Thursday, October 24, 2013 10:31 AM PDT
`Gal Shabtay
`
`Re: images for fusion algorithm evaluation
`
`From:
`Sent:
`To:
`CC:
`Subject:
`
`Hi Gal,
`
`My initial impression of the images CP provided has been quite positive. Thanks!
`
`I do have a few questions about the application:
`
`I assume that the application is being tuned for the Canon 40D with the 16-35mm lens, correct?
`What format will the application take (a mac app? windows? snapdragon? will it run on CPU /
`GPU). I feel like I should know the answers to these questions now but would like to have the
`details filled in.
`
`Will the application be able to process a pair of tele and wide images that have been captured
`with different f/#s? We expect that any practical camera system will have this.
`
`For the test system we are using with the Canon several things are over-simplified. Since the
`same sensor with the same CRA and pixel structure is being used the color & noise are the same
`for both wide and tele. Will I be able to add a color cast to the input images to simulate a real
`system that has two different sensors?
`
`Will the application work with only RAW images or can processed RGB images be passed in as
`well?
`
`Thanks in advance,
`
`
`
`@apple.com
`
`On Oct 24, 2013, at 5:44 AM, Gal Shabtay <gshabtay@corephotonics.com> wrote:
`
`Hi
`
`We are at the final stages of preparing the application (still on track for Oct. 27th).
`I was wondering if you have any feedback on the images we sent or anything else that we need to
`consider before sending the application.
`
`Thanks,
`Gal
`
`APPLE V COREPHOTONICS
`PROTECTIVE ORDER MATERIAL
`IPR2020-00906
`Exhibit 2020
`Page 1
`
`

`

`From: Gal Shabtay
`Sent: Sunday, October 20, 2013 1:13 PM
`To:
`Subject: RE: images for fusion algorithm evaluation
`
`Hi
`
`Thanks,
`Gal
`
`From: Gal Shabtay
`Sent: Friday, October 18, 2013 9:16 PM
`To:
`Subject: Re: images for fusion algorithm evaluation
`
`Hi
`
`Thanks,
`Gal
`
`
`
`Sent from my iPhone
`
`On 18 2013 קואב, at 20:49,
`
` <
`
`@apple.com> wrote:
`
`Hi Gal,
`
`Comments below,
`
`Thanks!
`
`
`
`
`
`
`
`
`
`
`
`
`APPLE V COREPHOTONICS
`PROTECTIVE ORDER MATERIAL
`IPR2020-00906
`Exhibit 2020
`Page 2
`
`

`

`
`
` :
`
`@apple.com
`
`On Oct 18, 2013, at 8:28 AM, Gal Shabtay <gshabtay@corephotonics.com> wrote:
`
`Hi
`
`Very good. Couple of items:
`
`
`
`
`
`
`
`
`
`Thanks,
`Gal
`
`On 18 2013 קואב, at 17:29, "
`
`" <
`
`@apple.com> wrote:
`
`Hi Gal,
`
`Thanks for clarifying. It was a mixup on my part. Thanks!
`
`
`
` :
`
`@apple.com
`
`On Oct 17, 2013, at 9:44 PM, Gal Shabtay <gshabtay@corephotonics.com> wrote:
`
`Hi
`
`Thanks for letting me know. Strange…that’s not what I am seeing.
`
`APPLE V COREPHOTONICS
`PROTECTIVE ORDER MATERIAL
`IPR2020-00906
`Exhibit 2020
`Page 3
`
`

`

`I am attaching a few snapshots from my screen.
`
`Can you check?
`
`Thanks,
`Gal
`
`From:
`@apple.com]
` [mailto
`Sent: Friday, October 18, 2013 4:01 AM
`To: Gal Shabtay
`Cc:
`Subject: Re: images for fusion algorithm evaluation
`
`Hi Gal,
`
`
`
`
`
`Thanks in advance!
`
`
`
` :
`
`@apple.com
`
`On Oct 17, 2013, at 5:07 AM,
`
` <
`
`@apple.com> wrote:
`
`Yes that's correct. Thanks for the rapid response!
`
`On Oct 17, 2013, at 3:46 AM, Gal Shabtay <gshabtay@corephotonics.com> wrote:
`
`Hi
`
`Thanks,
`Gal
`
`From: Gal Shabtay
`Sent: Thursday, October 17, 2013 11:17 AM
`To: '
`Cc:
`Subject: RE: images for fusion algorithm evaluation
`
`APPLE V COREPHOTONICS
`PROTECTIVE ORDER MATERIAL
`IPR2020-00906
`Exhibit 2020
`Page 4
`
`

`

`Hi
`
`Thanks,
`Gal
`
`
`
`
`
`From:
`@apple.com]
` [mailto:
`Sent: Thursday, October 17, 2013 4:19 AM
`To: Gal Shabtay
`Cc:
`Subject: Re: images for fusion algorithm evaluation
`
`Hi Gal,
`
`Is it possible to add the fusion results for the 3x zoom case for the following scenes?
`
`lab_800lx_100cm
`lab_10lx_100cm
`church_2
`target1_220cm
`
`Thanks again!
`
`
`
` :
`
`@apple.com
`
`On Oct 16, 2013, at 7:16 AM,
`
` <
`
`@apple.com> wrote:
`
`Hi Gal,
`
`This is fine. For the zero parallax case the best scenes for evaluation will be:
`
`lab_800lx_100cm
`lab_10lx_100cm
`church_2
`target1_220cm
`
`with zoom factors 1x, 1.5x and 2x.
`
`Thanks!
`
`APPLE V COREPHOTONICS
`PROTECTIVE ORDER MATERIAL
`IPR2020-00906
`Exhibit 2020
`Page 5
`
`

`

`
`
` :
`
`@apple.com
`
`On Oct 16, 2013, at 7:07 AM, Gal Shabtay <gshabtay@corephotonics.com> wrote:
`
`Hi
`
`We can do that.
`Can we compromise on ~3/10 scenes (of your choice)? Which zoom factors do you need?
`
`Thanks,
`Gal
`
`From:
`@apple.com]
` [mailto:
`Sent: Wednesday, October 16, 2013 4:51 PM
`To: Gal Shabtay
`Cc:
`Subject: Re: images for fusion algorithm evaluation
`
`Hi Gal,
`
`Will it be possible to expedite the processing of the 0mm (no parallax) baseline
`separation images?
`
`Thanks!
`
`
`
` :
`
`@apple.com
`
`On Oct 15, 2013, at 5:48 PM, Kevin Gross <kgross144@apple.com> wrote:
`
`Hi Gal,
`
`Thanks for the images. We’ll analyze the images further on our side but the first
`glance shows that the images look quite good. Thank you!
`
`A few comments below.
`
`APPLE V COREPHOTONICS
`PROTECTIVE ORDER MATERIAL
`IPR2020-00906
`Exhibit 2020
`Page 6
`
`

`

`Thanks,
`
`
`
` :
`
`@apple.com
`
`On Oct 15, 2013, at 11:47 AM, Gal Shabtay <gshabtay@corephotonics.com> wrote:
`
`Hi
`
`We have successfully downloaded the images and started to process them.
`I am uploading now 6 scenes (Target 1 & 2, Church 1 & 2, Lab-800lx 50cm & 100cm).
`
`Is it possible to include the 1.5x images as before?
`
`It is fine for now to look at just the 10mm separation. 15mm is extreme and I
`included it just so we can see how things brake down but in the short term looking at
`the 10mm separation is sufficient.
`
`APPLE V COREPHOTONICS
`PROTECTIVE ORDER MATERIAL
`IPR2020-00906
`Exhibit 2020
`Page 7
`
`

`

`Again, this is fine for now. Sticking with PS allows us to make apples to apples
`comparisons.
`
`I agree we won’t see this artifact in a real product and of course the images won’t be
`criticized in away way for the moving objects.
`
`We would be interested in your feedback on the images so that we can take it into account
`before the application delivery.
`Please let me know if you have any comments or questions.
`
`Thanks,
`Gal
`
`From:
`@apple.com]
` [mailto
`Sent: Tuesday, October 15, 2013 6:10 AM
`To: Gal Shabtay
`Cc:
`@apple.com;
`@apple.com;
`Subject: Re: images for fusion algorithm evaluation
`
`Hi Gal,
`
`@apple.com
`
`APPLE V COREPHOTONICS
`PROTECTIVE ORDER MATERIAL
`IPR2020-00906
`Exhibit 2020
`Page 8
`
`

`

`My attempt to send the images in batches of separate emails failed miserably. With any luck you
`should be able to download the images from the link embedded in this message. Please let me know if
`you can download this images.
`
`Thanks!
`
`Hi Gal,
`
`I’ve captured a new set of images for Core Photonics to use to evaluate the image fusion algorithm.
`The images are organized into 10 different experiments and I’ll email the sets separately to avoid
`having a single email message having too large of a file attachment. For each experiment there is a
`single “wide" image and 4 “tele" images. The “tele” images are from have varying amounts of parallax
`ranging from 0mm to 15mm as the baseline of the camera has been translated in 5mm steps. All
`together we’d like to have CP fuse together 80 images (10 experiments with 4 baselines per
`experiments and 2 raw conversions).
`
`The images in a given experiment are organized in the following way: All images were captured using
`a Canon 40D SLR using a Canon 16-35mm lens. For each image there is both a RAW file and the
`corresponding on-camera processed JPG image. All images were captured using an f/16 aperture (as
`per our discussion last week). With the exception of the low-light (10-lux) lab scene all images were
`captured at ISO 100 for very low noise. The low-light lab scene was captured at ISO 1600 for high-
`noise analysis.
`
`The differences between the experiments is the scenes (four scenes total: 3 in the lab and 2
`architectural), the light levels and the focus positions. The focus positions, light levels and baseline
`shifts should be clear from the file and folder names.
`
`For all “tele” images I have translated the camera forward by 6mm to correct for the shift of the
`entrance pupil of the Canon 16-35mm lens as the lens is zoomed from wide to tele. Previously we
`discussed that this axial translation should be 5mm but my own testing in the lab has determined that
`this axial translation is closer to 6mm.
`
`I would like for CorePhotonics to runs these images through the image fusion algorithm for each of the
`baseline separations. Specifically I would like to see the results for each of the baseline separations
`(0mm, 5mm, 10mm and 15mm cases). As per our discussion I understand that CP will use Adobe
`Camera RAW for the RAW conversion. Can you make sure that in addition to the fused images that
`you return to us the RAW-converted image files? Adobe Camera RAW has the capability to correct
`optical distortion, lateral chromatic aberration and lens vignetting in addition to the standard stuff
`(AWB etc). What I would like to see is the output of CP’s fusion algorithm when the RAW files are
`minimally processed (AWB only, no other histogram or geometry altering processing) and when the
`RAW files are maximally processed (AWB + distortion + lateral chromatic aberration + vignetting etc).
`This will represent the best and worst case scenarios.
`
`Please do not hesitate to contact me if you have any questions. I’ll be sending the images as
`attachments in separate emails.
`
`Thanks!
`
`------------------------------------------------------------
`
`APPLE V COREPHOTONICS
`PROTECTIVE ORDER MATERIAL
`IPR2020-00906
`Exhibit 2020
`Page 9
`
`

`

`The sender of this message has made the following files available to you at the
`Attaché website
` Church_1.zip
` Church_2.zip
` Lab_10lx_50cm.zip
` Lab_10lx_100cm.zip
` Lab_10lx_Infinity.zip
` Lab_800lx_50cm.zip
` Lab_800lx_100cm.zip
` Lab_800lx_Infinity.zip
` Target1_220cm.zip
` Target2_100cm.zip
`
`To download the files, please follow the link below.
`
`The files are available from 10/14/2013 00:00 PDT until 12/01/2013 22:59 PST.
`
`Important note: Each file can only be downloaded 12 times by all recipients.
`
`<Church_2_x2.jpg><Lab10lx_100cm_x2.jpg><Lab10lx_100cm_x1.jpg>
`
`APPLE V COREPHOTONICS
`PROTECTIVE ORDER MATERIAL
`IPR2020-00906
`Exhibit 2020
`Page 10
`
`