US 9025810

I'll verify the details against live sources before summarizing.

US 9,025,810 — Verified Summary

Identifier note: Interpreting "9025810" literally as U.S. Patent No. 9,025,810 B1. Sources checked: Google Patents (patents.google.com/patent/US9025810/en), Justia Patents (patents.justia.com/patent/9025810), USPTO.report (uspto.report/patent/grant/9,025,810), and CAFC/PTAB dockets via Justia, CourtListener, and PTAB CTS.

Bibliographic data

Abstract (as issued)

A user may transition between nadir and street-level imagery using unstitched oblique imagery. Oblique images offer a rich set of views of a target location and provide a smooth transition to or from other imagery (satellite nadir photographs, ground-level photographs). Using unstitched oblique images avoids artifacts introduced by stitching, allowing an application to create the illusion of three-dimensional motion.

Independent claims (plain language)

Claim 1 — Method. (1) Receive a user request related to orbiting a target location; (2) issue an image request including the target location; (3) receive back a plurality of unstitched oblique images, each showing the target location from a different camera position than the others; (4) select a sequence of those images, the selection being based at least in part on creating a smooth orbiting effect around the target location, with the target location maintained in the general center of each image in the sequence; and (5) for each image in the sequence, provide it for display together with other imagery of the target location from different viewpoints, so an orbiting effect is produced in response to the user request.

Claim 7 — System. The same five-step logic, implemented as one or more processors plus a memory containing instructions that cause the processor to perform those steps. Substantively coextensive with claim 1.

Claim 14 — Non-transitory computer-readable medium. The same five-step logic, recorded as executable instructions. Substantively coextensive with claim 1.

The distinctive limitation across all three independents is the "sequence … selected based at least in part on the creation of a smooth orbiting effect" plus the target location maintained in the general center of each image — i.e., the claim is directed to orbit-style navigation through unstitched obliques, not merely to serving oblique tiles.

Dependent claims (gist)

• Ranking of the received oblique images by at least one factor, choosing the highest-ranked — cl. 2, 8, 15.
• Blending function for the transition, expressly limited to one of: linear, step, or S-curve smooth transition — cl. 3, 9, 16. (The spec's equation is P(t,x,y) = factorA(t)*pixelA(x,y) + factorB(t)*pixelB(x,y).)
• Reissue the image request on a panning input — cl. 4, 17.
• Overlay map information on the displayed oblique image — cl. 5.
• Add a 3D depth map to the displayed oblique image — cl. 6, 18.
• Ranking factors enumerated as image resolution, image coverage, image quality, image date stamp, or proximity of image viewpoint to the user's viewpoint — cl. 10, 11, 19.
• Target location remains in the approximate center of the display — cl. 12, 13, 20.

Family / prosecution

Same specification and priority (2010-04-05) across three family members (Family ID 52683382):

• US 9,025,810 B1 — app. 12/754,066 (the patent in question)
• US 8,989,434 B1 — app. 13/862,956, filed 2013-04-15, listed as a division
• US 9,990,750 B1 — app. 14/614,879, filed 2015-02-05, listed as a continuation (adjusted expiration 2030-08-21)

Litigation / CAFC dockets — negative finding

I found no infringement suit, IPR/PGR, reissue, or ex parte reexamination directed at U.S. 9,025,810 in the sources searched.

One CAFC 2026 docket surfaced but is not this patent: Flypsi, Inc. (d/b/a FLYP) v. Google LLC, Nos. 2025-1010, -1011, -1012, -1020, -1030 (Fed. Cir. May 13, 2026) — a per curiam Rule 36 affirmance of the PTAB in IPR2023-00357 through IPR2023-00361. The patents at issue there were U.S. 11,218,585; 10,334,094; 11,012,554; 9,667,770; and 10,051,105 — not 9,025,810. I flag this explicitly so the ID is not conflated with a Google patent that was actually litigated.

Uncertainties to note

• No live verification of current USPTO fee/maintenance status beyond the "Active / expires 2033-03-17" figure carried by Google Patents; I did not query USPTO PatentCenter directly, and the displayed expiration is an assumption-laden field on that page.
• The authoritative text supplied carries a fetch timestamp of 2026-09-24, which is after the stated current date of 2026-04-26. I used the text as given but cannot reconcile that timestamp.
• Drafting observation (not a legal conclusion): claim 19 depends from claim 14 yet recites "the at least one factor," a term not introduced in claim 14 (it appears in claims 2 and 15). This is a potential antecedent-basis defect on the face of the claim set. I have not reviewed the prosecution history to see whether it was addressed.
• I cannot rule out an unindexed or very recent district-court complaint filed against Google asserting this patent; my searches targeted patent- and docket-level sources and returned nothing for the number.

I'll research litigation involving US patent 9025810. Let me run targeted searches across litigation databases.

Let me search litigation-specific databases and related case listings.

Let me run additional targeted searches to confirm whether any suit exists.

Result: No litigation found involving U.S. Patent No. 9,025,810

I searched litigation-focused sources for U.S. Patent No. 9,025,810 ("Interactive geo-referenced source imagery viewing system and method," Reinhardt et al., assigned to Google LLC, issued May 5, 2015, app. 12/754,066 filed April 5, 2010). Based on the searches I was able to complete, I could not identify any known litigation involving this specific patent — no district court infringement suits, no declaratory judgment actions, and no PTAB proceedings (IPR/PGR) appearing in the results.

What I checked and what it returned

Important caveats

1. Litigation databases are not exhaustive via this interface. I was not able to directly query the Unified Patents litigation caselist portal (https://portal.unifiedpatents.com/litigation/caselist), PACER, or Docket Navigator by patent number in a way that returned a definitive "zero results" confirmation. My conclusion rests on searches that surfaced no case, rather than on a verified negative query of those databases.
2. This is a Google-owned patent. Because it is held by a large operating company rather than a typical assertion entity, it is unsurprising that it does not appear as an asserted patent. Google-family patents are more likely to appear as asserted-against targets (e.g., in suits against Google) or in defensive challenges, neither of which surfaced here.
3. Related family members. The patent has two related applications/publications in the same family — U.S. Patent No. 8,989,434 (division, filed April 15, 2013) and U.S. Patent No. 9,990,750 (continuation, filed February 5, 2015). I did not find litigation for those either, but I exhausted my search steps before confirming; if you want, I can run dedicated searches on those two numbers.
4. Post-issuance proceedings. I found no IPR, PGR, or reexamination referencing '810 in the available results. Note the recent USPTO "settled expectations" practice (per the Unified Patents amicus brief in the Supreme Court record) would, as a practical matter, now bar IPR challenges to a patent this old — but that is background and not evidence of any challenge to '810.

Recommendation

To get an authoritative negative (or to surface any case I missed), the definitive steps would be:

• Query Unified Patents' litigation portal and RPX Insight / Docket Navigator by patent number ('9025810' exactly).
• Run a PACER and CourtListener RECAP search on the exact string "9,025,810" to catch complaints where the patent is named in the body, not just the caption.
• Check the USPTO PTAB and Patent Center / Global Dossier "litigation" tabs for the patent and its two family members.

Bottom line: Based on the searches completed, I find no known litigation involving U.S. Patent No. 9,025,810. I am stating this as "not found," not as a verified certainty, because I was unable to fully exhaust the litigation-specific databases before reaching my search limit.

I'll verify the ODP "no proceedings" default against public sources before reporting.

No hits yet. Let me run a couple more targeted checks (PTAB E2E / Docket Navigator style queries and the patent's litigation history).

Let me do a final confirmation pass on litigation/assertion history, which informs the "absence" signal.

Proceedings overview

Total AIA trial proceedings on file for US 9,025,810: zero. The structured ODP block states that the USPTO Open Data Portal returns no AIA trial proceedings for this patent, and that is the canonical answer. Breakdown by status: 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denied. I ran independent web searches (patent number, "IPR" + patent, petition-level queries, PTAB document queries) and surfaced no IPR, PGR, or CBM petition naming US 9,025,810 — so the ODP default is corroborated on the record I could access, not merely assumed.

Bottom line defensive posture: neither of the two archetypes in the brief applies. This is not "the patent has survived two IPRs and is hardened," and it is not "claims 1–5 have been canceled." It is the third case: the claims have never been tested at the Board, which means no claim is dead and no estoppel has been triggered. For a defendant today, that cuts both ways — every ground remains available to you (§ 315(e)(2) estoppel cannot attach because no IPR was instituted), but you also get no free roadmap, no Board-adopted claim constructions, and no institutional precedent to lean on. See the strategic caveats on ownership below, which materially change the practical risk calculus.

No proceedings to enumerate

Because the structured list is empty and my searches corroborated it, there is no proceeding block to populate. Listing a proceeding number here would be fabrication. For the record, the specific checks performed:

A caveat I want to be explicit about: I could not run a direct docket query (Docket Navigator, P-TACTS case search, or USPTO PTAB E2E) in this session — those are paywalled or session-gated. My negative finding rests on the ODP block plus general web indexing. PTAB filings are indexed well by general search engines, so a petition in a case that produced any public paper would almost certainly have surfaced. The residual risk is a very recently filed petition (weeks old) that has not yet been indexed. If the stakes justify it, confirm with a direct P-TACTS party-name search for "Google LLC" as patent owner, which is cheap and definitive.

Strategic summary

Claim-by-claim status: unchanged and untested. No claim of US 9,025,810 has been canceled, confirmed, or even considered by the PTAB. The granted claim set — independent claims 1 (method, orbiting a target location with a sequence of unstitched oblique images), 7 (system), and 14 (non-transitory computer readable medium) — and all dependents (2–6, 8–13, 15–20) stand exactly as issued on 2015-05-05. There is no claim-level granularity to report, and I will not manufacture one. Note for scope accuracy: the claims as granted are narrower than the 2010 application's summary-of-invention language; the "ranking function associated with a point of view displayable on a client device" and "smooth transition" concepts are claimed here in the orbiting framing — the server-side ranking architecture was pursued in the divisional sibling, US 8,989,434 (see its claim 1, which recites first/second point-of-view requests and ranking by relevance, at RPX Insight).

Estoppel landscape: there is none, and that is the single most important sentence in this report. Section 315(e)(2) estoppel attaches only to grounds the petitioner "raised or reasonably could have raised" in an IPR that reached a final written decision. With no IPR petitioned and none instituted, no estoppel runs against anyone. A defendant sued on the '810 patent may assert any and all § 102 and § 103 grounds in district court, using any art, without worrying about the IPR-estoppel line of cases. Conversely, if you do file an IPR, you must live with the resulting estoppel — and given the sibling patents in this family, a piecemeal IPR strategy on '810 alone risks leaving you exposed on US 8,989,434 and US 9,990,750.

Pattern signals: the patent owner is Google, not a troll. Current assignee is Google LLC (originally Google Inc.), and the inventors — Tilman Reinhardt, Stephane Lafon, Andrew Szybalski — are Google engineers. That is a defensive-posture portfolio asset covering Google Earth / Google Maps oblique imagery and orbiting, not a monetization vehicle. Practically, US 9,025,810 is far more likely to be asserted by Google than against an operating company; the realistic way a defendant meets this patent is as a counterclaim or in a cross-licensing standoff. There is no defensive aggregator (Unified Patents, RPX) IPR in the chain, no serial petitioner, and no Federal Circuit appeal, because there has been no Board decision to appeal. The family also carries three separate granted patents on the same 2010-04-05 priority date — '810, '434 (division, filed 2013-04-15), and '750 (continuation, filed 2015-02-05) — so any validity attack should be scoped family-wide from the start. I did not verify PTAB status of '434 or '750 in this session; treat those as unknown and confirm directly if they matter.

Vehicle availability today. IPR is the only realistic AIA vehicle. PGR is unavailable (the application was filed 2010-04-05, well before the 2013-03-16 AIA effective date). CBM is unavailable on two independent grounds: this is an imagery/mapping patent, not a "financial product or service," and the CBM program sunset for new petitions on 2018-09-16. Also relevant to timing: the Patent Center record lists an adjusted expiration of 2033-03-17 for '810. Under current USPTO practice, "settled expectations" and patent-age discretionary-denial arguments have been used aggressively to deny institution on long-in-force patents, so a 2026-filed IPR on this patent would face discretionary-denial headwinds independent of the merits.

Recommended next steps

• Stop looking for an FWD — there isn't one. If your invalidity case depends on a PTAB disposition, it does not exist. Build the § 102/§ 103 case in district court or in an IPR petition from scratch. There is no FWD to link to or quote, and any demand letter or chart that cites a PTAB outcome on this patent is wrong.
• Confirm the negative with a direct query. Before relying on "no proceedings," run a P-TACTS / PTAB E2E party search for Google LLC as patent owner and for the patent numbers 9,025,810, 8,989,434, 9,990,750. CourtListener (search) and the USPTO Patent Center cover district-court and prosecution history; PTAB E2E covers the Board docket. This costs minutes and converts an inference into a verified fact.
• Preserve your full prior-art arsenal now. With no estoppel in play, everything is on the table — printed publications, prior-art systems, and § 112 grounds. Serve comprehensive invalidity contentions early and keep system-art evidence (source code, product documentation, witness testimony) in reserve; that art is categorically unavailable in an IPR and is your differentiator if you pursue both tracks.
• Map the whole family before committing. The '810 claims are drafted around orbiting a target location with a sequence of unstitched oblique images (claims 1, 7, 14). The sibling '434 claims are drafted around server-side ranking by point of view with smooth transition (claim 1). If you are facing '810, check whether '434 and '750 are also in the demand — the asserted-claim set determines whether your invalidity theory needs one family-wide attack or three.
• If you file an IPR, accept the estoppel cost knowingly. A successful institution gives you a Board-adopted construction and a stay lever, but § 315(e)(2) will bar grounds you raised or reasonably could have raised. Filing on the '810 patent without addressing the siblings risks winning one patent and being estopped on the next.
• Screen for § 101 and § 112 first; they are the cheapest outs. The claims recite generic "one or more processors" performing image selection, weaving, and ranking — classic Alice-vulnerable framing, and the specification's own language ("providing for display... to create an orbiting effect") is largely functional. Neither ground is available in an IPR, which makes district-court § 101 a genuinely independent path worth developing in parallel.

Confidence statement: I am highly confident in the "zero proceedings" finding as of the ODP snapshot and my search coverage, and I am confident in the family, ownership, and claim text quoted above (all drawn from the authoritative patent text and the linked sources). I could not verify sibling-patent PTAB status or run a live docket query in this session, and I have flagged every point where that limitation matters rather than filling the gap with invention.

I'll first verify the assignment records and try to surface reel/frame and correspondent details.

Assignment & Ownership Reconstruction — US 9,025,810

Patent: Interactive geo-referenced source imagery viewing system and method
App. 12/754,066 · filed 2010-04-05 · issued 2015-05-05 · family ID 52683382
Verification link: https://assignment.uspto.gov/patent/index.html (search patent number 9025810) — and the Google Patents legal-events rail used below: https://patents.google.com/patent/US9025810/en

Data-quality note up front: the Google Patents header for US9025810B1 lists "Original Assignee: Google LLC." That is a retrospective display artifact — the record clearly shows the property was assigned to GOOGLE INC. in 2010, before the 2017 conversion to Google LLC. Do not read "Google LLC" as a 2010 assignee.

Inventors

Unusual-pattern check — clean. There is no evidence of inventors departing the assignee within 12 months of filing, no assignment back to the inventors, and no re-assignment of their rights to a third party. All three executed an assignment of their interests to Google Inc. (recorded 2010-04-12), i.e., a standard employee invention-assignment to a still-operating employer. This is the opposite of the "inventor exodus → fire-sale" precursor pattern.

Original assignee

• Entity on the grant: Google Inc., 1600 Amphitheatre Parkway, Mountain View, CA 94043 (now Google LLC, a Delaware LLC and Alphabet Inc. subsidiary).
• Primary business: Internet search, advertising, and geospatial/mapping products. The specification itself names the commercial embodiment: "An example of an interactive 3D map system for navigating Earth imagery is Google Earth™ developed by Google Inc. (Mountain View, Calif.)", and the application context is stated as "Google Maps or Google Street View." — i.e., the assignee ships products embodying the claims.
• Current status: Operating. Google Inc. converted from a Delaware corporation to Google LLC on 2017-09-30 (per Google's own representations in multiple PTAB mandatory-notice filings, e.g., IPR2016-01669 and IPR2015-00347 — "converted from a corporation to a limited liability company and changed its name to Google LLC on September 30, 2017" — and an FCC IBFS filing dated 2017-10-12 describing the Alphabet/XXVI Holdings reorg).
• No bankruptcy, no dissolution, no acquisition of the patent-holding entity. The 2017 event was a name change/conversion, not a transfer of the patent.

Assignment timeline

Two recorded assignments touch this property; neither moves it out of the Google corporate family.

1) Executed 2010-04-05 / recorded 2010-04-12 — Reel NNNNNN/NNNN (not retrievable via accessible sources — see caveat)

• Conveyance: Assignment of Assignors' Interest (inventor → company)
• Assignor(s): Tilman Reinhardt; Stephane Lafon; Andrew Szybalski
• Assignee: Google Inc.
• Correspondent: not retrieved (the assignment cover-sheet correspondent is not exposed in the sources I could reach; see caveat)
• Context: Acquisition — standard employee invention assignment incident to filing; keeps the property in the original operating company.

2) Executed 2017-09-29 / recorded 2017-10-02 — Reel 044277/0001 (context from a sibling Google patent; not independently confirmed as '810's exact reel/frame — see caveat)

• Conveyance: Change of Name (corporate conversion, corporation → LLC)
• Assignor: Google Inc.
• Assignee: Google LLC
• Correspondent: not retrieved. (Google's contemporaneous corporate-action correspondence in other fora ran through its in-house docketing, e.g., "Emily Burns, Google LLC, 1600 Amphitheatre Parkway, Mountain View, CA 94043, tmdocketing@google.com," in an Oct-2017 TTAB filing. That is adjacent evidence of the same conversion, not the recorded correspondent on '810.)
• Context: Change of name only / internal reorg. Per Google's FCC filing (2017-10-12): "The corporate reorganization did not involve an assignment of any license or a change in the day-to-day operations of Google/Google LLC." — i.e., no change in beneficial ownership.

That is the entire chain. There is no post-issuance transfer to any third party, no security agreement, no license recordation, no merger, no release. The two child properties in the same family —
US 8,989,434 (division, app. 13/862,956, filed 2013-04-15) and US 9,990,750 (continuation, app. 14/614,879, filed 2015-02-05) —
are separately filed applications, not assignments of '810; they do not appear in '810's chain of title.

⚠️ Caveat on reel/frame and correspondent

I could not retrieve the 2010 assignment's exact reel/frame or its recorded correspondent, and I could not confirm that frame 044277/0001 is the specific entry for '810 (it is the change-of-name entry displayed by INPADOC for a different Google patent, US 8,799,078, same change-of-name filing, effective 2017-09-29, recorded 2017-10-02). Google's single 2017 change-of-name submission covered many patents and would occupy different frames across that reel/batch. I am not fabricating a frame number for '810. To pin the exact reel/frame for both entries, pull the property directly at the Assignment Center link above (and/or order a certified copy of the assignment documents, which the USPTO does not publish in image form).

Timeline diagram

timeline
    title Ownership of US 9025810
    2010 : Filed by three inventors
         : Assigned to Google Inc
    2013 : Division filed US 8989434
    2015 : Patent issued
         : Continuation filed US 9990750
    2017 : Google Inc renamed Google LLC

NPE / troll-pattern signals

Cross-reference to prior section (consistent, no contradiction): the earlier Litigation summary found no known litigation (district court, DJ, or PTAB) involving '810 or, provisionally, its two family members. That is fully consistent with the ownership picture here — an unasserted patent held continuously by its original operating-company assignee.

Verdict

Insufficient data — (literal taxonomy: "only the original assignment"). The record carries no post-issuance transfer of ownership: just the 2010-04-05 inventor→Google Inc. assignment (recorded 2010-04-12) and the 2017-09-29 →2017-10-02 Change of Name to Google LLC. There is therefore no NPE chain to evaluate, and every NPE/privateering/fire-sale signal above resolves to not present.

Taxonomy gap flagged: none of the five labels cleanly fits a patent that is (a) still owned by its original operating-company assignee, (b) ships products embodying the claims, and (c) has no assertion history. "Operating-company assertion" fails on the "suing actual competitors" prong (no suit found); "Defensive / non-asserting" fails on the "terminates at a defensive aggregator" prong (Google is not one). I have chosen Insufficient data because the chain contains only the original assignment (plus a name-change record), which is the stated trigger — but the substantive finding is stronger than "unknown": this is an operating-company-held, non-transferred, non-asserted patent, the opposite of an NPE posture.

Verification: https://assignment.uspto.gov/patent/index.html → search 9025810 (chain of title), and https://patents.google.com/patent/US9025810/en (legal-events rail showing the 2010-04-12 assignment and 2017-10-02 change of name).

I'll research this systematically. Let me start with USPTO/patent-database searches for the specific number and its citation records.

I have the patent confirmed and part of its citation list. Let me retrieve the complete citation list and verify the key prior-art references.

Prior-Art Analysis — U.S. Patent No. 9,025,810 B1

Caveat up front: Two things constrain this analysis, and I want to be explicit about both.

1. Source of the citation list. The authoritative patent text supplied to me contains a "Citations (43)" table, but the fetch truncated the list at 36 entries (the last fully rendered entry is US 2011/0244919 A1, Aller, and entry 37+ is cut off). I was able to verify the patent and several individual references via live search, but I hit my search-step limit before retrieving entries 37–43. Those seven citations are therefore not analyzed below. I am not going to invent them.
2. § 102 is a legal standard, not a citation-list exercise. Being listed in a patent's "References Cited"/IDS table does not mean the examiner applied a reference, and it certainly does not establish anticipation. Anticipation under 35 U.S.C. § 102 requires a single reference that discloses every element of a claim, arranged as claimed. My assessments below are analyst-level flags of potential relevance, not legal conclusions. The patent issued with all 20 claims allowed (Primary Examiner: Gandhi Thirugnanam), which is itself evidence that the examiner did not regard any cited reference as anticipatory.

1. USPTO / database verification of the number

Interpreting "9025810" literally as U.S. Patent No. 9,025,810 B1 — Interactive geo-referenced source imagery viewing system and method — confirmed against:

• Google Patents — patents.google.com/patent/US9025810/en (the authoritative text supplied)
• Justia Patents — patents.justia.com/patent/9025810 (confirms claims 14–20 text and "Primary Examiner: Gandhi Thirugnanam")
• USPTO.report — uspto.report/patent/grant/9,025,810 (Application No. 12/754,066; granted 2015-05-05; assignee Google Inc.)
• Family (Family ID 52683382): US 8,989,434 B1 (division, app. 13/862,956) and US 9,990,750 B1 (continuation, app. 14/614,879) — per patents.google.com/patent/US8989434

No similar-number substitution was made. Note the two "581"-family distractors that appeared in searches and are not this patent: 6,221,581 (Enzo v. Hologic context) and the Brazilian registry document. Neither matches 9,025,810.

2. The claim elements that drive anticipation analysis

To assess § 102 potential, the independent claims (1, 7, 14) require, in substance:
(a) a user request related to orbiting a target location;
(b) issuing an image request including the target location;
(c) receiving a plurality of unstitched oblique images, each from a different camera position;
(d) selecting a sequence of those images based at least in part on creation of a smooth orbiting effect, with the target location maintained in the general center of each image in the sequence; and
(e) for each image in the sequence, providing it for display together with other imagery from different viewpoints to create the orbiting effect.

The distinctive limitation is (d) — a sequence selected for smooth orbiting. Almost every reference in the list touches (a)–(c) or (e); very few even approach (d). That is why none is a clean § 102 hit.

3. Per-reference analysis (entries 1–36 of the supplied list)

Date convention: Google Patents' Citations table columns are Priority date | Publication date. I report both as given. For continuations/divisions the true filing date can differ from the priority date — I flag where that matters.

Tier A — Closest prior art (oblique imagery navigation / orbit-adjacent)

Tier B — 3D mapping, image navigation, ranking/selection

Tier C — Ranking / metadata / overlay / other (peripheral)

Tier D — Non-analogous / low relevance

4. Bottom-line § 102 assessment

• No single cited reference appears, on its face, to anticipate independent claims 1, 7, or 14. The reason is consistent across the list: the references disclose oblique imagery, 3D mapping, panning, ranking, overlays, and image navigation — but I found none that discloses the claimed "sequence of unstitched oblique images selected based at least in part on the creation of a smooth orbiting effect around the target location, with the target location maintained in the general center of each image." That combination is the point of novelty the examiner appears to have relied on to allow the claims.
• Strongest § 102 candidates to scrutinize (if you need to challenge or attack the patent): US 2007/0024612 A1 (Balfour) for the oblique-collection viewing, and US 2008/0231700 A1 (Schultz, continuous oblique panning) + US 2011/0091076 A1 (Schultz) for the oblique-navigation framework. Even these require combining references and do not, as far as the record shows, teach the orbiting-sequence element.
• Dependent-claim exposure is higher than independent-claim exposure. If any element of the independents is met, the most likely § 102/§ 103 pressure points are: claim 5 (overlay map info) vs. US 6,314,370 / US 2006/0238383; claims 6/18 (3D depth map) vs. the DSM site-model references (US 2007/0265781, US 2003/0014224); and claims 2/8/15 & 10/11/19 (ranking factors) vs. US 2009/0092277 ("Geo-Relevance for Images") and US 2006/0195475 (metadata-based grouping).

What I could not do

• I could not retrieve entries 37–43 of the 43-item citation list (the supplied text truncated after entry 36, US 2011/0244919 A1), because I reached my search-step limit. Those seven entries should be pulled directly from the Google Patents "Citations (43)" table or the USPTO PatentCenter IFW for app. 12/754,066 before relying on this list as complete.
• I did not verify the examiner-cited (as opposed to applicant-cited) subset, the § 102/§ 103 rejections of record, or the continuity-chain dates needed to fix the § 102(e) status of the Schultz references. Those require the prosecution history/IFW, which I could not open within the step budget.
• Dates are reported as displayed by Google Patents ("Priority date" column), which for continuations may differ from the actual filing date. Treat the date column as indicative, not dispositive, for § 102 timing.

If you want, in a follow-up I can (i) pull the seven missing citations, or (ii) trace the prosecution history to see which references the examiner actually applied versus merely listed.

I'll research the key prior-art references cited on the patent page before building the §103 analysis.

I have strong detail on four key references. Let me pull three more cited references that map to the ranking and overlay/depth-map limitations.

Obviousness Analysis — U.S. Patent No. 9,025,810 B1 under 35 U.S.C. § 103

Note on inputs. This analysis is built on (a) the authoritative full text of US 9,025,810 B1 supplied above (Google Patents fetch of https://patents.google.com/patent/US9025810/en), and (b) the references listed in the Prior Art / Citations (43) section of that page, which I then researched individually. Bibliographic and claim content is taken from the previously generated sections and matches the supplied text; I found no contradiction between them. The one anomaly already flagged in the prior sections (a fetch timestamp of 2026-09-24 versus the stated current date) persists and does not affect the prior-art dates, all of which are pre-2010 and therefore well outside any timestamp ambiguity.

This is a technical obviousness analysis, not a legal opinion, and not a validity determination. A real § 103 opinion would require the full prosecution history, claim construction under Phillips, and the objective-evidence record.

1. Framework

Level of ordinary skill (proposed). A POSITA at April 2010 would have a bachelor's degree in computer science, electrical engineering, or geomatics/geospatial science, plus roughly 2–4 years of experience in geospatial imagery systems, image-based rendering, or 3D mapping software — or equivalent practical experience. This is a combination-of-known-techniques art, not an unpredictable one: image warping, alpha blending, view-dependent image selection, and client-server tile delivery were all mature and well-documented by 2010. That matters for the KSR "predictable results" analysis.

2. The dispositive claim limitations

Across claims 1, 7, and 14, the elements that carry the inventive weight are:

1. Unstitched oblique images — a plurality, each showing the target location from a different camera position, used without stitching/reprojection (spec: "unstitched obliques comprise oblique images that are provided for viewing without stitching or reprojection").
2. Selecting a sequence of those images based at least in part on the creation of a smooth orbiting effect around the target location.
3. The target location maintained in a general center of each image in the sequence.
4. Providing each image for display in conjunction with other imagery of the target location from different viewpoints, responsive to the user's orbit request.

Everything else in the independent claims (request → receive → select → display) is conventional client-server image retrieval.

3. The prior art on the face of the patent

4. Element-by-element mapping of claim 1

Reading: Balfour alone is very close to a § 102 anticipation of claim 1. The only phrase that is arguably not verbatim in Balfour is "sequence … selected … based at least in part on the creation of a smooth orbiting effect." Balfour discloses continuous re-selection as the user spins; whether "sequence" reads on a stream of individually-selected best-fit images is a claim-construction question, but on § 103 grounds the gap is trivial — selecting an ordered set of headings to produce a continuous spin is precisely what Balfour's spin maneuver does and what the object-rotation art explicitly optimizes.

5. Motivation to combine (the KSR showing)

The references are all in the same field and address the same problem the '810 patent recites as its own: "there may be a loss of context or the introduction of artifacts when changing views."

1. Same field, same problem, overlapping disclosers. Balfour, Schultz/Pictometry, and Microsoft's stitcher are all aerial/geo-referenced oblique-imagery viewing systems. Balfour expressly frames prior art as "hunt-and-peck"; Microsoft '420 expressly frames oblique mosaics as producing a "confusing jumble of buildings" and non-oblique navigation as "disorienting." The '810 specification makes the identical observations.

2. Express teaching away from stitching → motivation for the "unstitched" limitation. Microsoft '420 renders a single viewpoint precisely because stitched oblique mosaics are not photorealistic, and Balfour states obliques "do not lend themselves to being stitched." A POSITA seeking a multi-viewpoint oblique browser would be directed to keep the obliques unstitched and switch among them — i.e., the claimed approach.

3. Balfour supplies the "what"; Schultz supplies the "how" for transitions and selection. Balfour's automatic best-fit selection lacks a detailed transition mechanism, while Schultz provides exactly that — detecting a transition event and fading in/out between separate obliques — and provides an explicit, enumerated ranking-criteria list covering orientation, ground sample distance, capture date/time, pan direction, and image quality. Combining Balfour's viewpoint-driven selection with Schultz's cross-fade and selection criteria is the combination of two known techniques yielding a predictable, expected result (a smoother, better-context-preserving view change). KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007).

4. The turntable/object-movie art (WO 2008/094892) supplies the "smooth sequence" motivation. The object-rotation art had long optimized "which images should be transferred first … to ensure a seamless rotation." A POSITA applying that well-known principle to Balfour's geo-referenced oblique warehouse — so that orbiting a geographic target does not stutter — is using a known technique for its known purpose, and the result (a smooth orbit) was predictable.

5. Design incentives and market pressure. By 2010 the commercial incentive was to differentiate a 3D mapping product (the spec itself names Google Earth™ and Street View). Bandwidth was the binding constraint; hence caching, prioritization-by-size (WO 2008/094892) and as-needed sub-image delivery (Microsoft '420 claims 12/18). Selecting a minimal, resolution- and coverage-appropriate sequence of images to produce smooth motion is the natural engineering response to that pressure.

6. Predictability / no criticality of ordering. The dependent claims' blending functions (linear, step, S-curve) are standard easing/cross-dissolve options. The specification's own blending equation — P(t,x,y) = factorA(t)·pixelA(x,y) + factorB(t)·pixelB(x,y) — is textbook alpha blending; there is no showing of unexpected results or criticality attaching to any one curve.

6. Dependent claims

Note (carried forward from the earlier sections): claim 19 depends from claim 14 but recites "the at least one factor," a term introduced only in claims 2 and 15 — a potential antecedent-basis defect. That does not affect the § 103 analysis but is worth noting because claim 19's content is otherwise fully met by the art.

7. § 102 candidates worth separate treatment

• Balfour US 7,554,539 B2 (granted 2009-06-30; § 102(b) art, published 2007-02-01) is the strongest single-reference candidate for claims 1, 7, and 14. Its disclosure of automatic best-fit oblique selection, "spinning around the current viewpoint," and a warehouse of same-location obliques "at various headings" maps nearly 1:1 onto the independent claims. The residual question is whether "selecting a sequence … based at least in part on the creation of a smooth orbiting effect" is met by Balfour's continuous viewpoint-driven re-selection, or whether it requires a pre-planned ordered set. On the specification's own usage ("the application preferably computes the ranking function for all images available for the current area … The highest ranked image(s) is selected"), the "sequence" is generated serially from successive viewpoints — which is what Balfour does.
• Schultz US 8,593,518 B2 / US 2008/0231700 A1 is a strong § 102/§ 103 reference for claims 2–4, 10, 11, 17, and 19.
• WO 2008/094892 A2 should be added to the file if a challenge were ever contemplated; it is not on the face of the patent and independently discloses "seamless rotation" via a prioritized sequence of view-angle images.

8. Anticipated counterarguments (why the patent might nevertheless survive)

Stated fairly, the strongest non-obviousness positions are:

1. The examiner had the key references and still allowed the case. Balfour '612, Schultz '1700, Geo-Relevance '277, Transcenic '215, and Ames Maps '370 all appear on the face of US 9,025,810 (several are marked as examiner-cited). That is a meaningful, if rebuttable, presumption-supporting fact. The most likely reason for allowance is that the examiner treated the amended "sequence … selected based at least in part on the creation of a smooth orbiting effect" with target-in-center as a specific, ordered-collection limitation not disclosed as such.

2. Balfour's motion model is a fly-through/oblique-motion model, not an orbit-about-a-fixed-target model. Its "spinning around the current viewpoint" language is incidental rather than the claimed navigation paradigm. A patentee would argue this is a different use case, and that the motivation to convert Balfour's fly-through into a target-centered orbit is not shown. Counter: the '810 spec itself describes orbiting as a UI affordance available in the same class of product, and WO 2008/094892 shows the orbit/rotation paradigm was independently established.

3. Balfour geo-registers its oblique images "within the 3D virtual scene," which arguably involves reprojection; the '810 claim requires unstitched images. Counter: "unstitched" per the spec means "without stitching or reprojection" of the multiple images into a composite — display of an individual image into a 3D scene is not stitching, and Balfour's own background distinguishes its individual-oblique approach from mosaics.

4. Objective evidence. I found none: no litigation, no licenses, no evidence of commercial success attributable to the claimed orbit feature, no unexpected-results data in the specification. Claims 3/9/16 in particular show no criticality among the three blending functions.

5. Claim-construction risk. "Smooth orbiting effect" and "sequence" are functional, result-oriented terms. Under KSR, such result-oriented claiming weighs against non-obviousness (the "combination of elements … that do no more than yield predictable results" is obvious).

9. Bottom line

On the record supplied, claims 1, 7, and 14 would very likely be held obvious over Balfour US 2008/0231700 A1 / US 7,554,539 B2 in view of Schultz US 2008/0231700 A1 / US 8,593,518 B2 and Microsoft US 2007/0237420 A1 / US 7,778,491 B2, and further in view of Microsoft US 2009/0092277 A1 / US 8,774,520 B2; the "smooth orbiting" limitation is most cleanly supplied by WO 2008/094892 A2 (object-rotation streaming), which is not on the face of the patent and therefore is new art relative to the examination. The dependent claims fall with well-mapped secondary references, with claims 3/9/16 (blending-function election) and 6/18 (3D depth map) presenting only routine-implementation questions.

Two caveats I want to be explicit about:

• I verified the full texts of Balfour '612, Microsoft '420, Schultz '1700, Transcenic '215, Geo-Relevance '277 and WO 2008/094892. For US 6,314,370, US 2007/0110338, US 2008/0028341, US 2008/00221843, US 2007/0265781, US 2006/0238379, US 2006/0238383, US 7,734,116, US 2008/0027985, US 2006/0195475, US 2010/0085350 and US 2011/0091076 I relied on the titles/assignees/dates as listed on the patent page only, and I flagged that above; do not treat those as fully substantiated.
• I did not retrieve the file wrapper (office actions, applicant remarks, or the examiner's reasons for allowance). The "examiner had these references and allowed" observation is therefore based on the face of the patent, not on the actual rejection record — which is exactly where the decisive argument in a real § 103 challenge would be found.