• Microsoft Corporation v. X1 Discovery, Inc.No claims unpatentable
  PTAB · Jun 18, 2026 · IPR2025-00253 · APJ Scott C Moore · IPR · Final written decision

  Claims 1–10 were not shown unpatentable under either § 103 ground because the evidence did not establish a reason to combine the references or a reasonable expectation of success.

  RulingsFinal written decision

  • Skill levelSkill level patent owner adoptedDecided for patent owner

    “In our Decision on Institution, we adopted Mr. Haeberli’s formulation, finding it to be consistent with the evidence of record at this stage of the proceeding, including the asserted prior art and the ’035 patent. D.I. 8. Following a review of the full record, we agree with Mr. Haeberli’s statement that Petitioner’s definition of one of ordinary skill in the art is “not meaningfully different” from Mr. Haeberli’s for the purposes of this proceeding. We continue to view Mr. Haeberli’s proposed definition as consistent with the ’035 patent and the asserted prior art, and we apply it in our analysis below. However, our analysis of the issues below would have been the same had we instead adopted Petitioner’s proposed definition. C.” slip op. at 11

  • Claim Construction › Threshold & procedure › Whether a term needs construction (O2 Micro)Term construed board ownDecided for patent owner

    “In our Institution Decision, we construed the claim term “string separator character” to mean “a single-character string separator,” and invited the parties to raise any issues they had with this construction in future briefing. D.I. 10. Patent Owner subsequently indicated that it agreed with this construction. See PO Resp. 8. Petitioner did not raise any objections to this construction (see generally Pet. Reply). Accordingly, and in the absence of any dispute, we construe “string separator character” to mean “a single-character string separator.” Neither Petitioner nor Patent Owner ask us to adopt explicit constructions of any other claim terms. See Pet. 14, PO Resp. 8–9; Pet. Reply 1–23; PO Sur-reply 2–7. We also find that the issues in dispute can be resolved without construing any other claim terms. Accordingly, we do not adopt explicit constructions of any other claim terms. See Realtime Data, LLC v. Iancu, 912 F.3d 1368, 1375 (Fed. Cir. 2019) (“The Board is required to construe ‘only those terms … that are in controversy, and only to the extent necessary to resolve the controversy.’”) (quoting Vivid Techs., Inc. v. Am. Sci. & Eng’g, Inc., 200 F.3d 795, 803 (Fed. Cir. 1999)). D.” slip op. at 12

  • Obviousness (§103) › Motivation to combine / reasonable expectation of successNot shown unpatentableMotivation not foundNot shown unpatentable

    “search functionalities disclosed in the Lotus Notes References and Raskin are materially different in such a way as to render combining them anything but simple and straightforward.” See PO Resp. 26–27. Indeed, Petitioner has given little or no explanation of how one of ordinary skill in the art allegedly would have taken the “idea of incremental search” and applied it to Lotus Notes. See, e.g., Ex. 2021, 116:20–117:17 (Petitioner’s declarant, Dr. Terveen, testifying that “Raskin is not explicitly talking about searching across a set of documents, but the teaching that I relied on in general from Raskin is the idea of incremental searching,” and “I did not have to talk about implementation level details or incorporating a particular implementation. That wasn’t necessary for my opinion.”). Raskin simply does not teach a mode of incremental searching that applies to the query-style, document-finding searching of the type Petitioner cites from Lotus Notes. See generally Ex. 1021; see also Ex. 2020 ¶¶ 34–35 (Mr. Haeberli contrasting the Lotus Notes searches cited by Petitioner with Raskin’s searches). Raskin even states that it is “difficult or impossible to implement incremental searches” using interface-building tools. Ex. 1021, 126; Ex. 2020 ¶ 60 (Mr. Haeberli’s testimony that this passage of Raskin “directly undermines Dr. Terveen’s suggestion that a POSITA would have been motivated to implement Raskin’s incremental search approach in the Lotus Notes system”).” slip op. at 24

  Patent outcomesUS 7,370,035 not shown unpatentable
  Claim construction§ 103 obviousnesscites Phillips · Graham
  Patents in suitUS 7,370,035
• Microsoft Corporation v. X1 Discovery, Inc.No claims unpatentable
  PTAB · Jun 18, 2026 · IPR2025-00254 · APJ Lisa A Murray · IPR · Final written decision

  Claims 1–20 were not shown unpatentable under either §103 ground because Petitioner failed to establish motivation to combine and a reasonable expectation of success.

  RulingsFinal written decision

  • Skill levelSkill level patent owner adoptedDecided for patent owner

    “29. In my opinion, the POSITA at the time of the alleged invention of the Asserted Patents would have had a (1) a Bachelor’s degree in a pertinent discipline, such as computer science, electric engineering, computer engineering, or a related field and (2) have 2–4 years of experience in working with computing devices and have an understanding of indexed search systems utilized by such computing devices. 30. I am aware that Petitioner has provided its own definition . . . . 31. I believe Petitioner’s definition is not meaningfully different from my definition of a POSITA. My opinions would not change regardless of which definition of POSITA is applied. Ex. 2001 ¶¶ 29–31. In our Decision on Institution, we adopted Mr. Haeberli’s formulation, finding it to be consistent with the evidence then of record, including the asserted prior art and the ’490 patent. Dec. 9. Following a review of the full record, we agree with Mr. Haeberli’s statement that Petitioner’s definition of one of ordinary skill in the art is “not meaningfully different” from Mr. Haeberli’s for the purposes of this proceeding. We continue to view Mr. Haeberli’s proposed definition as consistent with the ’490 patent and the asserted prior art, and we apply it in our analysis below. However, our analysis of the issues below would have been the same had we instead adopted Petitioner’s proposed definition.” slip op. at 9

  • Claim Construction › Threshold & procedure › Whether a term needs construction (O2 Micro)No construction neededNot reached

    “In our Decision on Institution, the only term that we found it necessary to construe was the term “string separator character” in claim 1. We preliminarily construed this term to mean “a single-character string separator.” Dec. 11, 30–31. We invited the parties to address the appropriateness of this construction over the course of the trial. Id. at 11, 31. In its briefing, Patent Owner agrees with the Board’s preliminary construction, arguing that it is “consistent with the intrinsic evidence.” PO Resp. 8 (citing Ex. 1001, claims 8 and 20). Patent Owner further argues that a person of ordinary skill would have understood a multi-character Boolean operator such as “AND” to be “a search syntax instruction rather than a string separator.” Id. at 9 (citing Ex. 1059, 162; Ex. 2001 ¶ 112). Petitioner’s Reply Brief does not address the issue of claim construction. See generally Reply Br.; see also Pet. 11 (“Unless otherwise noted, Petitioner applies the plain and ordinary meaning of all claim terms below.”). Following our review of the full record, we determine that it is not necessary to construe this or any other claim term for the purposes of this Decision.2 See Nidec Motor Corp. v. Zhongshan Broad Ocean Motor Co., 868 F.3d 1013, 1017 (Fed. Cir. 2017) (Only terms that are in controversy need to be construed, and then only to the extent necessary to resolve the controversy.). In our analysis below, we interpret the language of the claims according to its plain and ordinary meaning.” slip op. at 10

  • Expert testimonyPetitioner expert discountedPatent owner expert creditedDecided for patent owner

    “Having reviewed the evidence of record, we find Mr. Haeberli’s testimony to be more persuasive and better supported by the evidence. According to Dr. Terveen, “[r]obust programming languages such as C and C++, together with well-developed operating system and UI frameworks, . . . allowed an ordinary programmer to build a UI that supported incremental searching, as evidenced by the numerous prior art systems that implemented such functionality, including Entourage, Eudora, Emacs, Canon Cat, SwyftWare, and others. Ex. 1100 ¶ 43 (citing Ex. 1004, 8–9, 161; Ex. 1021, 125, 127, 129; Ex. 1039, 102–103; Ex. 1043, 385–389). The relevant question, though, is not whether it would have been possible to build a user interface that supported incremental search, but whether it would have been “simple and straightforward” to execute an incremental search of multiple partial search strings on large, indexed multi-file databases such as those described in the Lotus Notes references, as Dr. Terveen alleges. None of the prior art discussed in the “State of the Art” section of Dr. Terveen’s initial declaration, taken individually or together, discloses incrementally searching a plurality of files using two partial search strings. See Ex. 1003 ¶¶ 32–33, 37, 40; Ex. 2001 ¶ 100. Thus, creating a user interface might be “simple,” but the evidence suggests that the back-end execution of an incremental search is not. See, e.g., Ex. 1021, 126 (Raskin, concluding that “almost all interface-building tools make it easy to implement delimited searches and difficult or impossible to implement incremental searches”).” slip op. at 23

  • Obviousness (§103) › Motivation to combine / reasonable expectation of successMotivation not foundNot shown unpatentable

    “different approach from that taken in Lotus Notes. The difference between Raskin’s sequential scanning method and the index-based system described in Wilcox is that sequential scanning by definition can only scan one document at a time as each document is loaded into a “buffer.” See, e.g., Ex. 2020 ¶ 35. Lotus Notes, however, only needs to load the index into memory for a user to be able to locate the single index entry identifying all instances of an indexed term across all indexed files. In fact, Raskin acknowledges that indexing may be necessary to maintain an acceptable response time. Ex. 1021, 132 n.10; see also id. at 129 (indexing speeds search times). Mr. Haeberli’s third and final reason for concluding that one of ordinary skill would not have been motivated to combine Raskin’s incremental search with Lotus Notes is that at the time using incremental searches would have presented “extreme technical difficulties.” Ex. 2001 ¶ 101. Here again we find Mr. Haeberli’s explanation more persuasive than Dr. Terveen’s opinion that combining the teachings from Raskin with Lotus Notes’ teachings “would have required only simple and straightforward software modifications.” Ex. 1003 ¶ 80. According to Mr. Haeberli, “there are serious technical challenges with implementing the complex search strategies of the challenged claims in the systems on which Dr. Terveen relies.” Id. ¶ 96. As Patent Owner argues: searching across a set of files to find specific files requires the system to identify which files are in the set, access those files (which are often not loaded in active memory), search their contents, determine which files contain matches to search criteria, and then present results to the user; all tasks that do not exist in the same way (or at all) for a search confined to text in a single open file already loaded into memory. Sur-Reply 9; see Ex. 2020 ¶¶ 15, 35–36. There would have been other challenges as well. For example, “[t]he type of sequential scanning typically employed to search a single open document is very time and resource intensive if employed to search a large set of documents.” Ex. 2020 ¶ 15. Also, “[i]mplementing incremental search in the Lotus Notes indexed system would necessarily require a larger index, which would slow down, rather than increase the search speed.” PO Resp. 23; see Ex. 2020 ¶ 33. Finally, Raskin itself concludes that “even if incremental searching is advantageous, ‘almost all interface-building tools make it easy to implement delimited searches and difficult or impossible to implement incremental searches.’” Ex. 2001 ¶ 96 (emphasis added by Mr. Haeberli) (quoting Ex. 1021, 126). In short, Mr. Haeberli presents three significant reasons why there would have been no motivation or expectation of success in combining the teachings of Wilcox with those of Raskin. Dr. Terveen’s opinion to the contrary, on the other hand, is not supported by evidence. Dr. Terveen testifies that “[a] POSITA would have been motivated to perform Lotus Notes’ multi-string search incrementally for the reasons stated by Raskin: speed, effectiveness, and ease of use.” Ex. 1003 ¶ 78 (citing Ex. 1021 ¶¶ 125–127). The evidence, however, not only fails to support Dr. Terveen’s testimony, but also demonstrates that one attempting to make the proposed combination would not have reasonably expected the proposed combination to result in greater speed or effectiveness.” slip op. at 27

  Patent outcomesUS 10,552,490 not shown unpatentable
  Claim construction§ 103 obviousnesscites KSR
  Patents in suitUS 10,552,490
• Microsoft Corp. v. X1 Discovery, Inc.No claims unpatentable
  PTAB · Jun 18, 2026 · IPR2025-00255 · APJ William V Saindon · IPR · Final written decision

  Claims 1–27 were not shown unpatentable under either §103 ground because Petitioner did not establish a persuasive reason to combine the cited teachings with adequate factual support.

  RulingsFinal written decision

  • Skill levelSkill level patent owner adoptedDecided for patent owner

    “Patent Owner does not comment on Petitioner’s formulation of the level of ordinary skill, or offer its own formulation in its Response. Patent Owner’s declarant, however, states: 29. In my opinion, the POSITA at the time of the alleged invention of the Asserted Patents would have had a (1) a Bachelor’s degree in a pertinent discipline, such as computer science, electric engineering, computer engineering, or a related field and (2) have 2–4 years of experience in working with computing devices and have an understanding of indexed search systems utilized by such computing devices. 30. I am aware that [Petitioner] has provided its own definition . . . . 31. I believe [Petitioner’s] definition is not meaningfully different from my definition of a POSITA. My opinions would not change regardless of which definition of POSITA is applied. Ex. 2001 ¶¶ 29–31. In our Decision on Institution, we adopted Patent Owner’s formulation, finding it to be consistent with the evidence of record at this stage of the proceeding, including the asserted prior art and the ’139 patent. Dec. on Inst. 8. Following a review of the full record, we agree that Petitioner’s definition of one of ordinary skill in the art is “not meaningfully different” from Patent Owner’s for the purposes of this proceeding. We continue to view Patent Owner’s proposed definition as consistent with the ’139 patent and the asserted prior art, and we apply it in our analysis below. However, our analysis of the issues below would have been the same had we instead adopted Petitioner’s proposed definition.” slip op. at 10

  • Claim Construction › Threshold & procedure › Whether a term needs construction (O2 Micro)Term construed patent owner constructionOther

    “Petitioner does not assert any explicit claim constructions. Pet. 12. In our Decision on Institution, the Board construed “string separator character” as “a single-character string separator.” Dec. on Inst. 10, 28–29. Patent Owner explicitly agreed with this construction in its Response (Resp. 8–9), and Petitioner offered no further comment in its Reply. Although no claim construction issue is squarely before us now, we maintain our prior construction of “string separator character,” and do not construe any further limitations. See Realtime Data, LLC v. Iancu, 912 F.3d 1368, 1375 (Fed. Cir. 2019) (“The Board is required to construe ‘only those terms . . . that are in controversy, and only to the extent necessary to resolve the controversy.’”) (quoting Vivid Techs., Inc. v. Am. Sci. & Eng’g, Inc., 200 F.3d 795, 803 (Fed. Cir. 1999)).” slip op. at 11

  • Obviousness (§103) › Motivation to combine / reasonable expectation of successMotivation not foundReasonable expectation not foundNot shown unpatentable

    “We agree with Patent Owner that a problem with Petitioner’s combination is that it proposes to take the idea of incremental searching from Raskin and divorce it from Raskin’s teaching of incremental searching. See PO Resp. 20 (arguing, “a [person of ordinary skill in the art] must consider all the teachings of a reference, and not pick out general ‘concepts’ disclosed in the art without considering the context in which they are used”); see also id. at 17–28 (setting out the whole argument); Ex. 2020 ¶¶ 39–62 (Patent Owner’s declarant explaining why Petitioner’s approach is “technically problematic”). Raskin’s incremental search is tied to finding and moving to character strings within a document. Ex. 1021, 125. For example, if a user were to type a string that was found in the text, “the instance [of that string] is selected and the cursor placed immediately after the end of the selection.” Id. The cursor can move about the text as characters of the search string are added or deleted. Id. at 125–126. Thus, Raskin discloses in-text search, even when searching across documents. Id. at 132 (“After the local document has been completely searched the search then proceeds through the following documents until the end of the folder is reached, when the search continues from the beginning of the first document in the folder until the current, already searched document is reached. After cycling through the folder, the next larger domain is treated similarly, and so forth.”). Raskin even distinguishes its text-finding search from Lotus Notes’ document-finding search, stating that “[i]t is usually easy to tell in what domain you are searching, because you see the results in context; what you see is not just a list of file names, as with many present search systems.” Id. (emphasis added); see also PO Sur-reply 16 (citing the same page and arguing that “Raskin recommends doing the opposite of what is claimed”). This context gives credence to Patent Owner’s argument that “the search functionalities disclosed in the Lotus Notes References and Raskin are materially different in such a way as to render combining them anything but simple and straightforward.” PO Resp. 26–27. Indeed, Petitioner has given little-to-no explanation of how one takes the “idea of incremental search” and applies it to Lotus Notes. Ex. 2021, 116:20–117:17 (Petitioner’s declarant testifying that “Raskin is not explicitly talking about searching across a set of documents [at once], but the teaching that I relied on in general from Raskin is the idea of incremental searching,” and “I did not have to talk about implementation level details or incorporating a particular implementation. That wasn’t necessary for my opinion.”). Raskin does not teach a mode of incremental searching that applies to the query-style, document-finding searching in Lotus Notes. See generally Ex. 1021; see also Ex. 2020 ¶¶ 34–35 (Patent Owner’s declarant contrasting Lotus Notes and Raskin). Raskin even states that it is “difficult or impossible to implement incremental searches” using standard interface-building tools. Ex. 1021, 126; Ex. 2020 ¶ 60 (Patent Owner’s declarant stating that this passage “directly undermines [Petitioner’s declarant’s] suggestion that a [person of ordinary skill in the art] would have been motivated to implement Raskin’s incremental search approach in the Lotus Notes system”).” slip op. at 34

  • Obviousness (§103) › Motivation to combine / reasonable expectation of success › HindsightMotivation not foundNot shown unpatentable

    “Entourage and what True is doing and how it is doing it, Petitioner asserts that it would have been obvious to use True’s two search strings “to provide more targeted searching, as was common sense and well known in the art.” Pet. 60 (citing Pet. 8–9; Ex. 1057, 27); see also id. at 56, 66 (similar). But we are not persuaded by a preponderance of the evidence that True provides more targeted searching or better search results when finding documents. True itself appears silent on the issue. Petitioner cites Exhibit 1057, but all that stands for is the general premise that doing a Boolean AND search in a Web search engine reduces the number of search results. Ex. 1057, 27. Petitioner does not allege that True uses a Boolean AND search of the kind discussed in Exhibit 1057. Thus, the statement in Exhibit 1057 regarding reducing the number of search results is tied to a manner of performing Boolean searches, something Petitioner does not use to modify Entourage. In other words, Exhibit 1057 is of little relevance to the Entourage-True combination, and does not serve as a factual underpinning in support of Petitioner’s reasoning. Petitioner also cites to several other references that discuss Boolean-style searches, but these generic background disclosures do not explain why a person of ordinary skill in the art would have considered it obvious to use one or more features of the probabilistic-based interest level search from True in a document-finding application like Entourage. See Pet. 8–9 (string citing over a dozen references with no pinpoint citations). Accordingly, there appears to be little reason, absent hindsight, to incorporate some aspect of a text-as-proxy-for-sounds system like True into Entourage’s filter or Find feature to provide more targeted searching. Accord Ex. 2020 ¶ 120 (“Patent Owner’s declarant stating that “a [person of ordinary skill in the art] working on the problem of searching large document collections would not look to [True,] a system designed for navigating imprecise transcriptions of single audio files, as the technical approaches and constraints are entirely different”). Patent Owner argues, and we agree, that another problem with Petitioner’s combination is that True does not appear to actually provide more targeted or better searching. PO Resp. 59–60 (“True’s search capabilities are inherently imprecise”) (citing Ex. 1009, 1:18–27; Ex. 2001 ¶ 137); see also Ex. 1009, 4:31–37 (explaining that True uses a speech recognition system that calculates confidence levels based on a “% likelihood that [a word] corresponds to the portion of the digitized audio [at a given time]”); Ex. 2020 ¶ 114 (Patent Owner’s declarant stating that “the purpose of True is not to definitively locate where a search string actually appears”). Petitioner does not explain what about True would make some aspect of Entourage’s filter or Find features more targeted. Even if it was known to use Boolean AND searches, for example, Petitioner is not citing True to modify Entourage to have a Boolean search. True uses an “interest level” search instead, which is more inclusive than a Boolean AND search (it adds together hits for “queen” OR “mother,” and may return hits containing neither). Ex. 1009, 5:29–50, Fig. 2 (showing results list 22 having several results which do not indicate any of the searched-for words); see Ex. 2020 ¶ 116 (Patent Owner’s declarant analyzing True’s interest level search would also include “text [that] does not show the query word”). Petitioner does not appear to be proposing to incorporate an interest level search or a Boolean-style search into the filter or Find features, and we are not persuaded that True in fact provides more targeted searching. Even if the quality of the transcripts in True happened to be perfect (see Pet. Reply 10, arguing that “True describes searching a text file that has 100% accuracy”), the fact remains that True attempts to find text in particular time windows using an interest level search based on the probability that a word or words were spoken in a given time window, which is completely different from what Entourage is doing with its search (see Ex. 2020 ¶¶ 114–116, 121). In its Reply, Petitioner states that True teaches “an incremental search for multiple search strings entered in a search field.” Pet. Reply. 10. This is a broad generalization of what True does, made with the benefit of hindsight, but as we have just explained, True’s interest level search is different from the kinds of searches done in Entourage or even the Boolean-style web searches performed in the background references. See, e.g., Ex. 1057, 27. This means it is not a matter of taking one known teaching and simply applying it to another system. True is doing something different, and it’s not clear how True’s search would actually provide “precise, targeted search results” (Pet. Reply 3) given that True appears to suggest a less-targeted search than claimed (see, e.g., Ex. 1001, 29:62–63, claiming “identifying one or more documents included in both the first one or more documents and the second one or more documents”). See also Ex. 2001 ¶¶ 135 (Patent Owner’s declarant stating that “Entourage and True are directed to completely different purposes”), 136–138 (further analysis); Ex. 2020 ¶ 114 (stating that “the purpose of True is not to definitively locate where a search string actually appears”), ¶ 116 (stating that True returns text that does not even have the search word). True does not teach, suggest, or support Petitioner’s general premise that there is a benefit for an incremental search for multiple search strings entered in a search field or that such searches would provide precise, targeted search results. The problem with Petitioner’s analysis becomes more apparent when the Petition turns to the “string separator character” of claim 1. Notably, it is the claimed string separator character that converts a field with a single search string into a field with two search strings. See, e.g., Ex. 1098, 37:12– 41–4 (deposition of Patent Owner’s declarant) (discussing that a string separator character is what instructs the computer to break up one string of characters into separate strings of characters). Thus, one cannot discuss the string separator character apart from the limitations requiring first and second search strings in a single search field. Nevertheless, Petitioner treats the “string separator character” in isolation. See Pet. 61–62. Petitioner’s reason for modifying one or both of Entourage’s filter and/or Find feature(s) to have a string separator character is because “it was common to [use string separator characters].” Id. at 61. However, even if something was known, it does not mean it is obvious to incorporate it into one or both of these features in Entourage. KSR, 550 U.S. at 418 (“a patent composed of several elements is not proved obvious merely by demonstrating that each of its elements was, independently, known in the prior art”). Adding a string separator character to divide a single string into two separately searched strings changes how the software would function. See Ex. 1098, 39:22–40:8 (Patent Owner’s declarant testifying that software has to look for a string separator character in order to know multiple search strings were entered in the field). The space character in “Orson Welles,” used in the Advanced Find example in Barber, for example, is not a string separator character because one looking to exclude hits with “Orson Welles” is searching for the string “Orson Welles”—the space is another character in the name string whose inclusion serves a specific semantic purpose (delineating between a first and last name). See Ex. 2001 ¶ 151 (“there is no evidence that when a user types ‘Orson Welles’ into this Find command text box that a search is performed on ‘Orson’ and ‘Welles’ separately as opposed to searching on the full text string ‘Orson Welles.’”); Ex. 1006, 387 (discussing how exclude the specific person “Orson Welles” from the search, not to find someone having the first name “Orson” or the last name “Welles”); see also Ex. 1003 ¶ 144 (Petitioner’s declarant implicitly acknowledging that the space in “Orson Welles” is not a string separator character). Thus, the space character in “Orson Welles” does not signal to the computer to break the string up into two parts at the space and perform” slip op. at 42

  Patent outcomesUS 9,633,139 not shown unpatentable
  Claim construction§ 103 obviousnesscites KSR · Graham
  Patents in suitUS 9,633,139
• Digital Global Systems, Inc. v. Deepsig, Inc.Some claims unpatentable
  PTAB · Mar 6, 2026 · IPR2024-01358 · APJ William V Saindon · IPR · Final written decision

  Claims 1–3, 7–11, 15–19, 23, and 24 are unpatentable as obvious; claims 4–6, 12–14, and 20–22 were not shown unpatentable because the prior art did not teach the claimed received-signal feedback and transmitter-model update.

  RulingsFinal written decision

  • Skill levelSkill level petitioner adoptedDecided for petitioner

    “Petitioner asserts that: Persons of ordinary skill in the art (“POSITA”) would have a bachelor’s degree in electrical engineering, computer engineering, or an equivalent degree and at least two years of experience working with of radio communication systems and machine learning. Additional work or research experience can substitute for less or different education, and vice-versa. Pet. 3–4 (citing Ex. 1003 ¶¶ 33–34). Patent Owner does not appear to challenge this level of skill. We use Petitioner’s definition for purposes of this Decision because it appears consistent with the level of skill reflected in the prior art and ’704 patent.” slip op. at 7

  • Claim Construction › Intrinsic evidence (Phillips canons) › Plain and ordinary meaning to a POSITATerm construed board ownDecided for petitioner

    “parameters “associated with the base station.” PO Resp. 31–32. In particular, Patent Owner argues that a transmitter “implemented in a base station” is not “associated with the base station.” Id. at 31. Petitioner argues that Patent Owner offers no reason to believe that a transmitter “implemented in a base station” is not “associated with the base station.” Pet. Reply 4–5; see also id. at 3–6 (entire argument). A plain reading of a “model associated with a base station” would encompass a model that operates in a base station. We do not see anything in the ’704 specification that would indicate “associated with” has any special meaning, or has a meaning contrary to our understanding of the plain meaning. We do not see the word “associated” outside the claims, whereas we see the word “implement” throughout the specification, generally in reference to an actual use of a model (consistent with our plain reading). See, e.g., Ex. 1001, 5:35–39, 6:3–17, 10:55–11:3 (discussing “implementations” where the model or some other feature of the system is used in a particular way). Accordingly, we construe a model “implemented in a base station” to be “associated with the base station.”” slip op. at 12

  • Obviousness (§103) › Graham factors (scope of art, differences, level of skill)UnpatentableUnpatentable

    “scenarios” (see PO Resp. 16–21), Patent Owner argues that “Jüschke does not disclose any association of the weight or bias parameters, . . . to any deployment scenario, and . . . base station” (id. at 19). For the reasons explained in our claim construction section above, we construe “deployment scenarios” to include “inline mode” models located within a base station, according to the list of deployment scenarios set out in the ’704 patent. See Ex. 1001, 2:38–65. The in-line deployment scenario depicted in Figure 6A of the ’704 patent bears a striking resemblance to the arrangement depicted in Figure 1 of Jüschke. In both figures, the analog radio signal is fed directly into the model and then the model directly outputs the signal to the amplifier. Compare Ex. 1001, Fig. 6A, with Ex. 1005, Fig. 1. As construed, the “deployment scenarios” represent the physical configurations of the device in which the model is found, such as “in-line.” Petitioner shows by a preponderance of the evidence that Jüschke’s similar in-line configuration is a “deployment scenario” as claimed because it teaches an inline model for a cellular system transmitter (i.e., base station). Pet. 26–29 (citing, e.g., Ex. 1005, Fig. 1, ¶¶ 6, 59; Ex. 1003 ¶¶ 146–147); see also Ex. 1001, 2:37–41 (stating that “[t]he particular deployment scenario[] may be . . . sending a radio transmission from a cellular base station that includes the . . . model . . . in an inline mode”).” slip op. at 26

  • BurdenNot shown unpatentable

    “Based on our claim construction analysis of claim 4, above, we agree with Patent Owner that claim 4 requires an actually transmitted signal and an update of the transmitter’s model. This claim construction eliminates Petitioner’s position regarding Holt teaching a modeled signal. See, e.g., Pet. Reply 17–23. We then turn to Petitioner’s argument that Holt teaches updating a model using an actually transmitted signal. See Pet. 59–60. There, Petitioner relies on a disclosure in Holt of an “actual communication channel 302 used to train the channel model 208.” Ex. 1006, 18:1–2. This is depicted in Figure 6 of Holt, in connection with a “fourth loss function” which is used to “train the decoder model.” Id. at 18:42–48; see generally id. at 17:58–18:51 (describing Figure 6). Whether training the channel model or the decoder model, Holt here does not disclose using the actual communication channel to train the encoder, i.e., transmitter. In sum, we find that Petitioner has not shown by a preponderance of the evidence that the prior art describes the claimed step of “obtaining a received radio signal,” and even if it did, Petitioner has not shown by a preponderance of the evidence that the prior art describes the claimed step of “updating the value of the model parameters” in the manner required by claim 4. Claims 12 and 20 are nearly identical to claim 4, and Petitioner’s ground fails for similar reasons. See Pet. 76. Claims 5, 6, 13, 14, 21 and 22 depend from claims 4, 12, and 20, and Petitioner’s ground fails for similar reasons as those claims. See id. at 77–80.” slip op. at 31

  Patent outcomesUS 11,018,704 mixed
  Claim construction§ 103 obviousness
  Patents in suitUS 11,018,704
• Cisco Systems, Inc. v. Lionra Technologies LimitedAll claims unpatentable
  PTAB · Mar 4, 2026 · IPR2024-01281 · APJ Steven M Amundson · IPR · Final written decision

  Claims 1–3, 5–10, 12–15, and 17–21 are unpatentable as obvious over Cornett, Paatela, and Nelson; claim 8 is also unpatentable with Russell.

  RulingsFinal written decision

  • Obviousness (§103) › Motivation to combine / reasonable expectation of successMotivation foundReasonable expectation foundUnpatentable

    “been motivated to combine the teachings of Cornett, Paatela, Nelson, and Russell in the way Petitioner proposes and would have had a reasonable expectation of success. See supra §§ III.E.2(a)–(b). Hence, Petitioner has shown by a preponderance of the evidence that claim 8 is unpatentable under § 103(a) because the claim would have been obvious over Cornett, Paatela, Nelson, and Russell.” slip op. at 122

  • Skill levelOther

    “B. Level of Ordinary Skill in the Art Factors pertinent to determining the level of ordinary skill in the art” slip op. at 13

  Patent outcomesUS 7,738,471 unpatentable
  § 103 obviousnesscites Phillips · KSR · Graham
  Patents in suitUS 7,738,471
• Guangzho Eko Trading Development Co., Ltd. v. Nine Stars Group (U.S.A.) Inc.Instituted
  PTAB · Jan 14, 2026 · IPR2025-01191 · APJ Kara L Szpondowski · IPR · Institution

  The Board instituted IPR on claims 1–24 and all asserted grounds because the petition showed a reasonable likelihood of prevailing on at least one claim.

  RulingsInstitution decision

  • PTAB / AIA Trials › Institution threshold (reasonable likelihood / more likely than not)InstitutedInstituted

    “Institution of an inter partes review is authorized when “the information presented in the petition . . . and any response . . . shows that there is a reasonable likelihood that the petitioner would prevail with respect to at least 1 of the claims challenged in the petition.” 35 U.S.C. § 314(a). Based on the current record, and for the reasons explained below, we determine that Petitioner has established a reasonable likelihood that it would prevail with respect to at least one challenged claim, and we institute an inter partes review.” slip op. at 2

  • PTAB / AIA Trials › Discretionary denial › §325(d) same art / argumentsNot reached

    “IV. PATENT OWNER’S 35 U.S.C. § 325(D) ARGUMENTS Patent Owner presents arguments directed to discretionary considerations under 35 U.S.C. § 325(d) and requests the Petition be denied based on these arguments. Prelim. Resp. 2–16. However, Patent Owner’s § 325(d) arguments are untimely under the “Interim Director Discretionary Process,”6 and, as such, will not be considered by the Board. See Section I.C. (“The . . . patent owner should not present discretionary considerations in the . . . Patent Owner Preliminary Response (POPR)”); Section I.E. (“[A] Board panel will issue a decision on institution addressing the merits and other non-discretionary considerations.”). 5 Available at https://www.uspto.gov/patents/ptab/trial-practice-guide#step1 6 https://www.uspto.gov/patents/ptab/interim-director-discretionary-process.” slip op. at 38

  • Novelty / Anticipation (§102) › Prior-art status of a reference › Printed publication / public accessibilityPrinted publication establishedDecided for petitioner

    “Petitioner bears the burden of establishing that Wang is a printed publication. Medtronic, 891 F.3d at 1380. The determination of whether a document is a “printed publication” “involves a case-by-case inquiry into the facts and circumstances surrounding the reference's disclosure to members of the public.” In re Klopfenstein, 380 F.3d 1345, 1350 (Fed. Cir. 2004). “[A]t the institution stage, the petition must identify, with particularity, evidence sufficient to establish a reasonable likelihood that the reference was publicly accessible before the critical date of the challenged patent and therefore that there is a reasonable likelihood that it qualifies as a printed publication.” Hulu, IPR2018-01039, Paper 29 at 13.” slip op. at 19

  • Novelty / Anticipation (§102) › Anticipation standard (single reference, every element)Instituted

    “A claim is unpatentable as anticipated under 35 U.S.C. § 102 “if each and every limitation is found either expressly or inherently in a single prior art reference.” WhitServe, LLC v. Computer Packages, Inc., 694 F.3d 10, 21 (Fed. Cir. 2012) (quoting Celeritas Techs., Ltd. v. Rockwell Int’l Corp., 150 F.3d 1354, 1361 (Fed. Cir. 1998)). An anticipatory reference must also “disclose those elements ‘arranged as in the claim.’” Net MoneyIN, Inc. v. VeriSign, Inc., 545 F.3d 1359, 1369 (Fed. Cir. 2008) (quoting Connell v. Sears, Roebuck & Co., 722 F.2d 1542, 1548 (Fed. Cir. 1983)).” slip op. at 11

  Claim construction§ 102 anticipation§ 103 obviousnessDiscretionary denialcites KSR · Graham
  Patents in suitUS 10,822,165
• Arashi Vision (U.S.) LLC (D/B/A Insta360) v. Gopro, Inc.No claims unpatentable
  PTAB · Dec 10, 2025 · IPR2024-01186 · APJ Brian D Range · IPR · Final written decision

  Claims 1–20 were not shown unpatentable because Kwatra does not teach or suggest minimizing rotational velocity and acceleration as required by the claims.

  RulingsFinal written decision

  • Skill levelSkill level petitioner adoptedDecided for petitioner

    “Petitioner contends that a person of ordinary skill in the art (POSITA) would have the following characteristics: “a bachelor’s degree in engineering or computer science, or equivalent field, and two years of experience working on image or video processing or digital cameras.” Pet. 6 (citing Ex. 1002 ¶¶ 36–37). Petitioner states that, “[a]lternatively, a POSITA would have four years of experience working on image or video processing 6 Patent Owner asserts objective evidence of nonobviousness. Prelim. Resp. 19–27. or digital cameras.” Id. Petitioner further contends “[a]n advanced graduate degree in the disciplines cited above may make up for the work experience indicated above.” Id. Patent Owner does not dispute the level of ordinary skill in the art. Prelim Resp. 14–15. For purposes of this Decision, we adopt Petitioner’s proposal as reasonable and consistent with the ’894 patent specification and the prior art. See Okajima v. Bourdeau, 261 F.3d 1350, 1355 (Fed. Cir. 2001) (the prior art may reflect an appropriate level of skill in the art).” slip op. at 10

  • Claim Construction › Threshold & procedure › Whether a term needs construction (O2 Micro)No construction neededDecided for patent owner

    “Petitioner “does not believe any terms require a formal construction.” Pet. 6. Patent Owner agrees “no claim terms presently require construction to resolve any current dispute between the parties.” PO Resp. 15. We do not discern a need to construe explicitly any claim language at this juncture because doing so would have no effect on our analyses below of Petitioner’s asserted grounds and will not assist in resolving the present controversy between the parties. See Realtime Data, LLC v. Iancu, 912 F.3d 1368, 1375 (Fed. Cir. 2019) (“The Board is required to construe ‘only those terms . . . that are in controversy, and only to the extent necessary to resolve the controversy.’”) (quoting Vivid Techs., Inc. v. Am. Sci. & Eng’g, Inc., 200 F.3d 795, 803 (Fed. Cir. 1999)).” slip op. at 11

  • Obviousness (§103) › Graham factors (scope of art, differences, level of skill)Not shown unpatentableNot shown unpatentable

    “Kwatra teaches, at Equation 7, minimizing velocity and acceleration of a function representing camera path (P). FF ¶¶ 6, 8–9. Kwatra does not state that this camera path function considers rotation. Rather, the words “camera path” suggest x, y, z movement of the camera rather than rotation of the camera. Dr. Foroosh’s testimony that Kwatra’s minimization of a smoothed camera path does not result in minimizing rotational velocity and acceleration is credible because this testimony is consistent with Kwatra as a whole and consistent with the other evidence of record. Id. ¶ 10. Kwatra recognizes that image rotation angle (“θ”) may not remain smooth. Id. ¶ 12. But rather than attempting to minimize velocity and acceleration of rotation of the camera housing, Kwatra states that applying a smoothness constraint on rotation angle “may be non-linear.” Id. ¶ 13. Consistent with Kwatra, Dr. Foroosh credibly explains the difficulty of applying linear techniques to rotation. Id. Thus, Kwatra proposes instead constraining the sine and cosine of theta. Id. ¶ 14. The Kwatra provisional application is part of the Kwatra reference (id. ¶ 15), and it similarly opts for simpler schemes rather than estimating all parameters for all frames jointly. Id. ¶ 16. The provisional explains that the rotational component is relatively small such that non-linear computations minimizing velocity and rotation can be avoided. Id. ¶¶ 17–18. The testimony of Patent Owner’s witness, Dr. Foroosh, explaining the provisional is credible (Ex. 2018 ¶ 72) because it is consistent with Kwatra’s complete disclosure. Dr. Foroosh’s testimony is more credible than the testimony of Petitioner’s witness, Dr. Katsaggelos, because Dr. Katsaggelos did not review the Kwatra provisional application and, thus, did not review the entire disclosure of Kwatra. Id. ¶ 19; see also In re Etter, 756 F.2d 852, 859 (Fed. Cir. 1985) (en banc) (obviousness analysis considers whether “the claimed inventions are rendered obvious by the teachings of the prior art as a whole” (emphasis added)).” slip op. at 28

  • Expert testimonyPatent owner expert creditedPetitioner expert discountedDecided for patent owner

    “(id. ¶ 15), and it similarly opts for simpler schemes rather than estimating all parameters for all frames jointly. Id. ¶ 16. The provisional explains that the rotational component is relatively small such that non-linear computations minimizing velocity and rotation can be avoided. Id. ¶¶ 17–18. The testimony of Patent Owner’s witness, Dr. Foroosh, explaining the provisional is credible (Ex. 2018 ¶ 72) because it is consistent with Kwatra’s complete disclosure. Dr. Foroosh’s testimony is more credible than the testimony of Petitioner’s witness, Dr. Katsaggelos, because Dr. Katsaggelos did not review the Kwatra provisional application and, thus, did not review the entire disclosure of Kwatra. Id. ¶ 19; see also In re Etter, 756 F.2d 852, 859 (Fed. Cir. 1985) (en banc) (obviousness analysis considers whether “the claimed inventions are rendered obvious by the teachings of the prior art as a whole” (emphasis added)).” slip op. at 28

  Patent outcomesUS 10,574,894 not shown unpatentable
  Claim construction§ 103 obviousnesscites KSR · Graham
  Patents in suitUS 10,574,894
• Arashi Vision (U.S.) LLC (D/B/A Insta360) v. Gopro, Inc.Some claims unpatentable
  PTAB · Dec 2, 2025 · IPR2024-01052 · APJ Brian D Range · IPR · Final written decision

  Claims 1, 2, 7–12, and 17–20 are unpatentable as obvious; claims 3–6 and 13–16 were not shown unpatentable because the asserted Watanabe combinations were unsupported.

  RulingsFinal written decision

  • Skill levelSkill level petitioner adoptedDecided for petitioner

    “cameras.” Pet. 6 (citing Ex. 1002 ¶¶ 36–38). Petitioner further contends “[a]n advanced graduate degree in the disciplines cited above may make up for the work experience indicated above.” Id. Patent Owner does not dispute and applies Petitioner’s proposed level of skill. PO Resp. 6. We find that Petitioner’s proposal is reasonable and consistent with the ’832 patent specification and the prior art. See Okajima v. Bourdeau, 261 F.3d 1350, 1355 (Fed. Cir. 2001) (the prior art may reflect an appropriate level of skill in the art). Thus, we apply Petitioner’s proposal in our analysis.” slip op. at 13

  • Prior constructionPrior construction adoptedDecided for petitioner

    “In our Institution Decision, we construed “determin[ing] a viewing window for the visual content as a function of progress through the progress length” as “requiring determining a viewing window for different times of the video.” Inst. Dec. 21. Neither party disputes this construction. PO Resp. 7; Pet. Reply 1–7. We maintain this same construction.” slip op. at 13

  • Obviousness (§103) › Motivation to combine / reasonable expectation of successMotivation foundReasonable expectation foundDecided for petitioner

    “Thomason are in the same field of endeavor. Thomason describes a camera and method for correctly orientating pictures from the camera. See, e.g., Ex. 1004 ¶¶ 1, 2 29–40, Figs. 3a–3b. In particular, Thomason is interested in rotating and cropping images to reduce the effect of poor orientation. Pet. 9 (citing Ex. 1004 ¶¶ 2, 34, 36). Pacurariu likewise describes aligning and resizing/cropping of a captured image to correct poor orientation. See, e.g., Ex. 1005 ¶¶ 1, 3, 45–56, Figs. 5A–5C. The preponderance of the evidence also supports that a person having ordinary skill in the art would have had reason to combine the Pacurariu’s teachings with Thomason to predictably improve Thomason. Thomason teaches correct rotation of video data and makes use of rotation sensor to correct rotated images. Pet. 19–21 (citing, e.g., Ex. 1004 ¶¶ 4, 27, 29, 35; Ex. 1002 ¶¶ 76–77). Pacurariu similarly uses sensors and measures tilt angle of the camera. Id at 21–22 (citing, e.g., Ex. 1005 ¶¶ 51–52, Fig. 6; Ex. 1002 ¶ 78). Pacurariu further teaches that its tilt data angle can be stored for each “video frame.” Id. at 22 (citing Ex. 1005 ¶ 52). Petitioner, with support from credible testimony of Dr. Katsaggelos, explains how the teachings of Thomason and Pacurariu could be successfully combined. Id. at 23 (citing, for example, Ex. 1002 ¶ 81). A person of ordinary skill in the art could have known that Thomason’s system could be improved by saving rotation data for each video frame and correcting images on a frame-by-frame basis. Id. at 23–24. The testimony of Dr. Katsaggelos is credible in this regard. Ex. 1002 ¶¶ 80–81. Petitioner and Dr. Katsaggelos explain the references’ teachings and, based on these teachings, assert that a person having skill in the art would have had reason to combine with reasonable expectation of success based on those teachings. Reason to combine may properly come from the references themselves. KSR Int’l Co., 550 U.S. at 418 (explaining reason to combine may come from “interrelated teachings of multiple patents”). Based on the evidence presented, Petitioner establishes by a preponderance of the evidence that a person having ordinary skill in the art would have had reason to combine the teachings of Thomason and Pacurariu with a reasonable expectation of success. Ex. 1002 ¶¶ 76–81, 89–92; Ex. 1004 ¶¶ 4, 8–9, 27, 29, 35; Ex. 1005 ¶¶ 26, 41, 46, 51–52, 55–57, Figs. 6, 7.” slip op. at 27

  • Obviousness (§103) › Objective indicia / secondary considerations › NexusObjective indicia no nexusDecided for patent owner

    “evaluating nexus between the claimed invention and objective evidence of nonobviousness, also referred to as secondary considerations. Lectrosonics, Inc. v. Zaxcom, Inc., IPR2018-01129, Paper 33 at 33 (PTAB Jan. 24, 2020) (precedential). We first consider whether Patent Owner has demonstrated “that its products are coextensive (or nearly coextensive) with the challenged claims,” resulting in a rebuttable presumption of nexus. Id. at 33. If not, that “does not end the inquiry into secondary considerations”; “the patent owner is still afforded an opportunity to prove nexus by showing that the evidence of secondary considerations is the ‘direct result of the unique characteristics of the claimed invention.’” Id. (quoting Fox Factory, 944 F.3d at 1373–75).” slip op. at 62

  Patent outcomesUS 11,336,832 mixed
  Claim construction§ 103 obviousnesscites KSR · Graham
  Patents in suitUS 11,336,832
• Arashi Vision (U.S.) LLC (D/B/A Insta360) v. Gopro, Inc.Some claims unpatentable
  PTAB · Nov 26, 2025 · IPR2024-00984 · APJ Nabeel U Khan · IPR · Final written decision

  Claims 6 and 7 are unpatentable as obvious over Meulen and Chao; claim 1 was not shown unpatentable because Meulen does not disclose transforming an image captured by its camera.

  RulingsFinal written decision

  • Skill levelSkill level petitioner adoptedDecided for petitioner

    “Petitioner argues that a person of ordinary skill in the art in the relevant field “would have a bachelor’s degree in engineering or computer science, or equivalent field, and two years of experience working on image or video processing, or digital cameras,” or alternatively “would have four years of experience working on image or video processing, or digital cameras.” Pet. 6. Petitioner further argues that an “advanced graduate degree” may “make up for the work experience.” Id. (citing Ex. 1002 ¶¶ 33–34). “Patent Owner applies Petitioner’s proposed level of skill.” PO Resp. 4–5 (citing Ex. 2027 ¶¶ 19–20). We adopt Petitioner’s proposed level of ordinary skill because it is consistent with the level of skill in the art at the time of the invention as reflected by the ’413 patent and the cited prior art.” slip op. at 9

  • Claim Construction › Intrinsic evidence (Phillips canons) › Plain and ordinary meaning to a POSITATerm construed petitioner constructionDecided for petitioner

    “From the outset we note that the words of the claim recite “non- uniformly shift[ing] the pixels” and do not recite a “non-linear” transformation. This favors Petitioner’s position, which is based directly off the definition of “non-uniform,” over Patent Owner’s position, which substitutes “non-linear” for “non-uniform.” Generally, then, we would determine the ordinary and customary meaning of the claim term in the context of the entire Specification. Here, however, the ʼ413 patent does not mention “uniform” or “non-uniform” shifting of pixels anywhere except in the claims. The Specification, therefore, provides limited guidance, by itself, as to the ordinary and customary meaning of the term. Nonetheless, even reviewing the portions of the Specification cited by the parties, we see no clear or unambiguous effort to limit the challenged claims to require only non-linear transformations in which the pixels to which a transformation is applied do not have the same amount of shift. is not non-linear. Pet. Reply 1. Specifically, Petitioner argues that the center circle of Figure 2B maintains the original circular shape after the transformation is applied to the image which contradicts Patent Owner’s argument that no two pixels will have the same amount of shift. Pet. Reply 1–2. We agree. In Figure 2B of the ʼ413 patent, the center of the image is chosen to be the region of interest. Ex. 1001, 4:19–21. After the input image goes through linear scaling (step 204) and then non-linear warping (step 206) the circles become more vertically compressed as they approach the top and bottom edges and more horizontally stretched as they approach the left and right edges. Id. at 4:22–30. But “the center circle is fully corrected so that it maintains the original circular shape of the corresponding circle in the input image 212.” Id. at 4:23–25. This means that at least some pixels, such as those of the center circle, undergo the same amount of shift after being transformed.” slip op. at 11

  • Claim Construction › Threshold & procedure › Whether a term needs construction (O2 Micro)No construction neededOther

    “readable instructions” of claims 1 and 2 must occur in a particular order. Specifically, Patent Owner argues that the step of “determine a portion within the input image” must occur after “obtain[ing] an image” and before “apply[ing] a transformation to the input image.” PO Resp. 17 (citing Ex. 2027 ¶ 59). Petitioner argues, on the other hand, that the specification does not require such an order and, in fact, describes an embodiment where such an order is not followed. Pet. 3. We determine that the “determine a portion within the input image” step does not need to occur after the “obtain an image” step and before the “apply a transformation to the input image” step. As Patent Owner itself points out, claims 1 and 2 are apparatus and system claims respectively, and an ordering of the limitations of such claims is not necessary. PO Resp. 17. Even if we were to treat the apparatus and system claims as reciting the steps of a method, “as a general rule the claim is not limited to performance of the steps in the order recited, unless the claim explicitly or implicitly requires a specific order.” Baldwin Graphic Systems, Inc. v. Siebert, Inc., 512 F.3d 1338, 1345 (Fed. Cir. 2008) (citing Interactive Gift Express, Inc. v. Compuserve Inc., 256 F.3d 1323, 1342–43 (Fed. Cir. 2001)). We see nothing in the claims that would cause us to break this general rule. Patent Owner argues that the claims imply a “specific order because the ‘determine a portion’ requirement operates on ‘the [obtained] input image’ and defines the ‘portion’ that is necessary for the 23; Ex. 2027 ¶ 59). Although the steps are recited in a specific order in the claim, as they necessarily must be, the claim does not expressly specify that the certain steps must occur before or after other instructions, nor does logic imply such a requirement. For example, if certain characteristics of the image are known before the image is obtained (e.g. the size, dimensions, resolution, aspect ratio, or content etc.) the determination of the portion of the image could be made before the image is obtained. These characteristics may be based on the specific device taking the image or the subject matter of the image, both of which may be known before the image is captured or obtained. It is, therefore, not necessary that the image be obtained in order for a portion of the image to be determined. Similarly, we do not find the written description to expressly or implicitly specify an order the claimed step of “determine a portion of the input image.” For example, the written description of the ʼ413 patent states that in Figure 2B “the center of the image is chosen to be the region of interest that is most fully corrected and the distortion increases in each direction as the frame boundary is approached.” Ex. 1001, 4:19–22. Patent Owner emphasizes that this disclosure implies that an input image is obtained before the center portion is chosen. PO Resp. 19 (citing Ex. 1001, 3:63–64, 4:19–22). But the description of this embodiment does not expressly recite an order. For example, the description of Figure 2A explains that “an input image is obtained” in step 202, “the image is then scaled” in step 204, and a “non-linear warping is then applied” in step 206. Ex. 1001, 3:63–4:3. This description does not mention “determin[ing] of a portion of the image” as a separate step or as part of the listed steps, and therefore, does not specify an order in which this step must occur. is configured so that some areas of the image are fully corrected geometrically, while some other areas are distorted. . . . [T]he center of the image is chosen to be the region of interest that is most fully corrected and the distortion increases in each direction.” Id. at 4:17–19. This excerpt from the ʼ413 patent, which the Patent Owner specifically relies on (PO Resp. 19), also does not specify an order in which the determination step (choosing of the region of interest) takes place. Moreover, because the ʼ413 patent states that the warp module is configured so that some areas are fully corrected while others are not, the determination step could take place at the time the warp module was configured, presumably before the image was obtained. Finally, the ʼ413 patent states that “in the case, of video, the area may dynamically change between images (e.g. to follow a moving object).” Ex. 1001, 4:34–36. Patent Owner argues that this description “requires first obtaining the image in order to locate the object that is being followed.” PO Resp. 19–20. This embodiment is related to the case of video or a moving object, which the challenged claims do not mention. We, therefore, decline to limit the scope of the claim based on this embodiment. Phillips v. AWH Corp., 415 F.3d 1303, 1323 (Fed. Cir. 2005) (en banc) (“[A]lthough the specification often describes very specific embodiments of the invention, we have repeatedly warned against confining the claims to those embodiments.”).” slip op. at 14

  • Obviousness (§103) › Motivation to combine / reasonable expectation of successMotivation foundDecided for petitioner

    “Having reviewed the parties’ arguments and cited evidence we determine Petitioner has established, by a preponderance of the evidence, that a POSITA would have been motivated to combine Meulen and Chao. For example, Petitioner argues that “[i]t would have been obvious that when performing image scaling, pixels from the input to output positions are shifted based on a scaling factor.” Pet. 91. Meulen teaches scaling but does not provide the details regarding scaling. E.g., Ex. 1007, 1:55–56, 4:37–49. We agree that in light of Meulen’s disclosure of scaling, a POSITA would have looked to other references, such as Chao, for further details on how to perform scaling. Dr. Katsaggelos provides credible support, testifying that “Meulen mentions ‘compression and/or scaling,’ and provides details on how to perform compression. But Meulen does not disclose the details of how to perform scaling. A POSITA when faced with Meulen and having the knowledge about scaling would have been motivated to turn to a prior art reference that discloses the details of scaling.” Ex. 1002 ¶ 262. Patent Owner argues that Petitioner provides no motivation to combine other than asserting that it involves combining prior art elements. PO Resp. 35 (quoting Pet. 92). As explained above, however, Petitioner does more than just assert that the motivation involves combining prior art elements. Instead, Petitioner explains that Meulen discusses both compression and scaling, and provides details regarding compression but does not provide details on how to scale. Pet. 91–92; Ex. 1002 ¶¶ 262–262. Thus, Meulen teaches that scaling was a known technique. See Hybritech, (“a patent need not teach, and preferably omits, what is well known in the art”). Dr. Katsaggelos testifies that, in this situation, a POSITA would have looked to Chao for details regarding scaling. Ex. 1002 ¶¶ 262–263. We determine this establishes, by a preponderance of the evidence, that a POSITA would have had reason to combine Meulen’s teachings with Chao. Patent Owner draws a distinction between Meulen and Chao, arguing that “Meulen is concerned with modifying content from a media source to be displayed on a device’s screen while Chao, by contrast, is directed to changing the aspect ratio of a document intended for printing.” Id. at 36 (citing Ex. 2027 ¶¶ 89–90). Patent Owner’s argument is contradicted, however, by Chao’s disclosure which states that “after all the background and foreground elements in the image have been resized . . . [t]he image may be output by displaying the image on or as part of: a display screen such as a computer monitor, camera screen, printer screen, printed page, projected image, or the like.” Ex. 1008 ¶ 32. Patent Owner argues that Meulen’s system avoids distortion near the middle where the user’s focus is more likely to be, while Chao, on the other hand, is directed to avoiding distortion of stylized borders on differently sized pages. PO Resp. 36–37 (citing Ex. 1006, 1:26–29; Ex. 1008 ¶¶ 1, 25; Ex. 2027 ¶¶ 91–92). This argument reads the prior art too narrowly and does not address Petitioner’s combination. Petitioner does not rely on Chao for its teaching of avoiding distortion near the borders of the image, but rather on its teaching of scaling the image while maintaining the design intent of the original image. Ex. 1008 ¶ 22; Ex. 1014 ¶ 67. Thus, Petitioner is applying Chao’s techniques of scaling (regardless of where the area of 92. Finally, Patent Owner argues that Chao’s approach of stretching/shrinking only the background and uniformly scaling entire foreground objects is suitable for graphic design but incompatible with Meulen which is concerned with adapting externally derived media source content for display on a fixed size screen. PO Resp. 37–38 (citing Ex. 2027 ¶ 93). We disagree with Patent Owner’s argument because it does not address Petitioner’s combination. Petitioner relies on Chao’s teaching of stretching/shrinking of background elements and applies that stretching/shrinking to Meulen’s images displayed on its screen. Pet. 91–92 (citing Ex. 1008 ¶¶ 28, 63; Ex. 1002 ¶ 263). The fact that Chao applies its teaching to graphic design does not prevent a POSITA from applying those teachings to Meulen’s media content displayed on its screen.” slip op. at 20

  Patent outcomesUS 10,015,413 mixed
  Claim construction§ 103 obviousnesscites Phillips · KSR · Graham
  Patents in suitUS 10,015,413
• Arashi Vision (U.S.) LLC (D/B/A Insta360) v. Gopro, IncNo claims unpatentable
  PTAB · Nov 25, 2025 · IPR2024-01165 · APJ William V Saindon · IPR · Final written decision

  Claims 1–21 were not shown unpatentable because Zhou does not teach the claimed low-light high-pass or stickiness parameter limitations.

  RulingsFinal written decision

  • BurdenFwd all claims patentableNot shown unpatentable

    “4. Conclusion for the Prior Art Analysis In sum, we find that Zhou does not disclose the low-light high-pass parameter or the stickiness parameter. Petitioner’s grounds for both independent claims 1 and 16 rely on Zhou for teaching a low-light high-pass parameter or a stickiness parameter. Pet. 23–28, 64. Because we find Zhou does not teach these limitations, all of Petitioner’s grounds fall for this reason. III. CONCLUSION For the reasons given above, we determine that Petitioner has not established by a preponderance of the evidence that any claim of the ’840 patent would have been obvious in view of the prior art cited.” slip op. at 18

  • Skill levelSkill level petitioner adoptedOther

    “A person of ordinary skill in the art (POSITA) in the field of the ’840 patent would have a bachelor’s degree in engineering or computer science, or equivalent field, and two years of experience working on image or video processing, or digital cameras. Alternatively, a POSITA would have four years of experience working on image or video processing, or digital cameras. An advanced graduate degree in the disciplines cited above may make up for the work experience indicated above. Pet. 6 (citing Ex. 1002 ¶ 32). Patent Owner applies the same standard. PO Resp. 26–27. We use this agreed-upon standard for purposes of this Decision.” slip op. at 7

  • Expert testimonyPatent owner expert creditedPetitioner expert discountedDecided for patent owner

    “We review this testimony based on the strength of the underlying factual basis. 37 C.F.R. § 42.65(a); Ashland Oil, Inc. v. Delta Resins & Refractories, Inc., 776 F.2d 281, 294 (Fed. Cir. 1985) (“Lack of factual support for expert opinion going to factual determinations, however, may render the testimony of little probative value . . ..”). Here, the underlying factual basis is not very strong for either witness. That said, we find that Dr. Foroosh’s testimony is more logically sound and consistent with the ’840 patent disclosure. Based on our reading of the specification, we find a clear indication that “stickiness” has to do with the past. In order to control “the extent by which preceding portion(s) . . . impacts determination of the smoothed trajectory” (emphasis added), one must control the degree by which the past frames impact determination. And to control the degree of the past’s influence, one must, as a consequence, favor the past over the future. That is, the frame of reference of the stickiness parameter is the past. The specification could have used a time-agnostic term, like a “nearness” or “adjacency” parameter, but it did not. Thus, the fact that “stickiness” is used in conjunction with “preceding” indicates that it deals with a preference for preceding frames. On this basis, we find Dr. Katsaggelos’s testimony less persuasive than Dr. Foroosh’s.” slip op. at 10

  • Claim Construction › Intrinsic evidence (Phillips canons) › Plain and ordinary meaning to a POSITATerm construed board ownDecided for patent owner

    “That said, we do not construe a stickiness parameter as prohibiting consideration of future frames. Instead, we determine that the stickiness parameter’s value determines the impact of the preceding portions, versus the following portions. The ’840 patent recognizes this, when it says that a stickiness parameter value of “1,” representing a “lazy” stabilizer that maximally favors the past, would not override a violation of the margin constraint based on the look-ahead (i.e., would not allow the viewing window to fall out of bounds in the event of large future motion). See Ex. 1001, 22:26–30; see also id. at 12:61–13:29 (acknowledging that favoring the past may lead to a conflict with the viewing window determination), 15:39–60 (adjusting the viewing window to ensure it does not violate a margin constraint). Thus, while a past-favoring parameter by necessity favors the past over the future, it does not prohibit consideration of the future. In view of the above, we construe the claimed “stickiness parameter” as a parameter that controls the extent by which preceding portions of the trajectory impacts determination of the smoothed trajectory, with an understanding that in order to favor preceding portions, to some extent, the parameter must relatively disfavor following portions.” slip op. at 10

  Patent outcomesUS 10,958,840 not shown unpatentable
  Claim construction§ 103 obviousnesscites Phillips · KSR · Graham
  Patents in suitUS 10,958,840

Holdings, summaries, rulings, issues, outcomes, winners and claim constructions shown for a decision were generated by a large language model reading the opinion. They are a guide to the decision, not the decision. Confirm every one against the court's own text before relying on it. Outcomes read from the patent owner’s side. Party names are extracted from the decision and are missing on about 900 short Director denials and scans, which carry the trial number instead.