US 10529052

US 10,529,052 B2 — Analyst Summary

Search basis / sourcing note: I could not query USPTO PatentCenter, PTAB, or the Federal Circuit's own CM/ECF directly from this environment. The findings below come from (a) the authoritative full-text record for US 10,529,052 supplied in your message (fetched from https://patents.google.com/patent/US10529052/en, which mirrors USPTO bibliographic data), (b) USPTO/PTAB and ITC documents surfaced by web search, and (c) secondary litigation trackers. Where a fact rests only on a secondary source, I flag it. No similar patent numbers were substituted — everything below is specific to 10,529,052 (the '052 patent). Note in particular the sibling patents 10,529,051, 10,535,115, 10,679,323, 10,186,012 and 10,395,338, which are separate members of the same family and are not the subject of this summary.

1. Bibliographic data (high confidence — from the patent record)

Caveat on assignee/ownership: the record also shows a security interest to JPMorgan Chase (2020), release (2021), and later security interests to Wells Fargo and Farallon Capital Management (both 2025). These are liens, not ownership transfers — GoPro, Inc. remains the assignee of record.

2. Abstract (verbatim)

"In a video capture system, a virtual lens is simulated when applying a crop or zoom effect to an input video. An input video frame is received from the input video that has a first field of view and an input lens distortion caused by a lens used to capture the input video frame. A selection of a sub-frame representing a portion of the input video frame is obtained that has a second field of view smaller than the first field of view. The sub-frame is processed to remap the input lens distortion to a desired lens distortion in the sub-frame. The processed sub-frame is the outputted."

3. Plain-language overview of the independent claims

Claim 1 — System

A system (one or more processors + non-transitory computer-readable storage medium) that does three things:

1. Accesses input images captured through a camera lens. Those images have fields of view of a scene and exhibit an input lens distortion centered in those fields of view (e.g., a fisheye/wide-angle curvature strongest at frame edges).
2. Selects sub-frames — cropped, smaller fields of view. Critically, the claim recites that these sub-frames contain different lens distortion effects, characterized as a function of three variables: (i) the input lens distortion present in the input images, (ii) the different positions of the sub-frame areas within the input images, and (iii) the size of those areas.
3. Generates output images using a desired lens distortion that is "consistent" in lens characteristics with the input lens distortion. The sub-frames are remapped from the input distortion (centered in the original fields of view) to the desired distortion (centered in the reduced fields of view), so the different distortions in the various sub-frames are normalized to one desired distortion and the scene appears as if it had been natively captured at the reduced field of view.

In plain terms: take a wide/fisheye shot, cut out a small window from anywhere in it, and re-render that window so it looks like a real camera had been pointed there and zoomed in — with the same lens "look," and without the window looking squashed, over-curved, or inconsistent with the other windows.

Claim 11 — Method

The same three steps (access → select → generate/remap), framed as a method rather than a system. Claim 11 is the method counterpart of claim 1 and is substantively parallel.

Dependent claims (themes)

• 2/12: image stabilization applied to the input images.
• 3–5/13–15: sub-frame selection based on metadata; metadata indicates camera motion (4/14) or orientation (5/15).
• 6/16: output images combined as frames of an output video.
• 7/17: remapping yields consistent frame-to-frame lens characteristics in the output video.
• 8/18: the varying sub-frame positions simulate virtual re-pointing of the camera.
• 9/19: the processing is performed in the camera itself.
• 10/20: the lens is a wide-angle lens causing straight lines in the scene to appear curved near image edges.

Note: the 2019/2020 '052 claims as issued do not expressly recite the two-step (inverse-then-apply) vs. direct-mapping distinction that appears in the specification — that express language shows up in sibling family members (e.g., claims 8–10/18–20 of US 10,535,115).

4. Enforcement and validity posture (search-derived; corroborate before relying)

• IPR2025-00017 — Arashi Vision (U.S.) LLC d/b/a Insta360 v. GoPro, Inc. Petition filed Oct. 15, 2024, challenging claims 1–20 over Okubo (US 2006/0017817), Okubo+Sokeila (US 8,994,838), and Eder (US 9,238,434). Institution denied April 28, 2025, on Fintiv-based § 314(a) discretion, in view of the parallel ITC investigation. The Google Patents "Family has litigation" line labels this "PTAB case IPR2025-00017 filed (Not Instituted – Procedural)." (One search result attributes the docket to "Unified Patents PTAB Data" — that is a data-provider attribution, not the petitioner; the petitioner is Insta360.)
• ITC Inv. No. 337-TA-1400, Certain Cameras, Camera Systems, and Accessories Used Therewith (instituted May 6, 2024, 89 FR 37242). GoPro asserted claims 1–10 of the '052 patent; claims were partially withdrawn/terminated. The ALJ's Final ID (July 10/11, 2025) addressed the '052 patent, and the Commission reviewed both the Element 1[g] limitation ("output images include the sub-frames remapped…") and the '052 invalidity findings in view of Okubo. In its final determination of Feb. 26, 2026, the Commission found no violation of § 337 with respect to the '052 patent (and the '840 patent), reversing the ALJ's infringement finding, and issued a limited exclusion order / cease-and-desist order based only on the D'435 design patent. Secondary Chinese-language reporting of Insta360's disclosure characterizes the final outcome as no infringement and invalidity of the '052 claims. Treat the precise invalidity holding as needing direct ITC-document confirmation.
• Court of Appeals for the Federal Circuit: the record lists a Federal Circuit case captioned GoPro, Inc. v. ITC, docket 26-1752 (full-form 2026-1752, docketed per a secondary tracker on Apr. 29, 2026). I could not independently confirm from Federal Circuit primary sources which patent(s) are on appeal or that the '052 patent is within the appealed issues — treat the '052-specific scope of this appeal as unverified. (There is also an unrelated 2026-2012 appeal, Contour IP Holding LLC v. GoPro, Inc., concerning different patents.)
• District court: GoPro, Inc. v. Arashi Vision Inc. et al., No. 8:24-cv-00681 (C.D. Cal.), automatically stayed under 28 U.S.C. § 1659 pending the ITC matter; reported as to resume after the ITC case closed.
• Family continuation activity: a further continuation, app. 19/455,238, published as US 2026/0154778 A1 (June 4, 2026) — i.e., GoPro continues to prosecute this family.

5. Explicit uncertainties

1. I do not have primary-source (PatentCenter/EDIS) verification of the ITC's exact claim-by-claim validity holding for the '052 patent; the "invalid" characterization comes from a respondent's press-disclosure summary, while the Commission's own notice emphasizes the no-violation finding.
2. The Federal Circuit docket 26-1752 / 2026-1752 is identified only via the patent record's litigation links and a secondary litigation database; I could not confirm its issues, briefing status, or that the '052 patent is among the appealed determinations.
3. There is a date discrepancy in my inputs: your instruction states the current date is April 26, 2026, whereas the source metadata timestamp is 2026-09-30 and one tracker lists the CAFC appeal as filed April 29, 2026. I have reported the docket as found rather than reconciling these.
4. Claim scope statements above are based on the issued claim text in the record you supplied; I did not perform an independent prosecution-history check for claim amendments or terminal disclaimers affecting the '052 patent.

I'll research litigation involving US 10529052 using the sources you specified.

Let me confirm the district court and Federal Circuit details.

Let me verify the Federal Circuit appeal details.

Litigation Involving U.S. Patent No. 10,529,052 ("the '052 patent")

Patent: US 10,529,052 B2 — "Virtual Lens Simulation for Video and Photo Cropping"; assignee GoPro, Inc.; priority 2015‑05‑20; issued 2020‑01‑07.

Based on my searches of PTAB/ITC dockets, ITC filings, litigation databases (Unified Patents, Law.com Radar, Ex Parte), USPTO PTAB records, and GoPro/Insta360 announcements, there are three U.S. proceedings and one PTAB proceeding involving the '052 patent. All trace to a single dispute: GoPro, Inc. v. Arashi Vision Inc. d/b/a Insta360 (and its U.S. subsidiary, Arashi Vision (U.S.) LLC). No other plaintiffs or defendants asserting the '052 patent were found.

1. ITC Section 337 Investigation (lead proceeding)

• Caption: In the Matter of Certain Cameras, Camera Systems, and Accessories Used Therewith
• Complainant: GoPro, Inc. (San Mateo, CA)
• Respondents: Arashi Vision Inc. d/b/a Insta360 (Shenzhen, China) and Arashi Vision (U.S.) LLC d/b/a Insta360 (Irvine, CA) (collectively "Insta360")
• Jurisdiction: U.S. International Trade Commission
• Investigation No.: 337‑TA‑1400
• Filed / Instituted: Complaint filed March 29, 2024; instituted May 6, 2024 (89 Fed. Reg. 37242)
• Patents asserted against Insta360: U.S. 10,015,413; 10,529,052 (claims 1–10); 10,574,894; 10,958,840; 11,336,832; and Design Patent D789,435
• Accused products: Insta360 "One X," "One R," "One X2," "One RS," "One X3," "Go 3," "Ace," "Ace Pro" camera systems and associated editing software/apps

Outcome / status:

• ALJ Final Initial Determination (July 11, 2025; ALJ Doris Johnson Hines): violation found only as to the D'435 design patent; no violation as to the five utility patents. As to the '052 patent, the ALJ found claims 1, 2, 5, and 6 were infringed (wide‑angle lens products only) but were invalid.
• Commission review (Sept. 11, 2025): the Commission reviewed, in part, the '052, '840, and D'435 findings — including the '052 "Element 1[g]" remapping limitation and the '052 invalidity finding over Okubo (U.S. Pub. 2006/0017817).
• Final Commission Determination (Feb. 26, 2026): violation of Section 337 only as to the D'435 design patent; no violation as to the '052 patent (and no violation as to '840). The Commission issued a Limited Exclusion Order and Cease‑and‑Desist Order directed to the design‑patent violation and terminated the investigation (91 Fed. Reg. / FR Doc. 2026‑04156, published Mar. 3, 2026). The '052 claims were additionally held invalid/partially invalid, and the Commission reversed the ALJ's finding that any accused '052 products infringed.
• Source: ITC Final ID; Fed. Reg. notice (FR‑2026‑03‑03, Investigation No. 337‑TA‑1400); GoPro/Insta360 announcements.

2. Parallel District Court Action (C.D. Cal.)

• Caption: GoPro, Inc. v. Arashi Vision Inc., d/b/a Insta360, and Arashi Vision (U.S.) LLC
• Plaintiff: GoPro, Inc.
• Defendants: Arashi Vision Inc. d/b/a Insta360; Arashi Vision (U.S.) LLC d/b/a Insta360
• Jurisdiction: U.S. District Court for the Central District of California
• Case No.: 8:24‑cv‑00681‑HDE‑JCV
• Filed: March 29, 2024 (filed same day as the ITC complaint, by Quinn Emanuel on GoPro's behalf)
• Patents asserted: the same six IP rights, including the '052 patent (complaint ¶140 alleged Insta360's reframing/zoom‑crop‑pan operations remap lens distortion to simulate consistent lens characteristics)
• Outcome / status: Stayed — on Insta360's motion under 28 U.S.C. § 1659, the court automatically stayed the action on June 3, 2024 pending the ITC investigation. With the ITC investigation terminated (Feb. 2026), GoPro has the right to seek to lift the stay and resume the case; Insta360's public announcements state the case "will resume" and that the outcome remains uncertain. (One litigation database lists the docket as "Closed," reflecting the stay posture rather than a merits resolution.) I could not confirm a post‑stay merits ruling.

3. PTAB Inter Partes Review — IPR2025‑00017

(This is technically a PTAB post‑grant proceeding, not "litigation," but it directly involves the '052 patent and is relevant to your request.)

• Caption: Arashi Vision (U.S.) LLC (d/b/a Insta360) v. GoPro, Inc.
• Petitioner: Arashi Vision / Insta360
• Patent Owner: GoPro, Inc.
• Patent: U.S. 10,529,052 (challenged claims 1–20)
• Case No.: IPR2025‑00017
• Decision: Institution DENIED — Paper 11, April 28, 2025 — the Board exercised discretion under 35 U.S.C. § 314(a) (Apple v. Fintiv) and denied institution because of the advanced parallel ITC investigation (five of six Fintiv factors favored denial; the Sotera stipulation did not cover the ITC).
• Note on identification: The Google Patents page's "Petitioner" field reads "Unified Patents PTAB Data" — this is the data‑source attribution, not the party. The actual petitioner is Arashi Vision / Insta360.

4. Federal Circuit Appeal

• Court: U.S. Court of Appeals for the Federal Circuit
• Case No.: 26‑1752
• Source: Listed in the Google Patents litigation data associated with US 10,529,052 (portal.unifiedpatents.com, CAFC case 26‑1752).
• Status / caveat: This appears to be GoPro's appeal of the ITC Final Determination (GoPro had 60 days from the effective date of the final determination to petition the Federal Circuit). I could not independently verify the docket details (parties, exact filing date, briefing status) in my searches, so treat the case number and posture as drawn from the patent's litigation metadata rather than a confirmed CAFC docket entry.

Summary Table

Notes / caveats

• Single dispute. Every U.S. proceeding involving the '052 patent arises from the GoPro–Insta360 conflict. I found no third‑party infringement suits asserting the '052 patent against any other defendant.
• Related but distinct proceedings. Insta360 also filed IPRs against other GoPro patents at issue in the same ITC case (e.g., IPR2024‑01164 as to the '840 patent), and Insta360 filed Chinese patent suits against GoPro (Jiangsu High People's Court, Shenzhen Intermediate Court, Changsha Intermediate Court). Those proceedings do not involve US 10,529,052, so I have excluded them from the table above.
• Verification limits. The ITC and PTAB records are well documented (ITC Final ID, Fed. Reg. notices, PTAB Paper 11). The C.D. Cal. docket details and the Federal Circuit appeal (26‑1752) are drawn from litigation databases and the patent's own Google Patents litigation metadata; I was unable to pull a full PACER/CAFC docket for those two in this session and flag them accordingly.
• Special note on the ITC outcome as to '052: The ALJ found the '052 claims both infringed and invalid at the initial determination; the Commission's final determination reversed the infringement finding and resolved the matter as to the '052 patent in Insta360's favor (no violation).

I'll start by searching for the proceedings flagged in the Google Patents data and any others on file.

Good — found the key IPR. Now let me check the Federal Circuit appeal and whether any other IPRs hit this patent.

Proceedings overview

One AIA trial proceeding is on file against US 10,529,052: IPR2025-00017 (Arashi Vision (U.S.) LLC d/b/a Insta360 v. GoPro, Inc.), institution DENIED 2025-04-28 — zero claims invalidated, zero claims sustained, zero FWDs, zero appeals. Note at the outset: the structured "PTAB proceedings on file" block returns no AIA trials for this patent, but the Google Patents family page independently flags "PTAB case IPR2025-00017 filed (Not Instituted - Procedural)," and web sources fully confirm that proceeding. I treat IPR2025-00017 as real and on file, and flag the ODP block as under-inclusive (most likely because a discretionary, non-instituted petition is not indexed as a "trial proceeding"). Both data sources attribute the entry to "Unified Patents" — that is the data vendor for the litigation feed, not the petitioner; the petitioner of record is Arashi Vision/Insta360, not Unified Patents.

Bottom line defensive posture: the PTAB has never adjudicated this patent. There is no cancellation, no estoppel, and no final written decision to cite. But a defendant today is not facing a hardened patent: the same three prior-art references Insta360 used (Okubo, Sokeila, Eder) remain fully available, and the ITC — on a full evidentiary record — already found claims 1, 2, 5, and 6 of the '052 patent invalid as obvious over Okubo. That is the strongest invalidity signal on this patent, and it came from a forum, not the PTAB.

IPR2025-00017 — Arashi Vision (U.S.) LLC (d/b/a Insta360) v. GoPro, Inc.

• Type: Inter Partes Review
• Filed: 2024-10-15 (petition assigned the 2025 IPR series number; docketed as IPR2025-00017)
• Status: "Not Instituted - Procedural" (structured data verbatim). Plain English: the Board denied institution on discretionary grounds — the petition was never tried on the merits and no Final Written Decision issued.
• Judge panel: Not publicly confirmed in the sources reviewed. The denial is Paper 11 (2025-04-28) per the PTAB Litigation Blog; I did not locate the three-APJ panel composition, so I do not name judges.
• Petition grounds (all § 103 obviousness, all against claims 1–20 — the entire claim set):
  • Ground 1: Claims 1–20 obvious over Okubo (U.S. Pat. App. Pub. No. 2006/0017817) — a digital camera performing two-stage lens distortion correction (correct to zero distortion, then re-introduce a desired distortion) on "image cutout ranges."
  • Ground 2: Claims 1–20 obvious over Okubo in view of Sokeila (U.S. Pat. No. 8,994,838) — Sokeila supplying adaptive, frame-by-frame, non-centric cropping (motion/face-based) to fix Okubo's fixed-position sub-frames.
  • Ground 3: Claims 1–20 obvious over Eder (U.S. Pat. No. 9,238,434) — vehicle rear-view-mirror simulation that rectifies a pixel subset and then applies a post-distortion.
  • Petitioner's expert was Dr. Aggelos Katsaggelos; Patent Owner's expert was Dr. Hassan Foroosh.
• Institution decision: DENIED — 2025-04-28 (Paper 11), exercising discretion under 35 U.S.C. § 314(a) and Apple Inc. v. Fintiv, Inc., IPR2020-00019. The Board analyzed the Fintiv factors against the co-pending ITC Investigation No. 337-TA-1400 and found five of six factors favored denial: Factor 2 (ITC completion ~7 months before the Board's FWD deadline) strongly favored denial; Factor 3 (ITC already through claim construction and a full evidentiary hearing) heavily favored denial; Factor 4 favored denial because 8 claims — including both independent claims — overlapped with the ITC; Factor 5 favored denial (same parties). Critically, the Board declined to credit Insta360's Sotera-type stipulation because it was worded to cover only the district court, not the ITC. GoPro's POPR (Paper 7, 2025-01-30) and a supplemental Fintiv brief (2025-04-03) preceded the denial. See PTAB Litigation Blog and Lexology.
• Final Written Decision: None. No FWD was ever issued and no claim was canceled by the PTAB. Any statement that the PTAB invalidated or upheld claims of the '052 patent would be false.
• Settlement / termination: No settlement. The proceeding terminated by discretionary denial of institution, not by the parties.
• Appeal: None — and none is available. Under 35 U.S.C. § 314(d), a determination whether to institute is non-appealable. The IPR2025-00017 denial was not appealed.
• Defensive value: Zero estoppel risk from this petition — § 315(e)(2) estoppel attaches only after an IPR "results in a final written decision," and there was none. That cuts both ways: Insta360 obtained no claim cancellation, but no defendant is estopped either. Practically, the denial means the Board has not blessed these claims; it only declined, for case-management reasons tied to the ITC, to review them. A defendant today can file a fresh IPR on the same Okubo/Sokeila/Eder grounds (subject to § 315(b)'s one-year bar and § 325(d) risk if the art was previously before the Office).

Related non-AIA proceedings bearing directly on the '052 patent

These are not PTAB proceedings and should never be cited as an "IPR outcome," but they drive the current defensive posture:

• ITC Inv. No. 337-TA-1400, Certain Cameras, Camera Systems, and Accessories Used Therewith (GoPro, Inc. v. Arashi Vision/Insta360). Instituted 2024-05-06, 89 FR 37242. ALJ Doris Johnson Hines. Final Initial Determination issued 2025-07-11 finding claims 1, 2, 5, and 6 of the '052 patent infringed (wide-angle lens products only), invalid, and satisfied for the technical prong. GoPro withdrew claims 3, 4, and 7–10 of the '052 patent (Order No. 9, 2024-09-30; Order No. 24, 2025-01-13). The Commission reviewed on 2025-09-11 (90 FR 44710) and, in its Final Determination of 2026-02-26, reversed the infringement finding as to '052 claim 1 element 1[g] (no infringement) and affirmed invalidity of claims 1, 2, 5, and 6 as obvious over Okubo under modified reasoning. Net: no violation of § 337 as to the '052 patent. See govinfo, FR-2026-03-03 (FR Doc. 2026-04156) and MoFo's client alert (2026-02-26).
• C.D. Cal. — GoPro, Inc. v. Arashi Vision Inc. et al., No. 8:24-cv-00681-HDE-JCV, asserted on the '413, '052, '894, '840, '832, and D789,435 patents. Automatically stayed under 28 U.S.C. § 1659 pending the ITC investigation; per the sources reviewed it remains stayed, with the C.D. Cal. case to resume now that the ITC has terminated.
• Federal Circuit — public litigation data for this family lists a Court of Appeals docket, 26-1752, on the Unified Patents litigation portal (https://portal.unifiedpatents.com/litigation/Court%20of%20Appeals%20for%20the%20Federal%20Circuit/case/26-1752). I could not verify the parties, the originating tribunal, or the issues from the sources reviewed, and I do not assert it is an appeal of anything in IPR2025-00017 (which is unappealable). Given the timing (a 2026 docket, consistent with the ITC's 2026-02-26 final determination), the most likely candidate is a § 1295(a)(6) appeal from the ITC determination — but treat that as an open item to confirm on the Federal Circuit docket and CourtListener, not as established.

Strategic summary

Claim status. By the PTAB: nothing is canceled and nothing is sustained — claims 1–20 were challenged in IPR2025-00017 but institution was denied, so all 20 claims are UNTESTED at the PTAB. By the ITC: on the Commission's 2026-02-26 final determination, claims 1, 2, 5, and 6 were held invalid as obvious over Okubo (affirming the ALJ's invalidity finding under modified reasoning) and not infringed as to element 1[g]; the Commission found no violation of § 337 as to the '052 patent. Claims 3, 4, and 7–20 are untested everywhere — GoPro affirmatively withdrew claims 3, 4, and 7–10 from the ITC, and the PTAB never reached any of 1–20. Two important caveats for a defendant: (i) an ITC invalidity finding is in rem and non-precedential as against the rest of the world — it does not legally cancel the claims, and the patent remains on the register as active with an anticipated expiration of 2036-05-17; and (ii) claims 3, 4, 7–20 have never been invalidated, so if a demand letter cites the metadata/orientation/motion-dependent claims (3, 4, 5, 6) or the method claims 11–20, you cannot point to any adjudication killing them — but you can point to the ITC record, where the same Okubo ground defeated the claims GoPro actually chose to press.

Estoppel landscape. There is no IPR estoppel on this patent. § 315(e)(2) applies only where an IPR "results in a final written decision," and IPR2025-00017 produced none. So no petitioner, privy, or real party in interest is barred from raising Okubo, Sokeila, Eder, or any other ground, in the PTAB or in district court. Conversely, Insta360's Sotera-type stipulation not to re-litigate IPR grounds in the district court — which the Board specifically found did not cover the ITC — was premised on an IPR that never happened; its continued effect in the C.D. Cal. case (and whether GoPro will seek to enforce it when that stay lifts) is an open question worth checking before you rely on it. Your available prior art is effectively unbounded, subject only to § 315(b)'s one-year bar from service of a complaint and § 325(d) discretion if you recycle art the examiner already considered.

Pattern signals. The petitioner on IPR2025-00017 was Insta360/Arashi Vision — a direct competitor and litigation opponent, not a defensive aggregator. "Unified Patents" appears only as the data source on the Google Patents litigation feed. Insta360 has been a serial IPR filer against the GoPro portfolio (multiple petitions across the '052, '840, and related assets; some reportedly still pending), and its overall record is mixed — GoPro has publicly countered that the PTAB upheld several GoPro patents, including HyperSmooth-related ones. GoPro itself litigates aggressively and in parallel (ITC + district court + multiple patent families), and this patent sits in a family of at least 11 co-pending U.S. applications (the '012, '338, '051, '115, '323, '777, '034, '184, and US20250045863A1 line), all sharing the 2015-05-20 priority date and all titled "Virtual lens simulation for video and photo cropping." Assume a defendant facing the '052 will also face siblings.

Recommended next steps

1. If you are a defendant and the assertion cites claims 1, 2, 5, or 6 — the strongest thing you have is the ITC Commission Final Determination of 2026-02-26, which affirmed invalidity of exactly claims 1, 2, 5, and 6 as obvious over Okubo. Cite the Commission's disposition: "affirm the Final ID's finding that claims 1, 2, 5, and 6 of the '052 patent are invalid as obvious under modified reasoning" and "reverse the Final ID's finding that the accused wide-angle lens products infringe 1, 2, 5, and 6 of the '052 patent because those products do not satisfy the limitation 'the output images include the sub-frames remapped from the input lens distortion' of element 1[g]." Full text: govinfo FR Doc. 2026-04156. Pair it with a § 282 invalidity defense and a fresh IPR on Okubo — and remember the ITC finding is persuasive, not preclusive.
2. Do not cite IPR2025-00017 as an invalidation. It is a denial of institution (Paper 11, 2025-04-28) with no FWD, no claim-level holding, and no appealable judgment. Representing otherwise to a court would be a sanction risk. Pull the paper from PTAB E2E (https://ptab.uspto.gov) to confirm the panel and the verbatim Fintiv reasoning before filing.
3. Confirm the Federal Circuit docket 26-1752 on the CAFC docket/CourtListener. If it is GoPro's appeal of the ITC's no-violation determination on the '052, the validity and non-infringement rulings described above are on appeal and could be reversed; that materially changes the weight of point 1. Treat the ITC outcome as not final until the appeal resolves.
4. Check the § 315(b) clock before filing any IPR. If you were served with a complaint asserting the '052 more than one year ago, your IPR window is closed. The C.D. Cal. case (No. 8:24-cv-00681) was stayed under § 1659; when that stay lifts, the § 315(b) clock restarts/is triggered for defendants served in that action — confirm your own service date rather than assuming.
5. Map the claim set, not just the independent claims. Claims 3, 4, 7–20 have never been tested by any tribunal. If the assertion reaches those claims, your IPR must target them specifically, and the ITC Okubo finding gives you no direct cover — though GoPro's withdrawal of claims 3, 4, and 7–10 from the ITC is itself a useful admission to raise.
6. Watch the sibling patents. With 11 family members sharing the 2015-05-20 priority date, expect continuation-level variants with comparable claims. Any clearance or invalidity opinion that addresses only the '052 is incomplete.

I'll research the assignment record for US 10529052, starting with the USPTO Assignment Center and cross-referencing public sources.

Let me dig for reel/frame specifics and the banking/lender assignment records.

Found reel/frame data patterns. Let me pin down the exact reel/frame entries for US 10529052 itself.

Ownership & Assignment Reconstruction — US 10529052 B2

Patent: Virtual lens simulation for video and photo cropping
App. No.: 16/535,940 · Filed: 2019-08-08 · Issued: 2020-01-07 · Priority: 2015-05-20 (Prov. 62/164,409)
Family: continuation of US 16/229,512 (US 10395338) ← US 15/157,207 (US 10186012)
Verification links: USPTO Assignment Center · Assignment Search (legacy) · Google Patents legal events

Sourcing caveat: I was able to confirm the events in the chain from Google Patents legal events, SEC/credit-agreement documents, and litigation filings, but I could not open the Assignment Center record page itself for the '052 patent, so I could not read every reel/frame directly off the Abstract of Title. Where a reel/frame is shown below, I state its provenance. Where I could not confirm one, I say so rather than infer it.

Inventors

Pattern note — no adverse signal. All four are GoPro engineers/employees, and all four appear as assignors on the recorded "ASSIGNMENT OF ASSIGNORS INTEREST" to GoPro (Google Patents legal event, recorded 2019-08-08; assignors "SIEVERT, OTTO KENNETH; MACMILLAN, TIMOTHY; BODINET, JOSHUA EDWARD; NEWMAN, DAVID A."). I found no evidence of inventor departure, no separate inventor-held assignment, and no inventor-owned entity in the chain. Caution: the 2019-08-08 recording date coincides with the continuation filing date, so it likely reflects a confirmatory inventors-to-GoPro assignment executed for the continuation; I could not retrieve the execution date.

Original assignee

GoPro, Inc. (Delaware corporation), 3000/3025 Clearview Way, San Mateo, CA 94402.

• Line of business: consumer action cameras and editing software (HERO, MAX lines; Quik / FX Reframe / GoPro Studio).
• Ships a product embodying the claims? Yes. GoPro's own complaint in GoPro, Inc. v. Arashi Vision Inc. (Insta360), C.D. Cal. 8:24-cv-00681, ¶¶140–145, alleges the '052 invention is practiced by GoPro's own "FX Reframe" plugin and by in-camera reframing/remap features, and states GoPro marks products practicing the '052 patent on its virtual marking site. It further alleges Insta360's accused editing apps "utilize virtual lens technology to remap input lens distortion in sub-frames" — i.e., GoPro asserts the patent against a direct competitor.
• Current status: Operating, publicly traded (NASDAQ: GPRO), but financially distressed. SEC filings show an asset-coverage covenant waiver dated 2026-05-08, a fourth amendment to the Wells Fargo revolver and a third amendment to the Farallon loan (July 2026), and a new amortization schedule beginning with $250k weekly payments from 2026-10-09. GoPro remains the owner of record; the IP is encumbered by liens, not transferred.

Assignment timeline

The chain contains no title transfers out of GoPro. Every post-issuance recording is a secured-party interest or its release. Chronological:

• 2016-03-25 (executed) / recorded 2016-03-28 — Reel 5759/0044

  • Conveyance: Security Interest (U.S. Patent and Trademark Security Agreement under the 2016 Credit Agreement)
  • Assignor: GoPro, Inc.
  • Assignee: JPMorgan Chase Bank, N.A., as Administrative Agent
  • Correspondent: not confirmed for the patent record. The contemporaneous GoPro trademark security-interest recording at reel 6348/0594 names correspondent Jay DaSilva, Cogency Global Inc., 1025 Vermont Ave NW, Suite 1130, Washington DC 20005 — a corporate-services filing shop, typical for secured-party recordings. Do not treat this as a verified patent-record correspondent.
  • Context: Securitization — collateral grant under the March 2016 credit facility. (Reel shown is the trademark record; the parallel patent security agreement covers GoPro patent schedules and should be verified against the '052 entry.)

• 2020-10-19 (recorded) — Reel/frame not confirmed

  • Conveyance: Security Interest (see document for details)
  • Assignor: GoPro, Inc.
  • Assignee: JPMorgan Chase Bank, N.A., as Administrative Agent
  • Correspondent: not determinable from available records
  • Context: Securitization — continuation/refresh of the JPMorgan collateral position covering newly issued GoPro patents.

• 2021-01-22 (executed) / recorded 2021-01-25 — Reel 7168/0156 (adjacent release of trademark security interest at 7168/0919)

  • Conveyance: Release of Patent Security Interest
  • Assignor: JPMorgan Chase Bank, N.A., as Administrative Agent
  • Assignee: GoPro, Inc.
  • Correspondent: not confirmed for the patent record
  • Context: Securitization unwind — JPMorgan lien released; title back to GoPro unencumbered. (Reel 7168/0156 is the "RELEASE BY SECURED PARTY" entry in the GoPro trademark Abstract of Title, exec. 2021-01-22; confirm the patent-side reel.)

• 2025-08-04 (recorded) — Reel 072358/0001 (reel read from a sibling GoPro patent's assignment data; confirm for '052)

  • Conveyance: Security Interest (Amended and Restated Guaranty and Security Agreement — first-priority lien on all GoPro IP registrations and applications)
  • Assignor: GoPro, Inc.
  • Assignee: Wells Fargo Bank, National Association, as Agent
  • Correspondent: not confirmed
  • Context: Securitization / refinancing — Wells Fargo given first-priority IP lien in connection with the Farallon second-lien deal.

• 2025-08-04 (recorded) — Reel 072340/0676 (reel read from a sibling GoPro patent's assignment data; confirm for '052)

  • Conveyance: Security Interest (Second Lien Credit Agreement, second-priority lien on substantially all assets including IP)
  • Assignor: GoPro, Inc.
  • Assignee: Farallon Capital Management, L.L.C., as Agent
  • Correspondent: not confirmed
  • Context: Securitization / distressed financing — $50M second-lien term loan at SOFR+7.50%, plus ~11.1M warrants at $1.25, used to repay the Nov-2025 converts.

There is no recorded assignment to any entity with an "IP / Holdings / Licensing / Ventures" suffix, and no assignment of title away from GoPro at any point.

Timeline diagram

timeline
    title Ownership of US 10529052
    2015 : Priority provisional filed
    2016 : Original application filed
         : Inventors assign rights to GoPro
         : JPMorgan records IP security interest
    2019 : Continuation application filed
    2020 : Patent issued to GoPro
         : JPMorgan records security interest
    2021 : JPMorgan lien released
    2024 : GoPro sues Insta360 for infringement
    2025 : Wells Fargo records first lien
         : Farallon records second lien
    2026 : ITC finds the claims invalid

NPE / troll-pattern signals

1. Shell-entity transfer — Not present. Every assignee other than GoPro is a bank or fund acting as collateral/administrative agent (JPMorgan Chase Bank N.A.; Wells Fargo Bank N.A.; Farallon Capital Management L.L.C.), each expressly "as Agent" and each under a named credit agreement documented in GoPro SEC filings. No LLC with an IP/Holdings/Licensing suffix ever appears as assignee. No registered-agent service address appears as an assignee address.

2. Known asserter in the chain — Not present. Neither GoPro nor JPMorgan/Wells Fargo/Farallon appears on the Acacia / Marathon / IV / IPNav / Wi-LAN / Conversant / Vringo / Pendrell / Round Rock lists. The one litigation in the family is GoPro as plaintiff, not an NPE — the family's PTAB/litigation docket (IPR2025-00017, ITC 337-TA-1400, CAFC 26-1752, C.D. Cal. 8:24-cv-00681) shows GoPro asserting its own patent against a competitor.

3. Repeat correspondent across the chain — Unclear. I could not read the correspondent field on the patent-side records. The only GoPro correspondent I could verify is Jay DaSilva, Cogency Global Inc., Washington DC, on the parallel trademark security-interest recording (reel 6348/0594, rec. 2018-06-07). Cogency Global is a commercial registered-agent/corporate-services filer — expected for lender-driven collateral recordings and not an NPE-associated filing attorney. I found no single recurring attorney name spanning multiple links, and I did not find GoPro's counsel on any NPE assertion list. (Note: a "Dusan Clark, Sidley Austin LLP" security-interest correspondent surfaced in my search, but that record belongs to Gogo LLC — a different company — and is not part of this chain.) No finding without verified recurrence.

4. Cascading transfers — Not present. Zero title transfers in the chain. The only clustered recordings are the two 2025-08-04 security interests, which are the two halves of a single, publicly announced financing (first lien to Wells Fargo + second lien to Farallon), not a chained LLC structure.

5. Pre-litigation transfer — Not present. No assignment to the plaintiff within 6 months of the March/June 2024 suits. GoPro has held title continuously since at least 2016 and pleads ownership "by assignment" in ¶136 of the C.D. Cal. complaint as a routine standing recital.

6. Bankruptcy fire-sale — Not present, but watch. GoPro has not filed Chapter 7/11, and no IP has been sold in proceedings. Distress precursors are observable: the 2025-08-04 second-lien grant over "substantially all assets including intellectual property," the 2026-05-08 asset-coverage waiver, lender-driven amendments, and the 180-day covenant requiring GoPro to refinance or repay the Wells Fargo facility. If GoPro defaults, the lenders' first/second liens on the '052 patent could convert into foreclosure — the mechanism by which such portfolios have historically moved.

7. Privateering — Not present. GoPro litigates in its own name (ITC and C.D. Cal.), does not appear to have transferred the patent to an NPE for assertion, and no Patent Progress/EFF coverage links this patent to an on-behalf-of assertion vehicle.

8. Defensive aggregator (anti-NPE) — Not present. The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. (Unified Patents is on the other side of this story: it filed the IPR2025-00017 petition, filed 2024-10-15 by Arashi Vision/Insta360 and not instituted on 2025-04-28 — i.e., a defensive challenge, not ownership.)

Verdict

Operating-company assertion.

GoPro, Inc. has held title continuously — the only non-GoPro recorded assignees are JPMorgan Chase (reel 5759/0044, 2016; recorded again 2020-10-19 and released 2021-01-25), Wells Fargo (reel 072358/0001, 2025-08-04) and Farallon (reel 072340/0676, 2025-08-04), each solely as secured collateral agent. GoPro ships products it says practice the claims and asserted US 10529052 against a direct competitor (Insta360/Arashi Vision) in ITC 337-TA-1400 and C.D. Cal. 8:24-cv-00681 — textbook operating-company enforcement, not NPE behavior. The real risk flag on this chain is financial distress, not troll structure: the 2025 liens and 2026 covenant waivers create a path to lender foreclosure if GoPro defaults.

Verify at: https://assignmentcenter.uspto.gov/ (search "10529052") — and please note that the reel/frame numbers marked not confirmed above should be read off the Abstract of Title before this memo is used in a formal opinion.

I'll research US Patent 10529052 and its prior-art citations. Let me start with searches.

Let me get the full front-page citation list and details on key references.

Let me pull the IPR petition art and the full citation list.

Prior-Art Analysis — US 10,529,052 B2

Methodological note first: I was able to retrieve the target patent's biblio data and a substantial portion of its cited references, but the Google Patents "Citations (169)" table is truncated in retrieval, and my search budget for this session ran out before I could open every individual reference. Below I give (a) verified bibliographic data, (b) the prior-art citations I could positively retrieve, and (c) an honest statement about what I could not verify. I do not assert anticipation for any reference I have not read claim-by-claim.

1. Target patent — literal identification

⚠️ Literal-identifier caution: searching the bare string "10529052" also surfaces an unrelated item in which 10/529,052 is an application serial number (US 2005/0255263 A1, "Vinylidene chloride polymer compositions and food casings," Lee et al., filed 2003-10-14). That is a different document and is not prior art to the GoPro patent. I did not auto-correct either identifier.

Claim structure (20 claims): independent claim 1 (system) and independent claim 11 (method) are substantively parallel, as are dependent pairs 2/12 (image stabilization), 3/13 (metadata-based sub-frame selection), 4/14 (motion metadata), 5/15 (orientation metadata), 6/16 (output video), 7/17 (frame-to-frame consistent lens characteristics), 8/18 (simulated virtual re-pointing), 9/19 (processing in-camera), 10/20 (wide-angle lens; straight lines curved near edges).

Critical framing for §102: the operative independent-claim elements are (i) input images bearing an input lens distortion centered in the input FOV; (ii) sub-frames with reduced FOV whose distortion varies as a function of both position and size; and (iii) remapping that distortion to a desired distortion re-centered on the sub-frame. Art that merely de-warps (rectilinearizes) does not meet this; art that merely crops does not meet this.

2. Prior art cited on the face of US 10,529,052

Important: the examiner's citation of a reference is not a finding that it anticipates. US 10,529,052 issued with all 20 claims intact after these 169 citations were considered, which means the examiner did not find any of them to be anticipatory as applied. References below marked * were flagged by the examiner (Google Patents "cited by examiner" asterisk) and were therefore substantively considered, not merely listed.

Tier 1 — the references most probative of claims 1/11

US 2002/0044153 A1 * — Hoch, Michael. "Method and apparatus for optimized distortion correction for add-on graphics for real time video." Pub. 2002-04-18 (earliest priority 2000-08-29). IPC H04N 5/222.
Description: real-time distortion correction applied to video so that graphics overlay correctly onto a distorted (e.g., wide-angle) image stream. Potentially bears on: claims 1, 11 (distortion is characterized and re-mapped) — but the reference corrects toward rectilinear, not toward a re-centered desired lens distortion in a reduced-FOV sub-frame, so anticipation appears incomplete.

US 2007/0229665 A1 * — Tobiason, Joseph D. "Robust field of view distortion calibration." Pub. 2007-10-04 (filed 2006-03-31).
Description: calibrating the distortion of a field of view (fisheye/wide-angle optics) to produce a distortion model. Potentially bears on: claims 1, 3, 11, 13 (the step of determining the input lens distortion / using stored lens characteristics). No teaching of selecting sub-frames or re-centering the distortion.

WO 2009/146319 A1 — Adobe Systems Inc. "Method and apparatus for rendering images with and without radially symmetric distortions." Pub. 2009-12-03 (filed 2008-05-28).
Description: rendering an image both with and without a radially symmetric distortion, i.e., controlled add/remove of a radially symmetric (lens-like) distortion. Potentially bears on: claims 1, 11 — it is the citation closest in kind to the "apply a desired lens distortion" half of the remap. Still lacks the reduced-FOV sub-frame selection and distortion-centered-on-the-sub-frame limitation.

US 6,323,934 B1 * — Enomoto et al., Fuji Photo Film Co. "Image processing method and apparatus." Pub. 2001-11-27 (priority 1997-12-04). Classified G06T 5/006. Potentially bears on: claims 1, 11 (computational image transformation to alter lens-induced geometry).

US 9,105,080 B2 * — McAndrew, A. (8/2015). Classified G06T 7/80 (camera calibration). Title/abstract not retrieved in this session — cannot responsibly characterize. Listed here only because the examiner flagged it; it is a plausible calibration/distortion-model reference.

US 9,262,807 B2 * — Stec, P. (2/2016). Classified G06T 5/006 (geometric correction). Title/abstract not retrieved — same caveat as above.

Tier 2 — swept-up art on which the claims were most likely allowed (panoramic/virtual-camera and auto-selection)

• US 6,337,683 B1 — Gilbert et al., Imove Inc. "Panoramic movies which simulate movement through multidimensional space." Pub. 2002-01-08 (filed 1998-05-13). Bears on: claims 6, 8, 16, 18 (simulated camera movement/virtual re-pointing) and, by extension, 1/11.
• US 5,130,794 A — Ritchey, K. "Panoramic display system." Pub. 1992-07-14. Bears on: claims 6, 8, 16, 18.
• US 2010/0045773 A1 — Ritchey, K. "Panoramic adapter system and method with spherical field-of-view coverage." Pub. 2010-02-25. Bears on: claims 10, 20 (spherical/wide-angle FOV).
• US 6,593,956 B1 — Pomerantz et al., Polycom Inc. "Locating an audio source." Pub. 2003-07-15. Bears on: claims 3, 13 (automatic sub-frame selection — the spec expressly contemplates microphone-array/audio-source-based selection).
• US 2008/0208791 A1 — Das, M. et al. "Retrieving images based on an example image." Pub. 2008-08-28. Bears on: claims 3, 13.
• US 2004/0128317 A1 — Sull, Sanghoon. "Methods and apparatuses for viewing, browsing, navigating and bookmarking videos and displaying images." Pub. 2004-07-01. Bears on: claims 3, 13, 6, 16.
• US 7,483,618 B1 — Edwards et al., YesVideo, Inc. "Automatic editing of a visual recording to eliminate content of unacceptably low quality and/or very little or no interest." Pub. 2009-01-27. Bears on: claims 3, 13, 6, 16.
• US 2005/0025454 A1 — Nakamura, Nobuo. "Editing system and control method thereof." Pub. 2005-02-03. Bears on: claims 6, 16.
• US 2010/0183280 A1 — Muvee Technologies. "Creating a new video production by intercutting between multiple video clips." Pub. 2010-07-22. Bears on: claims 6, 16.
• US 2010/0231730 A1 — Ichikawa, Yuka. "Image sensing device and camera." Pub. 2010-09-16; US 2010/0245626 A1 — Woycechowsky, D.B. "Digital Camera," pub. 2010-09-30. Bear on: claims 9, 19 (processing within the camera).

Tier 3 — the remaining cited U.S. patents/publications (retrieved, with dates)

US 7,222,356 B1 (Canon, 2007-05-22); US 8,446,433 B1 (Mallet, 2013-05-07); US 8,611,422 B1 (Yagnik, 2013-12-17); US 8,718,447 B2 (Yang, 2014-05-06); US 8,730,299 B1 (Kozko, 2014-05-20); US 8,763,023 B1 / shown in one source as 8,763,032 B1 (Goetz, 2014-06-17 — OCR-ambiguous); US 8,910,046 B2 (Matsuda, 2014-12-09); US 8,988,509 B1 (Macmillan, 2015-03-24); US 9,032,929 B2 / shown elsewhere as 9,032,299 B2 (Lyons, 2015-05-12 — OCR-ambiguous); US 9,036,001 B2 (Chuang, 2015-05-19); US 9,077,956 B1 (Morgan, 2015-07-07); US 9,111,579 B2 (Meaney, 2015-08-18); US 9,142,253 B2 (Ubillos, 2015-09-22); US 9,151,933 B2 (Sato, 2015-10-06); US 9,204,029 / 9,204,039 B2 (He, 2015-12-01 — OCR-ambiguous); US 9,208,821 B2 (Evans, 2015-12-08); US 9,245,582 B2 (Shore, 2016-01-26); US 9,253,333 / 9,255,333 B1 (Morgan, 2016-02-02 — OCR-ambiguous); US 9,317,172 B2 (Lyons, 2016-04-19); US 9,423,944 B2 (Eppolito, 2016-08-23); US 9,473,758 B1 (Long, 2016-10-25); US 9,479,697 B2 (Aguilar, 2016-10-25); US 9,564,173 B2 (Swenson, 2017-02-07); US 10,156,012 B2 (Newman, 2019-01-22 — same family, not prior art).

Publications: US 2002/0044153 (Hoch); US 2004/0128317 (Sull); US 2005/0025454 (Nakamura); US 2006/0122842 (Herberger, Magix, 2006-06-08); US 2007/0173296 (Canon, 2007-07-26); US 2007/0204310 (Microsoft, 2007-08-30); US 2007/0230461 (Samsung, 2007-10-04); US 2007/0229665 (Tobiason); US 2008/0044155 (Kuspa, 2008-02-21); US 2008/0123976 (Reuters, 2008-05-29); US 2008/0152297 (Apple, 2008-06-26); US 2008/0163283 (Tan, 2008-07-03); US 2008/0177706 (Yuen, 2008-07-24); US 2008/0208791 (Das); US 2008/0253735 (Adobe, 2008-10-16); US 2008/0313541 (Yahoo!, 2008-12-18); US 2009/0213270 (Ismert, 2009-08-27); US 2009/0274339 A9 (Cohen, 2009-11-05); US 2009/0327856 (Mouilleseaux, 2009-12-31); US 2010/0045773 (Ritchey); US 2010/0064219 (Gabrisko, 2010-03-11); US 2010/0086216 (Samsung, 2010-04-08); US 2010/0104261 (Liu, 2010-04-29); US 2010/0183280 (Muvee); US 2010/0231730 (Ichikawa); US 2010/0245626 (Woycechowsky); US 2010/0251295 (AT&T, 2010-09-30); US 2010/0278509 (Nagano, 2010-11-04); US 2010/0281375 (Pendergast, Adobe, 2010-11-04).

Foreign: WO 2001/020466 A1 (Hotv Inc., 2001-03-22); WO 2009/040538 A1 (British Telecommunications, 2009-04-02); WO 2009/146319 A1 (Adobe, 2009-12-03).

Non-patent literature cited: Ernoult, Emeric, "How to Triple Your YouTube Video Views with Facebook," SocialMediaExaminer.com, Nov. 26, 2012, 16 pages — marketing, not technical; no §102 relevance to claims 1/11.

3. The genuinely most relevant prior art — from the IPR challenge, not the examiner's list

US 10,529,052 has been challenged. Arashi Vision Inc. (d/b/a Insta360) v. GoPro, Inc., IPR2025-00017, petition filed 2024-10-15; the Board denied institution on 2025-04-28 (35 U.S.C. §314; disposed on procedural/Fintiv-type grounds, not on the merits).

• https://portal.unifiedpatents.com/[ptab](/ptab)/case/IPR2025-00017
• https://ipverse.greyb.com/ptab-web/cases/case-details/IPR2025-00017

The petition's technical exhibits identify the challenger's chosen art (Ex. 1004–1012):

Because institution was denied, no claim of US 10,529,052 has been held unpatentable over this art, and there is no PTAB merits record I can cite for §102 element-by-element mapping. That denial is also not a merits adjudication that the art fails.

Related proceedings (per Google Patents family litigation data): ITC Inv. No. 337-TA-1400; Fed. Cir. No. 26-1752; C.D. Cal. 8:24-cv-00681.

4. Direct answers to the four requested fields

1. Full citations / dates — given in §2 above for the ~70 references I could retrieve, and §3 for the IPR art. The remaining ~100 entries of the 169-item citation table were not retrievable in this session; I will not invent them.
2. Which claims they potentially anticipate — mapped per reference above (claims 1/11 for distortion-remapping art; 3/13–5/15 for metadata/auto-selection art; 6/16–8/18 for panoramic/virtual-camera art; 9/19 for in-camera art; 10/20 for wide-angle/FOV art). These are relevance assessments keyed to titles/abstracts/classifications.
3. Anticipation under §102 — bottom line: I found no reference on the face of US 10,529,052 that discloses every element of claim 1 or claim 11, and it would be inconsistent with the grant of all 20 claims to assert one. The closest single-reference candidates are US 2002/0044153 A1 (Hoch) and WO 2009/146319 A1 (Adobe), each of which supplies the distortion add/remove concept but not the reduced-FOV sub-frame selection with distortion re-centered on the sub-frame, nor the position-and-size-dependent distortion characteristic. Any serious §102 case would more likely be built on the IPR-style combination (Okubo / Sokeila / Eder), which I could not verify element-by-element.
4. Not prior art: US 10,186,012 (15/157,207); US 10,395,338 (16/229,512); US 10,529,051; US 10,535,115; US 10,679,323; US 10,817,977; US 11,164,282; US 11,688,034; US 12,243,184; US 2025/0045863 — all share the 2015-05-20 priority and are continuations of the same disclosure, not §102 art against each other. Likewise the Google Patents "Cited By" entry US 2020/0118240 A1 (GoPro) is a family member.

5. Explicit limitations on this analysis

• I could not complete: full text of all 169 cited references; the IPR2025-00017 petition's claim charts; and bibliographic verification of Okubo US 2006/0017817 A1, Sokeila US 8,994,838 B2, and Eder US 9,238,434 B2. I have flagged these rather than guessing.
• Google Patents renders several cited patent numbers inconsistently between its HTML and the patent PDF (noted inline as OCR-ambiguous). Per instruction, I did not silently "correct" them; the authoritative numbers should be confirmed against the granted front page at https://patents.google.com/patent/US10529052/en or the PTAB Exhibit 1001 at https://www.docketalarm.com/cases/PTAB/IPR2025-00017/.
• This is a technical prior-art survey, not a legal opinion; no §102/§103 conclusion should be drawn without reading each reference in full against the issued claim language.

I'll research the prior-art references cited on the '052 patent page and the ones used in the PTAB/ITC challenges, then build the §103 analysis.

Now let me gather details on the other cited prior art from the patent's reference list and the ITC/PTAB findings.

Obviousness Analysis under 35 U.S.C. § 103 — US 10,529,052 B2

Sourcing note. This analysis is built on (a) the authoritative full-text/citation record for US 10,529,052 supplied in the prior section (fetched from https://patents.google.com/patent/US10529052/en), (b) the actual prior-art grounds litigated in IPR2025-00017 (Okubo, Sokeila, Eder) and in the parallel ITC 337-TA-1400 record, and (c) reference texts surfaced by web search for those three references. It builds on, and does not repeat, the previously generated summary and litigation sections.

Two limitations on completeness I want on the record before the analysis:

1. The "Citations (169)" list on the Google Patents page was truncated in the supplied text (only the first ~40 references are visible). My secondary-reference combinations therefore rely on the visible portion of that list plus the three references actually asserted by Insta360. Findings keyed to references beyond the visible portion cannot be made.
2. The "Families Citing this family" list (EP3345184A1, US10506221B2, JP2020521348A, US12536627B2, etc.) is post-priority art by definition — those documents cite the '052 family, and their 2015–2022 priority dates fall on or after May 20, 2015. They cannot be § 102/§ 103 prior art against the '052 claims (apart from any earlier-published parent content). I have excluded them from the obviousness grounds and use them only as evidence of what the field was doing.
3. The date discrepancy flagged in the prior section persists (instruction says April 26, 2026; source metadata timestamp is 2026-09-30). Nothing below turns on it.

I. Governing framework and priority date

• The '052 claims are post-AIA (earliest priority May 20, 2015; provisional 62/164,409), so § 102(a)(1)/(a)(2) and § 103 apply as amended.
• All three primary references pre-date that date:
  • Okubo, US 2006/0017817 A1 ("Image pick-up apparatus and image restoration method") — published Jan. 26, 2006 → § 102(a)(1).
  • Sokeila, US 8,994,838 B2 ("Motion adaptive cropping for video stabilization"; Sokeila & Gren) — filed Apr. 16, 2013, issued Mar. 31, 2015 → § 102(a)(1).
  • Eder, US 9,238,434 B2 ("Rear view mirror simulation"; SMR Patents/Audi) — filed Apr. 19, 2011, published as US 2011/0254957 A1 Oct. 20, 2011 → § 102(a)(1).
• The Graham/KSR framework controls: scope and content of the prior art, differences from the claims, level of ordinary skill, and secondary considerations. Under KSR Int'l v. Teleflex, a combination is obvious where the improvement is a "predictable use of prior art elements according to their established functions," where a "known technique" is applied to improve a similar device in the same way, or where there is a design incentive/market pressure and a finite number of identified, predictable solutions.

Authenticity point: per the IPR record, none of Okubo, Sokeila or Eder was considered during prosecution of the '052 family (the examiner's cited art is the 169-reference list, of which the visible portion is largely video-editing, panoramic-display, and content-indexing art). That matters because it neutralizes any argument that the examiner already weighed and rejected these teachings.

II. Level of ordinary skill (inferred)

I could not retrieve a verbatim POSITA definition from the primary pleadings in this session, so the following is inferred from the art and the parties' framing: a POSITA would hold at least a bachelor's degree in electrical engineering, computer science, or equivalent, plus roughly 2–5 years of experience in digital image/video processing, computational photography, or camera systems, or equivalent work experience. The disputed subject matter (lens-distortion modeling, crop/mosaic stabilization, pixel remapping) was mature by 2015 — a point GoPro itself concedes in the '052 specification, which describes wide-angle edge distortion and rectilinear/fisheye projections as known phenomena.

III. Claim 1 (and method claim 11) — limitation mapping

Critical claim-construction observation. The issued claims recite only a "remap" from input distortion to desired distortion. They do not require a direct single-function mapping, and they do not exclude an intermediate rectilinear (zero-distortion) stage. The specification itself discloses the two-step inverse-then-apply alternative ("In an alternative embodiment, a two-step transformation may be used instead of a direct mapping…"). This is decisive: no limitation in claims 1 or 11 distinguishes Okubo's two-stage "distortion correcting 1 → distortion correcting 2" sequence, and GoPro's principal non-obviousness theme in IPR2025-00017 — that the '052's direct mapping distinguishes the prior art — does not map onto the claim language as issued. Likewise, "a size of the areas" is not "variable size"; the indefinite article does not require the sub-frames to differ in size, and the '052's own Fig. 1 (last two rows, argued by Insta360) shows same-sized sub-frames producing different distortion patterns by virtue of position alone.

IV. Grounds of rejection

Ground A — Okubo alone renders claims 1–10 and 11–20 obvious

Okubo discloses a camera that (i) captures a full-frame image with centered lens distortion, (ii) selects a reduced "image cutout range," (iii) corrects that cutout's distortion to zero, and (iv) re-applies a controlled barrel distortion so that the resulting image "is restored without any sense of incongruity." That is every functional step of claim 1. Dependent-claim overlays:

• Claims 2/12 (image stabilization): Okubo's entire purpose is vibration restoration — the cutout is shifted among positions to counteract camera motion. Expressly disclosed.
• Claims 3/13, 4/14 (selection based on metadata; metadata indicates motion): Okubo's angular velocity sensors 108/109 generate a vibration locus used to compute the restorative function (Fig. 12, steps S304–S306); mode key 7 selects still vs. moving-image processing.
• Claims 5/15 (orientation): Okubo derives a "vibration locus" from angular-velocity signals — orientation-type metadata.
• Claims 6/16 (output images as video frames): Okubo's fourth embodiment expressly addresses "image deteriorations by vibrations between frames in moving images"; the moving-image mode follows a distinct processing path (Fig. 7/S106 → S120).
• Claims 7/17 (consistent frame-to-frame characteristics): This is Okubo's stated design goal — outputs "without any sense of incongruity," with the same re-applied distortion level applied across frames.
• Claims 8/18 (positions simulate virtual re-pointing): Okubo's shifted cutout windows are a virtual re-pointing of the view within the larger captured field of view.
• Claims 9/19 (processing in the camera): Okubo is an in-camera implementation.
• Claims 10/20 (wide-angle lens; straight lines appear curved near edges): Okubo addresses lens distortion generally; Eder and the '052's own admitted background supply the wide-angle/curvature expression.

Honest caveat: GoPro's Preliminary Response (Paper 7, Jan. 30, 2025) argues the two-stage distortion correction is confined to still-image mode and that in moving-image mode Okubo's cutout is fixed at 70%. That is a genuine factual dispute. It is the pivot on which Ground A lives or dies, and it is the same dispute on which the ITC's Final Initial Determination apparently resolved in Insta360's favor (the ID's analysis concludes that "the evidence supports that element 1[e] is disclosed in Okubo" and that "consideration of indicia of non-obviousness do[es] not alter this outcome," while rejecting only the Eder-based ground).

Ground B — Okubo in view of Sokeila

Adds Sokeila where a decision-maker credits GoPro's argument that Okubo's sub-frame selection is too rigid (fixed 70%, mode-driven).

What Sokeila supplies: an adaptive cropping factor computed frame-by-frame, adjusted based on motion and/or detection of an object of interest (e.g., a face), with non-centric cropping to avoid cutting out the subject — precisely the "different positions … and a size of the areas" aspect of element 1[e].

Motivation to combine (KSR rationales):

1. Common problem, analogous art: both are video-stabilization/cropping systems with overlapping components (camera, sensor, image processor, motion sensors). Both are in the same field of endeavor.
2. Known technique improving a similar device in the same way: replacing a fixed-ratio crop with an adaptive, content-aware crop is a recognized improvement that Sokeila expressly frames as minimizing unnecessary cropping to optimize image quality — a result a POSITA would predict.
3. Design incentive / finite solutions: the only ways to avoid cropping subjects of interest out of a stabilized crop are to move and/or resize the crop window — a finite, predictable design space.
4. Reasonable expectation of success: integrating Sokeila's crop-selection logic into Okubo's existing image-processing pipeline is a straightforward substitution of one crop-selection stage for another; no new hardware or untested principle is required.

Note the PTAB never reached the merits — institution was denied April 28, 2025 on Fintiv § 314(a) discretion (the same Google Patents page labels it "Not Instituted – Procedural"). So Ground B has been briefed but not adjudicated by the Board.

Ground C — Okubo + Sokeila + Eder

Eder is best used not as an independent anticipation reference (where GoPro's rebuttal is strong) but as corroborating evidence of the "desired lens distortion" step and of virtual re-pointing.

What Eder supplies: a wide-angle automotive camera whose continuous video is (i) rectified to a "low-error" distortion-free image, then (ii) re-distorted ("post-distortion") so the display reproduces the exact look of the desired mirror glass (plane, aspheric, or convex); and (iii) whose selected pixel region ("area A") is moved across the sensor to adapt the field of view to the driver's position — with the size of the section unchanged. Item (iii) is direct support for dependent claims 8/18 (virtual re-pointing) and reinforces that the claim does not require variable-sized sub-frames.

Motivation to combine: Eder articulates the very commercial/ergonomic rationale for re-applying a lens distortion after correction — a viewer's familiarity and perceptual consistency with the native optical presentation. A POSITA seeking to make post-processed crops look natively captured (the '052's stated goal) would look directly to that teaching. Alternatively, Eder alone plus the knowledge of a POSITA is a viable § 103 combination for claims 1–3, 5–6, 11–13, 15–16 (this was Insta360's Ground 3, briefed at Insta360 Br. 76–83).

Why Ground C is weaker than A/B: the ITC Final ID concluded that "Insta360 has not made a clear and convincing case of invalidity" on Eder, reasoning that Eder's "area A" is only ever de-distorted to simulate a plane mirror and used to define a display region — the "post-distortion" of Eder runs on the full wide-angle image to simulate a convex/aspheric mirror, not on a sub-frame. That is a fair reading of Eder 4:60–5:12 and Fig. 8. Use Eder as secondary teaching, not as the primary reference. (Google Patents' "Cited By" list also shows a third-party citation of Eder-family art to GoPro's own later US 20200118240 A1, indicating the family itself tracked this space.)

Ground D — Okubo + WO 2009/146319 A1 (Adobe) and/or US 2007/0229665 A1 (Tobiason)

These are on the face of the '052 patent (visible in the citation list) and are pointed directly at the distortion-removal-and-re-application concept:

• WO 2009/146319 A1 (Adobe Systems), "Method and apparatus for rendering images with and without radially symmetric distortions." By its title and the family it belongs to (Adobe's lens-distortion work, e.g., US 2007/0098296 A1, "Estimating and removing lens distortion from scenes," which describes removing distortion and then applying inverse transformations to restore an image to its original distorted form), this reference teaches both halves of the claimed remap — and expressly contemplates restoring a distorted presentation after editing. Motivation: same problem (rendering images with controlled radial distortion), same field.
• US 2007/0229665 A1 (Tobiason), "Robust field of view distortion calibration." Supplies the calibration/modeling layer: how to characterize distortion across the field of view, which is what a POSITA needs to compute a position-dependent remap. Motivation: you cannot remap what you cannot model; Tobiason is the known, standard tool for that step.

This combination is attractive to a challenger because both references are already of record on the '052 face — minimizing the "new art" burden and squarely addressing the "consistent lens characteristics" limitation.

Ground E — Okubo (or Sokeila) + the panoramic-simulation art of record (Ritchey US 5,130,794; Imove US 6,337,683; Ritchey US 2010/0045773)

For dependent claims 8/18 (positions simulate virtual re-pointing) and 7/17 (consistent frame-to-frame look): Imove US 6,337,683 ("Panoramic movies which simulate movement through multidimensional space") and Ritchey US 5,130,794 / US 2010/0045773 ("Panoramic display system"/"panoramic adapter system with spherical field-of-view coverage") expressly teach simulating camera motion and re-orientation by translating a virtual view window over a wide/panoramic capture. Motivation: these references exist to solve exactly the problem of making post-capture viewpoint changes look like native capture — the '052's stated purpose. They also independently establish that virtual re-pointing over a wide-FOV capture was a known technique by 2015, which supports the § 103 rationale under KSR ("known technique applied to a similar device").

Ground F — Okubo + Sokeila + Eder + Tobiason (belt-and-suspenders)

Where the sub-frame-size dispute is contested, a challenger can combine: Okubo (primary; two-stage distortion, cutouts, stabilization), Sokeila (adaptive/variable cropping and non-centric selection), Eder (re-distortion for perceptual consistency; movable pixel region), and Tobiason (distortion calibration/modeling). Each reference supplies a distinct, well-understood element with an articulated, art-recognized function — the textbook KSR combination.

V. Why the combinations are motivated (consolidated)

VI. Where the obviousness case is vulnerable (and GoPro's counterarguments)

1. Okubo's two-stage distortion correction may be limited to still images. GoPro's POPR (Paper 7, § XI) argues Okubo's "distortion correcting 1 / distortion correcting 2" pipeline is a still-image mode, so applying it to video frames — a requirement of claims 1/11 — is not disclosed, and there is no motivation to modify. A challenger must rebut this with Okubo's fourth embodiment (inter-frame vibration in moving images) and/or Sokeila's video-stabilization teaching. This is the single most important factual battleground.
2. Eder cuts against the challenger on element 1[g]. The ITC ID found that Eder only de-distorts "area A" to plane-mirror level; the post-distortion is applied to the wide image. Don't lead with Eder.
3. Claim construction. GoPro's non-obviousness story leans heavily on the direct single-function mapping disclosed in the specification. Because that limitation is absent from claims 1 and 11 as issued, this argument should fail — but if a court imports the "higher quality and less loss" direct-mapping advantage from the specification into the claims, the analysis shifts. Worth watching on appeal (Fed. Cir. 26-1752, as identified in the prior section; still not independently verified there).
4. Secondary considerations. GoPro asserted commercial success (HERO line), long-felt but unmet need, unexpected results, and competitor copying in IPR2025-00017. The classic vulnerabilities: the '052 claims are not coextensive with a whole HERO camera (the camera embodies many other patented features), and "copying" evidence consisting of competitors implementing the broadly-known correct-then-reapply distortion technique is weak absent proof it was copied from the '052. The ITC ID appears to have found these arguments did not change the outcome on the Okubo ground.
5. The Board never ruled on the merits. Institution was denied on Fintiv, so no PTAB precedent binds a district court. A challenger in the stayed C.D. Cal. case (GoPro v. Arashi Vision, 8:24-cv-00681) must prove invalidity de novo by clear and convincing evidence.
6. Truncated citation list. Because only ~40 of 169 cited references are visible in the record I was given, additional on-face combinations (particularly in the video-editing and spherical-imaging portions of that list) may exist. The grounds above are those I can ground in references I can actually cite.

VII. Bottom line

Strongest § 103 case: Okubo alone, or Okubo + Sokeila, against claims 1–10 and 11–20. Okubo supplies the full access → select-reduced-cutout → correct-to-zero → re-apply-desired-distortion sequence with centered, position-dependent lens distortion, in-camera implementation, stabilization, motion/orientation metadata, and video mode; Sokeila cures any "fixed crop" objection with adaptive, non-centric, content-aware cropping. The motivation is a same-field, same-problem, known-technique improvement with predictable results — squarely within KSR. This is also the theory the ITC ALJ credited at the Final Initial Determination.

Secondary (corroborative) combinations: Okubo + Eder (re-distortion for perceptual consistency; movable pixel region → virtual re-pointing); Okubo + Adobe WO 2009/146319 and/or Tobiason US 2007/0229665 (both already of record on the '052 face) for the distortion-modeling and remove-then-re-apply steps; Okubo/Sokeila + Imove US 6,337,683 or Ritchey US 5,130,794 for virtual re-pointing under claims 8/18.

Weakest: any ground led by Eder alone (element 1[g] fails on the ITC record) or any ground that depends on construing the claims to require a direct single-function remap or variable-sized sub-frames — neither limitation is present in the issued claim text.

Uncertainty flags carried forward: the Commission's precise claim-by-claim validity holding for the '052 remains something I could substantiate only from the ALJ's Final ID excerpts and secondary sources, not from a primary ITC EDIS document; and the Federal Circuit appeal (26-1752) has not been verified from a CAFC docket in this session.