` 22
`
` 23
`
` 24
`
` 25
`
`Lexitas
`
`Page 2 of 190
`
`
`
`Henry H. Houh, PhD
`
`3
`
` 1 INDEX
`
` 2 DEPOSITION OF HENRY H. HOUH, Ph.D. PAGE
`
` 3 Examination by Attorney Robinson 4
`
` 4
`
` 5
`
` 6
`
` 7
`
` 8
`
` 9
`
` 10
`
` 11
`
` 12
`
` 13 EXHIBITS
`
` 14 Number Page
`
` 15 (None.)
`
` 16
`
` 17
`
` 18
`
` 19
`
` 20
`
` 21
`
` 22
`
` 23
`
` 24
`
` 25
`
`Lexitas
`
`Page 3 of 190
`
`
`
`Henry H. Houh, PhD
`
`4
`
` 1 HENRY H. HOUH, Ph.D.,
`
` 2 having been duly sworn by the reporter, was
`
` 3 deposed and testified as follows:
`
` 4 EXAMINATION
`
` 5 BY MR. ROBINSON:
`
` 6 Q. Dr. Houh, thank you for being here today.
`
` 7 How many times have you been deposed
`
` 8 previously?
`
` 9 A. I've probably been deposed maybe 40 -- around 40
`
` 10 times.
`
` 11 Q. Okay.
`
` 12 A. I don't remember the exact number.
`
` 13 Q. All right. I won't spend a lot of time on
`
` 14 instructions or ground rules then.
`
` 15 The only thing I would ask is if I ask a
`
` 16 question and you don't understand it, please just
`
` 17 ask me to clarify.
`
` 18 A. Okay.
`
` 19 Q. I'd like to start with some basics.
`
` 20 In the 2006 time frame, were you working with
`
` 21 image and video capture with mobile phone
`
` 22 cameras?
`
` 23 A. In my -- as part of the declaration submitted
`
` 24 there was a copy of my CV.
`
` 25 Could I take a look at that, please.
`
`Lexitas
`
`Page 4 of 190
`
`
`
`Henry H. Houh, PhD
`
`5
`
` 1 Q. Sure.
`
` 2 (Recess was taken from 9:09 a.m. until 9:10 a.m.)
`
` 3 BY MR. ROBINSON:
`
` 4 Q. So, Dr. Houh, I'll hand you what's been marked
`
` 5 Exhibit 1004.
`
` 6 I believe that's the CV you mentioned.
`
` 7 A. So around 2006 I was working partly on a project
`
` 8 called PodZinger, and PodZinger is -- was a
`
` 9 website where we did search in video; so video
`
` 10 and audio files, and so we actually had a website
`
` 11 and set up a website to provide hosting and
`
` 12 streaming for media and audio and a search site
`
` 13 for searching for spoken words within video and
`
` 14 audio files that we index on the Internet, and we
`
` 15 also eventually provided hosting services for
`
` 16 video and audio as part of PodZinger as well.
`
` 17 So in 2006 that was what I was doing. I
`
` 18 think those sites were accessible from mobile
`
` 19 phones, but I was working on the back-end sites
`
` 20 primarily as part of the PodZinger project.
`
` 21 Q. Okay. Do you know how cameras on cell phones
`
` 22 operated to capture video in 2006?
`
` 23 A. I mean, I am a -- I've been traditionally a
`
` 24 camera person starting with film in the 1980s,
`
` 25 early 1980s; and I think by 2006 I also had a
`
`Lexitas
`
`Page 5 of 190
`
`
`
`Henry H. Houh, PhD
`
`6
`
` 1 digital camera.
`
` 2 I don't recall what type of phone I had at
`
` 3 the time. It's just too long ago now. But I was
`
` 4 aware -- I've become aware of cameras; you know,
`
` 5 digital cameras. I'd been aware of them at the
`
` 6 time and that they -- there's sensors and lenses
`
` 7 and focusing mechanism that I understood -- I
`
` 8 don't remember when I understood the focusing
`
` 9 mechanism to be based on, you know -- so I've
`
` 10 read about the technology for camera phones, but
`
` 11 I don't know when.
`
` 12 Q. Okay. Do you know how camera phones in 2006
`
` 13 determined the frame rate at which video would be
`
` 14 captured?
`
` 15 MR. SIEGEL: Objection.
`
` 16 Form.
`
` 17 A. I mean, at that time I believe I was aware of --
`
` 18 of -- I mean, starting in the early '90s and -- I
`
` 19 was working on video -- digital video capture
`
` 20 around '93 or '94 as part of my graduate research
`
` 21 work, and so I was aware back then of capture
`
` 22 techniques for video and also compression
`
` 23 techniques for video, although our project back
`
` 24 then didn't rely on compression techniques or
`
` 25 video.
`
`Lexitas
`
`Page 6 of 190
`
`
`
`Henry H. Houh, PhD
`
`7
`
` 1 I took a course in -- a graduate course at
`
` 2 the MIT media lab on video compression
`
` 3 technologies. I can't remember the title of that
`
` 4 course, but we learned a lot more -- if I didn't
`
` 5 already know somewhat about video compression, I
`
` 6 understood a lot more about video compression as
`
` 7 part of that course.
`
` 8 And part of the -- one of our projects in
`
` 9 that course was to write a -- use some of the
`
` 10 standards to compress a single frame in a
`
` 11 JPEG-like fashion which is similar to the way
`
` 12 iFrames are compressed in MPEG-type video
`
` 13 compression algorithms.
`
` 14 And so -- I had also taken other classes on
`
` 15 video. I took another class at the media lab
`
` 16 called -- I think it was called Elastic Video
`
` 17 Time or Elastic -- something about video or -- I
`
` 18 can't recall exactly the title now, but it was
`
` 19 about using and editing a video together for a
`
` 20 media project.
`
` 21 So I was aware of video compression
`
` 22 techniques starting in -- I believe that was in
`
` 23 the mid '90s, and I did -- I was aware of video
`
` 24 compression standards that came out.
`
` 25 I don't recall exactly the dates now without
`
`Lexitas
`
`Page 7 of 190
`
`
`
`Henry H. Houh, PhD
`
`8
`
` 1 being able to do a little bit of background on
`
` 2 it, and I understood at some point there were
`
` 3 chips involved for compressing video.
`
` 4 I don't know offhand when video compression
`
` 5 chips became available.
`
` 6 BY MR. ROBINSON:
`
` 7 Q. In 2006 how did camera phones set the frame rate
`
` 8 at which video would be captured?
`
` 9 MR. SIEGEL: Objection.
`
` 10 Form.
`
` 11 A. I mean, I think there'd be -- if you have a -- if
`
` 12 you have the code or the -- or the -- code and/or
`
` 13 circuit diagram for some phones, I'd have to
`
` 14 research that; but it's hard to make a general
`
` 15 statement, especially since I don't know about
`
` 16 every single phone available in 2006.
`
` 17 Q. Can you give me some examples of the way in which
`
` 18 camera phones in 2006 would have determined the
`
` 19 frame rate at which video was captured?
`
` 20 A. I mean, I can -- I can review some high level,
`
` 21 you know, compression algorithms with you in
`
` 22 terms -- so I'm not aware of every specific
`
` 23 algorithm that every phone utilized back then.
`
` 24 I mean, what's clear is that, you know, these
`
` 25 phones, the CCD devices which I understand -- I'm
`
`Lexitas
`
`Page 8 of 190
`
`
`
`Henry H. Houh, PhD
`
`9
`
` 1 not aware of any other technology to do --
`
` 2 capture images, but the CCD devices in cameras
`
` 3 could be read out and then algorithms could be
`
` 4 run on images; and depending on how fast either
`
` 5 the software or hardware algorithms were, that
`
` 6 would determine maximum possible frame rate
`
` 7 for -- for the video capture.
`
` 8 Q. Okay. I want to make sure I understand; so let's
`
` 9 start with the CCD.
`
` 10 What does CCD stand for in this context?
`
` 11 A. I would be intending it to stand for a
`
` 12 charge-coupled device.
`
` 13 Q. And at a high level, what -- the charge-coupled
`
` 14 device is a sensor that captures individual
`
` 15 images?
`
` 16 A. At a high level, a charge-coupled device would be
`
` 17 made up of a number of sensors, usually arranged
`
` 18 in an array, on which light is focused and
`
` 19 there's some relationship between the level of
`
` 20 charge -- and I understand that there's some
`
` 21 relationship between the level of charge and the
`
` 22 amount of light, and that the charge-coupled
`
` 23 device can be cleared to represent a new frame
`
` 24 and read on a periodic basis.
`
` 25 Q. So to capture a video, the CCD captures a
`
`Lexitas
`
`Page 9 of 190
`
`
`
`Henry H. Houh, PhD
`
`10
`
` 1 sequence of images?
`
` 2 A. It would be kin to -- well, I mean, there's a lot
`
` 3 of intermediate steps between making up the image
`
` 4 and reading from the charge-coupled device, but
`
` 5 essentially it captures an image that can be
`
` 6 represented in digital form -- that is
`
` 7 represented in digital form and stored.
`
` 8 Q. Okay. And the frame rate generally refers to the
`
` 9 number of images per second that is captured to
`
` 10 generate a video?
`
` 11 A. So the frame rates in a video typically refers to
`
` 12 how many frames are able -- are displayed per
`
` 13 second, and that frame rate -- if the hardware
`
` 14 and software are capable of capturing the video
`
` 15 at a certain rate, the rates often don't need to
`
` 16 go beyond, say, 30 frames a second.
`
` 17 Q. So I believe a moment ago you mentioned two
`
` 18 things within a camera phone that could affect
`
` 19 the frame rate; the hardware and the software.
`
` 20 I'd like to start with the hardware.
`
` 21 How does that impact the frame rate?
`
` 22 A. I mean, there are various different parts of the
`
` 23 hardware; so there's -- there -- in a complex
`
` 24 system, oftentimes the -- the slowest device for
`
` 25 a critical step is the most limiting; so that if
`
`Lexitas
`
`Page 10 of 190
`
`
`
`Henry H. Houh, PhD
`
`11
`
` 1 other parts of the system are faster or fast
`
` 2 enough or can be done faster, it doesn't matter
`
` 3 as long as, you know, that one part may limit the
`
` 4 speed.
`
` 5 So what I said -- earlier I was discussing
`
` 6 either the software compression algorithms for
`
` 7 MPEG or the hardware chips for MPEG compression.
`
` 8 One of the projects that we did in our lab
`
` 9 group at -- in the early '90s that my office mate
`
` 10 did, it wasn't -- I didn't design it, but he
`
` 11 designed a digital video capture card; and so I
`
` 12 was familiar with those -- many of those steps by
`
` 13 the early 1990s and that there -- the compression
`
` 14 can be done in software; so there are a number of
`
` 15 steps, for example, for MPEG compression.
`
` 16 Each frame is not compressed individually
`
` 17 based on the data in a single frame.
`
` 18 A lot of the algorithms involved what's known
`
` 19 as motion compensation so that a frame may be
`
` 20 derived from a previous frame.
`
` 21 Because if you think about a video in, say, a
`
` 22 30th of a second, a lot of things don't move
`
` 23 around that much. You know, talking heads.
`
` 24 The exception might be sports games or, you
`
` 25 know, high speed panning a camera; but a lot of
`
`Lexitas
`
`Page 11 of 190
`
`
`
`Henry H. Houh, PhD
`
`12
`
` 1 those you can also adjust if it's a very fast
`
` 2 pan; if it's a slower pan, a lot of the
`
` 3 algorithms are very good at compressing that kind
`
` 4 of video.
`
` 5 And the way it does that is by a method known
`
` 6 as -- it uses a previous frame or a future frame
`
` 7 and it -- there are blocks that are searched for
`
` 8 as part of these algorithms that see if it can
`
` 9 find a similar block in a different frame and,
`
` 10 instead of encoding the whole new block, encoding
`
` 11 the whereabouts of the previous block as well as
`
` 12 a motion vector that identifies how the previous
`
` 13 block moved from the previous frame to the
`
` 14 current frame.
`
` 15 So oftentimes that requires -- that and
`
` 16 similar methods of encoding between frames
`
` 17 results in data that's less voluminous than
`
` 18 encoding that block because you've already
`
` 19 encoded that information in that block, often
`
` 20 with some level of detail; so you can then encode
`
` 21 the motion of that block and any slight
`
` 22 differences to that block.
`
` 23 And so the algorithms for video compression
`
` 24 are often very asymmetrical in that they require
`
` 25 a lot more processing to compress the video
`
`Lexitas
`
`Page 12 of 190
`
`
`
`Henry H. Houh, PhD
`
`13
`
` 1 because you have to do searching for these
`
` 2 blocks, matching blocks, and then they do the --
`
` 3 decompress the video because -- and that's the
`
` 4 way the algorithms evolved and were designed at
`
` 5 the time.
`
` 6 I think many similar methods are used today,
`
` 7 and certainly in 2006 those methods were known
`
` 8 and used to produce much better compression than
`
` 9 simply just compressing each frame independently;
`
` 10 and in many of these compression algorithms
`
` 11 the -- the frames that are compressed
`
` 12 independently are often known as iFrames, and
`
` 13 they -- every iFrame can be decoded with
`
` 14 information only from a particular iFrame; and
`
` 15 then there are these other types of frames called
`
` 16 B- and P-frames which are dependent on other
`
` 17 types of frames, and there's often a sequence of
`
` 18 frames that repeat starting with an iFrame and
`
` 19 that all the other frames in the sequence are B-
`
` 20 or P-frames which may depend on other frames
`
` 21 within that sequence, and so in that way -- and
`
` 22 those can all be -- the sequence can be changed
`
` 23 based on the -- on the -- on parameters that --
`
` 24 in terms of if you want to have a lot of frames
`
` 25 in between iFrames or fewer frames, and there are
`
`Lexitas
`
`Page 13 of 190
`
`
`
`Henry H. Houh, PhD
`
`14
`
` 1 pros and cons for those selections.
`
` 2 And so there are -- there could be software
`
` 3 algorithms involved in performing many of these
`
` 4 steps that I've outlined.
`
` 5 And there are other steps I probably have not
`
` 6 discussed, but at a high level, they're -- that's
`
` 7 one of the ways to get good video compression.
`
` 8 And I understand at some point, I don't know
`
` 9 exactly when or -- you know, off the top of my
`
` 10 head, many of those algorithms were put into
`
` 11 hardware chips that were available so that, in
`
` 12 place of a software running on a general purpose
`
` 13 CPU, one could design in a hardware chip into the
`
` 14 system to perform capture; and oftentimes there
`
` 15 are ways to configure the hardware to adjust for
`
` 16 the -- potentially frame rate or the exact
`
` 17 sequence of frames in a group of pictures as it's
`
` 18 known or the resolution or whatnot.
`
` 19 So -- so I think one has to look at the
`
` 20 architecture and -- you know, of each specific
`
` 21 type of compression circuit and associated
`
` 22 software to be able to look at where many of
`
` 23 these parameters would be derived from generally
`
` 24 speaking.
`
` 25 Q. So it could be that the -- the software's baked
`
`Lexitas
`
`Page 14 of 190
`
`
`
`Henry H. Houh, PhD
`
`15
`
` 1 into the hardware in the form of a chip or the
`
` 2 firmware on a chip, or it could also be a
`
` 3 separate software application?
`
` 4 Is that generally correct?
`
` 5 A. So these compression algorithms could be baked
`
` 6 into a chip in the firmware and oftentimes
`
` 7 that -- there are configuration parameters that
`
` 8 hardware chips can be configured with to adjust
`
` 9 certain parameters and -- or it could all be done
`
` 10 in -- on a general purpose CPU.
`
` 11 It's oftentimes less efficient, but if a CPU
`
` 12 is fast enough, then it can -- it can just be
`
` 13 done in software without much impact to the CPU.
`
` 14 So you have to look at, you know, the
`
` 15 software of a system and to understand where it's
`
` 16 done or the -- or the hardware architecture
`
` 17 software to see, you know, where or whether it's
`
` 18 all software or there's some chip involved.
`
` 19 Q. Okay. You described compression algorithms just
`
` 20 now.
`
` 21 Is that an example of encoding video content?
`
` 22 A. That's what I would say is taking something
`
` 23 that's raw and encoding or compressing it to go
`
` 24 from kind of the original raw video into a
`
` 25 compressed video form, and I think encoding is an
`
`Lexitas
`
`Page 15 of 190
`
`
`
`Henry H. Houh, PhD
`
`16
`
` 1 acceptable word to use there. I think more
`
` 2 specific would be, you know, compressing or -- it
`
` 3 is compressing and it is encoding in a sense.
`
` 4 Q. Is there -- just to make sure I understand.
`
` 5 Is there a difference between encoding and
`
` 6 compressing in that context?
`
` 7 A. I think encoding can mean more things than
`
` 8 compressing.
`
` 9 Like you can encode data without compressing.
`
` 10 You can expand it, but you're encoding it for
`
` 11 transmission.
`
` 12 Oftentimes there are encoders that take 4
`
` 13 bits and convert them to 5 bits because they're
`
` 14 concerning error correction information, and that
`
` 15 would also be a form of encoding.
`
` 16 Q. Okay. You mentioned raw video data just now.
`
` 17 What is that?
`
` 18 A. What I mean by "raw" is, you know, if one were to
`
` 19 display the video in terms of pixels to a display
`
` 20 specifying the, you know, color and -- you know,
`
` 21 all the required information; it's either color
`
` 22 and luminance or luminance and chrominance, but
`
` 23 there's combinations of ways that video can be
`
` 24 displayed, but that -- that output I would --
`
` 25 that drives the display, I would call that -- I
`
`Lexitas
`
`Page 16 of 190
`
`
`
`Henry H. Houh, PhD
`
`17
`
` 1 would think of that as kind of the raw video,
`
` 2 but it's -- in decoded form, but the raw video
`
` 3 captured from the CCD encodes similar for the
`
` 4 same information.
`
` 5 Some information's lost in the compression
`
` 6 and decompression process, but -- so there's --
`
` 7 the original capture video I would consider raw
`
` 8 if you had, you know, the -- the luminance and
`
` 9 chrominance values for every pixel; or the
`
` 10 information that, you know, an MPEG compressor
`
` 11 would take to compress the video, I would call
`
` 12 that the raw video.
`
` 13 Q. Is it accurate to say that in raw video, digital
`
` 14 video of course, each frame stands alone as
`
` 15 enough information to be displayed by itself?
`
` 16 A. If you're talking about an uncompressed raw video
`
` 17 stream, there are -- there are -- there are
`
` 18 standards that encode that as well, but that
`
` 19 would be -- I don't know if people -- the cameras
`
` 20 have a raw mode, right; so like a digital camera
`
` 21 has a raw mode where you're capturing a huge file
`
` 22 that really represents the original data right
`
` 23 off the CCD oftentimes, and if you were to have
`
` 24 that for every single frame of video at whatever
`
` 25 frame rate your video is I would say that's your
`
`Lexitas
`
`Page 17 of 190
`
`
`
`Henry H. Houh, PhD
`
`18
`
` 1 raw input stream.
`
` 2 Q. Okay. And you mentioned the encoding or
`
` 3 compression algorithm impacting frame rate.
`
` 4 What -- in -- you know, for digital
`
` 5 cameras -- digital camera phones as of 2006 can
`
` 6 you give me some examples of the types of things
`
` 7 that would have determined the frame rate at
`
` 8 which the raw video was captured at the CCD.
`
` 9 MR. SIEGEL: Objection.
`
` 10 Form.
`
` 11 A. Well, I mean -- I mean, generally speaking, when
`
` 12 you have a chip, chips are often rated at a
`
` 13 certain speed for certain operations; and so
`
` 14 if -- I think you'd have to look at the
`
` 15 specifications for the CCD chip and review it to
`
` 16 understand how fast it -- you can actually read a
`
` 17 whole frame off the CCD chip or how often -- how
`
` 18 fast that CCD chip can cycle through various
`
` 19 frames.
`
` 20 I think if you knew what the CCD chip was and
`
` 21 you had the specifications for the chip, you
`
` 22 could review those.
`
` 23 I think there are lots of different types of
`
` 24 CCDs that have different requirements, and so I
`
` 25 think you'd have to look and know what the chip
`
`Lexitas
`
`Page 18 of 190
`
`
`
`Henry H. Houh, PhD
`
`19
`
` 1 was and look at the specifications for that chip
`
` 2 to understand the maximum rate at which it
`
` 3 could -- could perform the operation that someone
`
` 4 was interested in.
`
` 5 BY MR. ROBINSON:
`
` 6 Q. Is that an example of what you mentioned earlier
`
` 7 as a hardware limitation?
`
` 8 Sort of setting the maximum available frame
`
` 9 rate?
`
` 10 A. So if a CCD chip can only, you know, read one --
`
` 11 you know, produce one frame a second, that's what
`
` 12 it was designed to do, you know, for some
`
` 13 application, then that would impose a limit on
`
` 14 how fast you could read from that particular CCD
`
` 15 chip a frame of video.
`
` 16 Some CCD chips would be -- may be available
`
` 17 in 2006 to do video at 30 frames a second or
`
` 18 faster. I think you have to review the
`
` 19 specifications for the chips.
`
` 20 Q. So just as an example, if you have a camera phone
`
` 21 in 2006 that the hardware's capable of capturing
`
` 22 30 frames per second, what would have determined
`
` 23 the rate at which video was actually captured?
`
` 24 A. So, again, I think you have to look at the
`
` 25 architecture and APIs and the software because
`
`Lexitas
`
`Page 19 of 190
`
`
`
`Henry H. Houh, PhD
`
`20
`
` 1 sometimes, even though hardware can be capable of
`
` 2 capturing say, for example, 30 frames per second,
`
` 3 it's quite possible that it could be configured
`
` 4 to configure at various different rates --
`
` 5 capture rates, including just a single frame.
`
` 6 And so I think one has to review the CCDs and
`
` 7 the chip specifications to understand how to
`
` 8 configure them and look at the software to see
`
` 9 whether the software or the firmware actually
`
` 10 interacts with that part of the chip to be able
`
` 11 to control the frame rate, and I think you'd have
`
` 12 to understand all that for any particular device
`
` 13 based on the version of software, the version of
`
` 14 hardware, whatnot.
`
` 15 Q. So in 2006 it would have depended on the
`
` 16 particular configuration of any given model of
`
` 17 camera phone?
`
` 18 Is that accurate?
`
` 19 A. I mean, to do a thorough investigation, I think
`
` 20 one should look at all the different classes of
`
` 21 circuitry and software loads for that to
`
` 22 understand exactly the frame rate at which the --
`
` 23 the device can run and at which it does run in
`
` 24 order to determine how that device operates with
`
` 25 respect to the frame rate of the video being
`
`Lexitas
`
`Page 20 of 190
`
`
`
`Henry H. Houh, PhD
`
`21
`
` 1 captured.
`
` 2 Q. And I -- I think what I'm hearing you say is that
`
` 3 you would have to investigate the particular
`
` 4 configuration because there were different
`
` 5 configurations available from different
`
` 6 manufacturers.
`
` 7 Is that accurate?
`
` 8 A. I mean, every different -- different circuit
`
` 9 board architectures and circuit boards may
`
` 10 have -- they may have different CCD chips for the
`
` 11 design of the camera, right.
`
` 12 They may have different firmware used to --
`
` 13 drivers for the CCD chip which may be available
`
` 14 to use or set the -- to control the capturing of
`
` 15 the video based on higher-level software, and
`
` 16 then the software may or may not interact with
`
` 17 those drivers or may use default values.
`
` 18 There's a lot to investigate to determine --
`
` 19 if I were doing a thorough investigation of a
`
` 20 particular camera and you were asking me a
`
` 21 question of how does that CCD know -- how is that
`
` 22 CCD controlled or, you know, any other hardware
`
` 23 component as part of the video capture work flow
`
` 24 and you were asking me these specific questions,
`
` 25 this is one way that I could think of to try to
`
`Lexitas
`
`Page 21 of 190
`
`
`
`Henry H. Houh, PhD
`
`22
`
` 1 figure out how to get to those answers is to look
`
` 2 at the architecture, look at the firmware, look
`
` 3 at the driver.
`
` 4 And when I mean "look at," I mean, you know,
`
` 5 look at the diagrams and the source code of the
`
` 6 software and then with access to that I think --
`
` 7 I think that's one way to provide the answer to
`
` 8 some of the questions for any particular camera
`
` 9 device.
`
` 10 Q. So without looking at that specific information,
`
` 11 as we sit here today can you tell me in 2006
`
` 12 whether all camera phones governed the frame rate
`
` 13 at which they captured video in the same way?
`
` 14 A. I mean, to be able to answer the question about
`
` 15 all camera phones, I think one -- whether they
`
` 16 govern it in the same way or not, again, I think
`
` 17 you have to look at individually each of the
`
` 18 different classes of camera phones, potentially
`
` 19 each different mode of software, and do an
`
` 20 investigation as to how each of those govern --
`
` 21 what was the word that you used again in that
`
` 22 question?
`
` 23 Q. The frame rate at which video is captured.
`
` 24 A. Yeah. And look at how each of those govern the
`
` 25 frame rate at which video's captured, and only --
`
`Lexitas
`
`Page 22 of 190
`
`
`
`Henry H. Houh, PhD
`
`23
`
` 1 if you're asking about all of the phones
`
` 2 available in 2006, I think to answer that
`
` 3 question you have to look at all the classes of
`
` 4 phones available and look at the circuit
`
` 5 backgrounds, the firmware, the drivers, the
`
` 6 software for each; and then -- and then if one of
`
` 7 them doesn't govern it that way, then you could
`
` 8 say, no, it doesn't do it for all.
`
` 9 But then once you have found one by one that
`
` 10 it does it the way you're asking, then after
`
` 11 you've done it for all those classes and all
`
` 12 those software loads, then one could say that all
`
` 13 of them work that way; but I think without
`
` 14 thorough investigation, I don't think anyone'd be
`
` 15 able to answer that question.
`
` 16 Q. Can you tell me how any particular camera phone
`
` 17 in 2006 governed the frame rate at which it
`
` 18 captured video?
`
` 19 A. Well, at a high level I -- I think it's possible
`
` 20 that some software loads may have been able to
`
` 21 govern the rate, and it depends also -- I mean,
`
` 22 there may have been some phones available with
`
` 23 kind of burst mode photos, like five in a row or
`
` 24 something, in 2006. I'd have to look.
`
` 25 But, you know, I don't know if you consider a
`
`Lexitas
`
`Page 23 of 190
`
`
`
`Henry H. Houh, PhD
`
`24
`
` 1 five-frame sequence of burst mode photos a video
`
` 2 or you intend it to be, you know, a longer
`
` 3 sequence or longer time.
`
` 4 So you'd have to answer those questions, and
`
` 5 I think you'd have to actually look at a specific
`
` 6 camera and look at what software's available for
`
` 7 the camera and see if there's an application
`
` 8 which may be able to set the frame rate and then
`
` 9 see if videos captured with that particular
`
` 10 application indeed captured videos at the desired
`
` 11 requested frame rate through the software
`
` 12 interface that a user could set.
`
` 13 And so if there was such software available,
`
` 14 I think I'd start with that; and I could use that
`
` 15 software to tell you if it captured the video as
`
` 16 expected.
`
` 17 Q. You mentioned a burst of photos.
`
` 18 Would those photos captured in a burst like
`
` 19 that be saved as a common video file, or would it
`
` 20 have been more common for them to be saved as
`
` 21 individual photos?
`
` 22 A. I think it depends on -- I'm aware that today
`
` 23 sometimes, you know, some cameras can save or at
`
` 24 least treat burst mode photos as an animated
`
` 25 sequence akin to a video.
`
`Lexitas
`
`Page 24 of 190
`
`
`
`Henry H. Houh, PhD
`
`25
`
` 1 I don't off the top of my head know the
`
` 2 particular -- particulars of the file as to
`
` 3 whether it's one file or a sequence of files
`
` 4 offhand.
`
` 5 Q. Do you know whether there was any of that
`
` 6 functionality in 2006?
`
` 7 A. I believe -- I think -- I don't recall when that
`
` 8 feature was available.
`
` 9 I think that in a -- I think that in 2006 I
`
` 10 had a regular camera that could do burst mode
`
` 11 photos.
`
` 12 I don't know whether it was available on
`
` 13 camera phones at the time.
`
` 14 Q. Okay. So for the purposes of our discussion
`
` 15 today on 2006 camera phones, let's assume that
`
` 16 video is sort of traditional video; so, as we've
`
` 17 been talking about, video that's encoded and then
`
` 18 can be played as a -- as a video file.
`
` 19 Were there different frame rates at which
`
` 20 camera phones in 2006 could capture video files?
`
` 21 A. I mean, oftentimes phone manufacturers would
`
` 22 publish various specs of how fast a video could
`
` 23 perform or be captured, and it might just say
`
` 24 that you can capture, you know, full HD at 30
`
` 25 frames a second; or maybe they would tell you
`
`Lexitas
`
`Page 25 of 190
`
`
`
`Henry H. Houh, PhD
`
`26
`
` 1 that you could capture standard definition at 30
`
` 2 frames a second and high def at maybe a reduced,
`
` 3 but I think if you looked at phone specs at the
`
` 4 time from camera phones, I think some of them
`
` 5