Imagination Technologies Limited

Patent Trials and Appeals Board

Sep 4, 2020

2019006809 (P.T.A.B. Sep. 4, 2020)

UNITED STATES PATENT AND TRADEMARK OFFICE
UNITED STATES DEPARTMENT OF COMMERCE
United States Patent and Trademark Office
Address: COMMISSIONER FOR PATENTS
P.O. Box 1450
Alexandria, Virginia 22313-1450
www.uspto.gov
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR ATTORNEY DOCKET NO. CONFIRMATION NO.
14/936,986 11/10/2015 Luke T. Peterson 2645-0156US02 2062
125968 7590 09/04/2020
Potomac Law Group PLLC (IMGTEC)
8229 Boone Boulevard Suite 430
Vienna, VA 22182
EXAMINER
BROOME, SAID A
ART UNIT PAPER NUMBER
2699
NOTIFICATION DATE DELIVERY MODE
09/04/2020 ELECTRONIC
Please find below and/or attached an Office communication concerning this application or proceeding.
The time period for reply, if any, is set in the attached communication.
Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the
following e-mail address(es):
Eofficeaction@appcoll.com
Patents@potomaclaw.com
vdeluca@potomaclaw.com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
________________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
________________

Ex parte LUKE T. PETERSON, JAMES ALEXANDER McCOMBE,
RYAN R. SALSBURY, and STEVEN J. CLOHSET
________________

Appeal 2019-006809
Application 14/936,986
Technology Center 2600
________________

Before JOHN A. EVANS, JAMES W. DEJMEK, and
RUSSELL E. CASS, Administrative Patent Judges.

CASS, Administrative Patent Judge.

DECISION ON APPEAL
Appellant1 appeals under 35 U.S.C. § 134(a) from the Examiner’s
final rejection of claims 1–20 under 35 U.S.C. § 112, and claims 1, 2, 4–10,
14–19 under 35 U.S.C. § 103. Appeal Br. 5, 8.2 We have jurisdiction under
35 U.S.C. § 6(b).
We affirm.

1 We use the word “Appellant” to refer to “applicant” as defined in 37 C.F.R.
§ 1.42 (2018). Appellant lists Imagination Technologies Limited of Kings
Langley, United Kingdom as the real party in interest. Appeal Brief filed
April 11, 2019 (“Appeal Br.”) at 1.
2 Rather than repeat the Examiner’s positions and Appellant’s arguments in
their entirety, we refer to the above mentioned Appeal Brief, as well as the
following documents for their respective details: the Final Action mailed
January 18, 2019 (“Final Act.”); the Examiner’s Answer mailed July 22,
2019 (“Ans.”); and the Reply Brief filed September 17, 2019 (“Reply Br.”).
Appeal 2019-006809
Application 14/936,986

2
BACKGROUND AND PROCEDURAL HISTORY
The present invention relates to using ray tracing to render two-
dimensional representations of three-dimensional scenes. Spec. ¶ 2.
Appellant’s Specification explains that rendering photo-realistic 2-D images
from 3-D scene descriptions with ray tracing is well-known in the computer
graphics art. Id. ¶ 3. According to the Specification, ray tracing usually
involves obtaining a scene description composed of geometric shapes, which
describe surfaces of structures in the scene, called primitives. Id. Using ray
tracing, virtual rays of light are traced into a scene from a view point (“a
camera”), and each ray is issued to travel through a respective pixel of the 2-
D representation, on which that ray can have an effect. Id. ¶ 4. The rays are
tested for intersection with scene primitives to identify a first intersected
primitive for each ray, if any. Id. After identifying an intersection for a
given ray, a shader associated with that primitive determines what happens
next, such as how the ray is reflected from a surface. Id. ¶ 5. According to
the Specification, most conventional algorithms build a tree of rays in flight
when ray tracing a scene, where the tree continues along each branch until it
leaves the scene or hits a luminaire that does not issue new rays. Id. ¶ 6.
Appellant’s Specification describes a variety of purported improvements to
ray tracing architectures. Id. ¶ 7.
The present application was the subject of a previous appeal to the
Board. Ex Parte Peterson, Appeal No. 2017-009671 (PTAB Mar. 27, 2018)
(“-9671 Appeal Dec.”). Claim 1 in that decision is representative of the
claims at issue in that appeal:
1. A method of rendering a plurality of images of a
particular instance of a 3-D scene from a respective plurality of
Appeal 2019-006809
Application 14/936,986

3
perspectives by ray tracing in a computer system, the method
comprising:
receiving shape data defining shapes to be rendered in the
particular instance of the 3-D scene from said plurality of
perspectives;
defining rays for the plurality of perspectives to be tested
for intersection in the particular instance of the 3-D scene;
processing rays from the different perspectives together, the
processing comprising at least one of:
testing the rays against common geometric shapes
within the particular instance of the 3-D scene, and
performing shading operations using a common shader
module.
-9671 Appeal Dec. 2.
In the previous appeal, the Examiner rejected the claims over Sven
Woop et al., RPU: A Programmable Ray Processing Unit for Realtime Ray
Tracing, ACM SIGGRAPH 2005 Papers (SIGGRAPH ’05) 434–444 (2005)
(“Woop”). In rejecting the claims, the Examiner relied on Woop’s ray
processing unit (“RPU”) that performs real-time rendering of three
dimensional scenes using real-time ray tracing. -9671 Appeal Dec. 3 (citing
Final Action mailed Aug. 22, 2016, at 2–3; Woop at 434–35, 441–43, Figs.
1, 2, 6). On appeal, Appellant argued that Woop does not disclose
“rendering a plurality of images of a particular instance of a 3-D scene from
a respective plurality of perspectives by ray tracing” because
“representations from different viewpoint perspectives in Woop are of
different scenes, as a scene may change over time,” and “[f]or any particular
instance of a 3-D scene in Woop, there is only one image rendered from only
one viewpoint.” Id. (citing Appeal Brief filed Jan. 24, 2017 (“2017 Appeal
Br.”) at 5, 6).
Appeal 2019-006809
Application 14/936,986

4
The Board rejected Appellant’s argument, determining that “[a]n
ordinarily skilled artisan would have understood that ‘a particular instance of
a 3-D scene’ can persist over time, and nothing in Appellant[’s]
Specification contravenes this plain and ordinary meaning.” -9671 Appeal
Dec. 4. “Thus,” the Board explained, “as the camera position in Woop
moves around an otherwise unchanging instance of a scene, Woop discloses
rendering a plurality of images of that particular instance of the 3D scene
from a respective plurality of perspectives, as recited.” Id.
Following the Board’s decision, Appellant amended the claims to
include additional language, which led to another final rejection and to the
present appeal. See May 29, 2018 Amdt. 2–5; Final Act.; Appeal Br. Claim
1 is illustrative of the claims presently on appeal, and is reproduced below
with italics indicating the language added to claim 1 in the -9671 Appeal.
1. A method of rendering a plurality of images of a
particular instance of a 3-D scene from a respective plurality of
perspectives by ray tracing in a computer system, wherein the
particular instance of the 3-D scene is an instance of the 3-D
scene at a particular instant of time, the method comprising:
receiving shape data defining shapes to be rendered in the
particular instance of the 3-D scene at the particular instant of
time from said plurality of perspectives;
defining rays for the plurality of perspectives to be tested
for intersection in the particular instance of the 3-D scene;
processing rays from the different perspectives together, the
processing comprising at least one of:
testing the rays against common geometric shapes
within the particular instance of the 3-D scene, and

Appeal 2019-006809
Application 14/936,986

5
performing shading operations using a common shader
module.
Appeal Br. 12 (Claims Appendix).
THE EXAMINER’S REJECTIONS
In the Final Office Action presently on appeal, the Examiner rejected
claims 1–20 under 35 U.S.C. § 112, first paragraph, as failing to comply
with the written description requirement. Final Act. 2–3. The Examiner
also rejected claims 1, 2, 4–10, 14, 15, and 17–19 under 35 U.S.C. § 103(a)
as being unpatentable over Woop in view of Ard (US 2005/0017971 A1;
published Jan. 27, 2005). Id. at 3. The Examiner further rejected claim 16
under § 103(a) as being unpatentable over Woop, Ard, and McGill (US
6,313,908 B1; issued Nov. 6, 2001). Id. at 6. The Examiner objected to
claims 3, 11–13, and 20 as being dependent upon a rejected base claim, and
indicated that they would be allowable if amended to overcome the § 112
rejection and rewritten in independent form. Id. at 7.
ANALYSIS
The Section 112 Rejection
In the § 112 rejection, the Examiner determines that the Specification
does not sufficiently describe the limitation in claim 1 reciting “wherein the
particular instance of the 3-D scene is an instance of the 3-D scene at a
particular instant of time, the method comprising: receiving shape data
defining shapes to be rendered in the particular instance of the 3-D scene at
the particular instant of time from said plurality of perspectives.” Final
Act. 2.
Appellant argues that the Specification “clearly conveys to those
skilled in the art that the invention relates to improvement in parallelization
Appeal 2019-006809
Application 14/936,986

6
of ray tracing processing to increase efficiency, with respect to a scene, i.e. a
single instance of a scene to be rendered,” relying on paragraphs 7 and 116
of the Specification. Appeal Br. 7; see Reply Br. 3–4. Appellant relies on
the following disclosure in paragraph 7:
“Ray tracing can be naively parallelized by providing many
processing resources that operate on different portions of pixels
of a 2-D scene to be rendered. However, simply providing
more computation capability does not necessarily allow a
suitable scaling of ray tracing speed and efficiency. One reason
for this is that such parallelization does not account for how
data composing the scene, or an acceleration structure that
increases tracing efficiency can be accessed in an efficient
manner. . . .”
Appeal Br. 7 (quoting Spec. ¶ 7).
With respect to paragraph 116 of the Specification, Appellant argues
as follows:
[P]aragraph [0116] of the original application specifically
discloses methods used in rendering representation of a 3-D
scene for use in holographic imaging systems wherein a
plurality of images of a given scene are to be rendered, each
from a different perspective. This clearly signifies to a person
having skill in the art that the “given scene” is a static set of
geometry, i.e. a “snapshot” at an instant of time. Viewing the
static scene from a plurality of different perspectives is what
enables a holographic image to be rendered. A holographic
image cannot be obtained by combining different viewpoints of
different temporal instances of scenes (such as in Woop where
different views of a scene as the scene changes over time are
obtained).
Appeal Br. 7.
The Examiner responds that Appellant’s “mere mention of a ‘scene’”
in paragraph 7 of the Specification “does not provide sufficient subject
matter to clearly convey to one skilled in the art at the time of invention that
Appeal 2019-006809
Application 14/936,986

7
a ‘scene’ is equivalent to a ‘particular instance of time’ or an ‘instant of
time,’” as recited in claim 1. Ans. 8. The Examiner further states that
although paragraph 116 “mentions that rays of each perspective are collected
together for intersection testing, [Appellant’s] Specification fails to mention
an ‘instant of time’ or a ‘particular instance of the 3-D scene’ in which the
intersection would be performed,” as recited in claim 1. Final Act. 7–8.
In order for a claim to satisfy the written description requirement of
§ 112, the written description “‘must clearly allow persons of ordinary skill
in the art to recognize that [the inventor] invented what is claimed.’” Ariad
Pharms., Inc. v. Eli Lilly & Co., 598 F.3d 1336, 1351 (Fed. Cir 2010) (en
banc) (citing In re Gosteli, 872 F.2d 1008, 1012 (Fed. Cir. 1989)). “[T]he
test for sufficiency is whether the disclosure of the application relied upon
conveys to those skilled in the art that the inventor had possession of the
claimed subject matter as of the filing date.” Id.
We agree with the Examiner’s finding that the Specification fails to
provide an adequate written description of the claim limitation at issue.
Appellant has not persuaded us that paragraphs 7 and 116, upon which
Appellant relies, adequately describe these claim limitations. Paragraph 7
describes the use of ray tracing for a “scene,” but fails to disclose receiving
shape data defining shapes to be rendered in “an instance of the 3-D scene at
a particular instant of time” from a “plurality of perspectives.” Appellant
appears to rely on the use of the term “scene” in paragraph 7, but fails to
point to any evidence in the Specification or elsewhere that a “scene” must
be interpreted to mean something that exists for only a particular “instant in
time,” as opposed to something that can persist for a period of time.

Appeal 2019-006809
Application 14/936,986

8
Turning to paragraph 116, that paragraph states as follows:
In one application, these systems and methods can be
used in rendering representations of a 3-D scene for use in
holographic imaging systems. In an example approach to
rendering for holographic imaging systems, a plurality of
images of a given scene are to be rendered, each from a
different perspective. In rendering such images, each
perspective can be considered to be an origin of rays to be
intersection tested. The rays of each perspective can be
collected together for intersection testing, such as collecting
rays of different origins and their progeny together, without
regard to their origins, but rather with respect to commonality
of intersection testing and/or shading to be performed.
Allowing collection of rays from a plurality of such origins
allows systems and methods to provide for setup of the 3-D
scene once, so that such scene setup is amortized over a large
number of image renderings. Also, combining rays to be traced
from different origins may allow for greater coherence and
overall processor utilization. Thus, in the above examples,
where collections of rays are formed, outputted, or otherwise
handled according to the disclosures, these rays can be
attributed to a plurality of camera positions. For example, rays
of a given collection can be tested against child nodes of a
parent node of a common acceleration structure.
Spec. ¶ 116.
We find that paragraph 116 discloses holographic imaging systems in
which a plurality of images of a scene are to be rendered from a different
perspective, but fails to disclose rendering images from a plurality of
perspectives “at a particular instant in time.” Indeed, paragraph 116 says
nothing about the particular timing for the various actions described.
Appellant’s argument that the discussion of “holographic imaging systems”
“clearly signifies” that images from multiple perspectives are being rendered
at a particular instant in time lacks sufficient evidentiary support in the
Specification or otherwise.
Appeal 2019-006809
Application 14/936,986

9
Consequently we sustain the Examiner’s § 112 rejection of claim 1,
and of independent claims 9 and 18, which include similar limitations, and
are not separately argued. We also sustain the Examiner’s § 112 rejection of
claims 2–8, 10–17, and 19–20, which are dependent on claims 1, 9, or 18.
The Section 103 Rejections
In the § 103 rejections, the Examiner found, inter alia, that Woop
teaches rendering a plurality of images of a 3-D scene from a plurality of
perspectives by ray tracing in a computer system, including receiving shape
data defining shapes to be rendered in the particular instance of the 3-D
scene from the plurality of perspectives, defining rays for the plurality of
perspectives to be tested for intersection in the particular instance of the 3-D
scene, and processing rays from the different perspectives together, where
the processing includes at least testing the rays against common geometric
shapes within the particular instance of the 3-D scene, and performing
shading operations using a common shader module. Final Act. 3–4.
The Examiner finds that Ard teaches providing a user “with the ability
to generate an image of [a] three dimensional scene of models acquired from
a real time interactive scene” and, therefore, “the image acquired during user
interaction is an imaged instance of time of the real time 3D scene.” Id. at 4
(citing Ard ¶¶ 45, 46) (italics omitted). The Examiner determines that it
would have been obvious to one of ordinary skill to modify the ray tracing
of Woop with the time instance image of Ard because this modification
would enable a user to interactively improve the visualization of a complex
three dimensional scene over any sequence of time through enabling a user
to image the scene at a given selected viewpoint at a particular instance of
time. Id. at 4–5.
Appeal 2019-006809
Application 14/936,986

10
Appellant argues that, unlike the claimed invention, “Woop is directed
to rendering only a single image from a single perspective, for any particular
instance of a 3-D scene.” Appeal Br. 8. “In Woop,” Appellant asserts,
“representations from different viewpoint perspectives are of different
scenes, i.e. as a scene may change over time,” and thus, “[f]or any particular
instance of a 3-D scene in Woop, there is only one image rendered from only
one viewpoint.” Id. Thus, Appellant argues, “[t]here is nothing in Woop
that relates to rendering a plurality of images of the same instance of a scene
from different perspectives (i.e. different viewpoints).” Id. at 9. Appellant
further argues that the cited portions of Ard “do not disclose processing rays
from different perspectives of an instance of a 3-D scene together.” Id.
Rather, according to Appellant, “Ard merely discloses rendering the scene
from a single viewpoint after the objects are arranged.” Id. at 9–10.
Appellant further argues that the rejection does not explain how Woop
would be modified by Ard to arrive at the claimed subject matter. Id. at 10.
The Examiner responds that “Woop does not merely rely upon
rendering a scene from a single perspective or instance, but clearly teaches
real time rendering of scenes, such as scenes generated from a video game.”
Ans. 9 (citing Woop Fig. 1 (“Realtime renderings on the RPU prototype
using a single FPGA running at 66MHz . . . and UT2003 a scene from a
current computer game (7.5 fps, precomputed illumination)” (italics
omitted)), § 5, second paragraph (“More realistic examples are taken from
computer games . . . such as Castle, UT2003, and Quake3. The latter one is
used for two benchmarks: as a single object, and again with several moving
players.”), § 5.2, third paragraph (“as shown in Figure 3, is possible for
scenes with a reasonable number of visible triangles as the RPU units are fed
Appeal 2019-006809
Application 14/936,986

11
from primary caches . . . small changes in the set of visible scene parts must
be transfer[r]ed per frame . . . the camera abruptly changes the view (e.g. by
walking around a corner)”)). Thus, the Examiner finds that “the scenes of
Woop may be rendered from different perspectives of the scene as the view
changes during traversal through the video game scene.” Id. “Therefore,”
according to the Examiner, Appellant’s arguments are unpersuasive because
“one skilled in the art at the time of invention would have clearly recognized
that the video game perspectives of a scene taught by Woop provides
continual perspective change to the players during traversal through a real
time video game scene.” Id.
As to Appellant’s argument that Ard fails to disclose “processing rays
from the different perspectives together,” the Examiner states that “Woop
was relied upon to teach processing rays from the different perspectives
together by processing several different perspectives from a plurality of
players using a ray processing unit during movement together through a
video game scene.” Id. at 9–10.
In its Reply, Appellant argues that “[r]endering a scene from different
perspectives as the view changes during a real time traversal by a player
through the scene” does not meet the claim language because “a traversal by
a player through a scene requires the passage of time and cannot be
accomplished during an instant of time.” Reply Br. 4. Appellant agrees
with the Examiner that “one skilled in the art ‘would have clearly recognized
that the video game perspectives of a scene taught by Woop provides
continual perspective change to the players during traversal through a real
time video game scene,’” but argues that this “fails to disclose rendering a
plurality of images of an instance of a 3-D scene at a particular instant of
Appeal 2019-006809
Application 14/936,986

12
time from a respective plurality of perspectives,” as claim 1 requires. Id. at
5.
We are not persuaded that the rejection is in error. As we found in the
-9671 Appeal Decision, Woop discloses the use of its RPU for efficiently
rendering views of a particular scene, with a focus on the “realtime” rate of
the rendering of each view. Woop 434 (Abstract), 441 (Results section,
Table 4); see -9671 Appeal Dec. 3. A person of ordinary skill would have
understood that Woop’s use of spatial index structures and “global” shading
effects to render those views discloses rendering views of a scene from
different perspectives, particularly with Woop’s contrasting of minor
viewpoint changes with “abrupt” changes and dynamic scenes. See Woop
442 (describing that, as the camera position changes, “only small changes in
the set of visible scene parts must be transferred [sic] per frame, unless the
camera abruptly changes the view (e.g., by walking around a corner)”); see
also Woop at 436 (Scalable Design section), Fig. 3. As we found in the
-9671 Appeal Decision, these disclosures indicate the use of Woop’s RPU
for a scene in which the camera view has changed, separate and apart from
changes to the scene itself—i.e., “a particular instance of a 3-D scene from a
respective plurality of perspectives.” See -9671 Appeal Dec. 4. Appellant
does not appear to dispute this. See Reply Br. 5 (“The Answer goes on to
state that one skilled in the art ‘would have clearly recognized that the video
game perspectives of a scene taught by Woop provides continual perspective
change to the players during traversal through a real time video game scene.’
. . . . Appellant agrees with this statement . . . .”).
The question before us now is what effect should be given to the
additional language added to claim 1 stating that the particular instance of
Appeal 2019-006809
Application 14/936,986

13
the 3-D scene is “an instance of the 3-D scene at a particular instant of
time.” Applying the broadest reasonable construction of the claim language,
we determine that “an instance of the 3-D scene at a particular instant of
time” encompasses a point in a graphical rendering (such as a video game)
in which no action is occurring and the scene appears to be static (or “frozen
in time”) to the user, but that the user can move their view and thereby see
this scene from different angles or perspectives. We see nothing in the
Specification or elsewhere that would limit the “particular instant in time” to
time in the “real world” as opposed to time in the context of the video game
or other graphical rendering being run on the system. Moreover, Appellant
has not pointed to any evidence in the Specification or elsewhere that the
“particular instant in time language” requires the production of a
holographic image.
We find that Woop discloses that the 3-D scenes which a player can
view from multiple perspectives may be static (or “frozen in time”). For
example, Woop refers to “closed room scenes of low complexity,” including
“a single object” and a “Conference scene [that] shows a conference room
with several instantiated chairs.” Woop at 441 (§ 5). Figures 1 and 6 of
Woop also show static scenes. Id., Figs. 1, 6. That a user may move around
a static scene is also consistent with Woop’s statement that “only small
changes in the set of visible scene parts must be transferred per frame, unless
the camera abruptly changes the view (e.g., by walking around a corner).”
For the above reasons, we sustain the Examiner’s § 103 rejections of
claim 1, and of claims 2, 4–10, and 14–19, which Appellant does not
separately argue.
Appeal 2019-006809
Application 14/936,986

14
CONCLUSION
We affirm the Examiner’s § 112 rejection of claims 1–20, and the
Examiner’s § 103 rejections of claims 1, 2, 4–10, and 14–19.
In summary:
Claims
Rejected
35 U.S.C.
§
Reference(s)/Basis Affirmed Reversed
1–20 112 Written
Description
1–20
1, 2, 4–10,
14, 15, 17–
19
103(a) Woop, Ard 1, 2, 4–10,
14, 15, 17–
19

16 103(a) Woop, Ard,
McGill
16
Overall
Outcome
1–20

No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a)(1)(iv). See
37 C.F.R. § 41.50(f).

AFFIRMED