US 8035644

I'll verify the patent record and check for any 2026 CAFC docket activity before summarizing.

Let me check specifically for 2026 litigation and PTAB/appeal activity naming this patent.

USPTO / Court Records Check — US 8,035,644

What I searched and what I did not find

• The authoritative record is the patent document itself (Google Patents full text, fetched 2026-09-23): https://patents.google.com/patent/US8035644/en
• I searched for a 2026 Court of Appeals for the Federal Circuit docket naming this patent. I did not find one. The current CAFC scheduled-cases list (April 2026 revision) and the March–May 2026 Federal Circuit round-ups (Gibson Dunn, WilmerHale, Fed Circuit Blog) contain no entry for 8,035,644 or for a party associated with it. Caveat: my search coverage of the CAFC docket is web-search based, not a direct PACER/CAFC CM-ECF query; I cannot rule out a 2026 appeal I did not surface, but I found no evidence of one.
• Older, non‑2026 litigation does exist and is reflected on the patent's own record page: E.D. Tex. 6:17‑cv‑00428; N.D. Ill. 1:21‑cv‑01840; S.D.N.Y. 1:21‑cv‑07917, 1:22‑cv‑04963, 1:22‑cv‑07114. A 2021–2022 Unified Patents "Patroll" prior-art contest for US 8,035,644 (ended ~March 2022) describes the patent as then owned by Cine Graphic Solutions and being asserted against Motorola, Samsung and Fujifilm: https://patroll.unifiedpatents.com/contests/HcLEbgtYaYeC863Bz — treat the "Cine Graphic Solutions" ownership statement as a search result, not a verified assignment record.

Patent Summary — US 8,035,644 B2

Continuity. This patent is a continuation‑in‑part of 11/586,016 (now US 7,388,587), itself a CIP of 11/403/374 (now US 7,629,977), which claims benefit of provisional 60/670,402. It is the parent (via CIP) of 13/232,306 → US 8,487,939, and through that line to US 11,232,768, 11,263,998, 11,832,768, 11,893,965, 12,051,391, etc.

Abstract (verbatim gist). Provided is a method for providing animation in electronic communications. An image is generated by capturing multiple photographs from a camera or video camera typically fixed in one position. The first photograph is called the "naked photo." Using a graphics program, photos subsequent to the naked photo are edited to cut an element common to the subsequent photos. The cut images are pasted into the naked photo as layers. The modified naked photo, including the layers, is stored as a web‑enabled graphics file, which is then transmitted in conjunction with electronic communication. When the electronic communication is received, the naked photo is displayed and each of the layers is displayed and removed in the order that each was taken with a short delay between photos. In this manner, a movie is generated with much smaller files than is currently possible.

Independent Claims in Plain Language

Claim 1 — Method for providing animation in an electronic message. Capture a "first image" of a scene from a particular location using a defined set of photographic parameters (i.e., a fixed camera position/optics). Capture a further set of images in sequence from the same location with the same parameters. Set a sensitivity level defined in terms of pixel blocks. Identify the portions of the sequential images that differ from the first image by an amount that meets that sensitivity level. Cut out those identified portions. Superimpose the cut images onto the first image as layers — each layer placed where the element appeared in its source sequential image, and shown in a time order matching the capture timing. Save the first image plus the layers together as a single, web‑enabled graphic file.

The two elements that distinguish this claim family are (a) the pixel‑block sensitivity threshold used to auto‑identify the moving/most‑changed regions, and (b) the "flattening" of base image + cut layers into one web‑enabled graphic file that needs no plug‑in to play.

Claim 8 — Electronic message (article of manufacture). A single web‑enabled graphic file stored in memory, the file containing: a first image of a scene captured from a particular location with defined photographic parameters; a plurality of sequentially captured images from the same location/parameters; identified portions of those sequential images that differ from the first image; and cut images corresponding to those portions — with the cut images superimposed on the first image as layers, positioned to match their source images and ordered in time to match capture timing. (Claims 9–14 depend on this and add, e.g., a file name without a file extension, embedding in email/web page, display-and-remove logic, and HTML embedding.)

Claim 15 — Computer programming product. A memory with logic that: defines a pixel‑block sensitivity level; identifies portions of sequential images differing from a first image to that degree (the first image and the sequential images having been captured from the same location with the same photographic parameters, per the claim's internal "wherein" clauses); cuts out those portions; superimposes them on the first image as layers with matching position and matching time sequence; and saves the first image and layers as a web‑enabled graphic file. (Claim 17 adds the file‑extension‑deletion limitation; claims 18–21 add email, web page, and HTML embedding.)

Dependent-claim highlights

• Cl. 2 / 16: the sensitivity level corresponds to a degree of movement of the identified portions.
• Cl. 6 / 13: recite the display cycle — show the first image, successively superimpose each layer in sequential order at its original position, and remove each layer before displaying the next.
• Cl. 7 / 14 / 21: embed the graphic file in an HTML file transmitted with an email.
• Cl. 9 / 17: saving the file with the extension portion of the file name deleted (specification explains this prevents recipients from saving the animation as anything other than a "snapshot," e.g. a .bmp).

Specific Uncertainty Flags

• Assignee today. The Google Patents record says the current assignee is "Individual" (after the 2017 reassignments back to Richardson), while the 2021 Unified Patents contest page states the patent is owned by "Cine Graphic Solutions, an NPE." I could not verify a recorded assignment to Cine Graphic Solutions from the documents reviewed, so I am not treating that ownership statement as authoritative.
• 2026 CAFC activity. Not found (see top section). Absence of search evidence is not proof of absence.
• Claim text formatting. In the supplied full text, claim 15's ordering is garbled by the extraction (the "wherein" clauses for the first/sequential images appear mid‑claim). The substantive scope is as I summarized, but if exact claim language matters for an infringement or invalidity chart, pull the text from USPTO PatentCenter or the granted‑patent PDF rather than this rendered web version.

I'll research litigation involving US patent 8035644. Let me start with the sources you mentioned and the litigation links surfaced in the patent record.

Let me search for more specific case details on each identified litigation.

Let me dig deeper into the specific case dockets.

I was able to gather a good deal of grounded information, though I hit tool limits before confirming every case's final outcome. Below is what I can support with specific sources, with confidence levels flagged. Where I could not verify something, I say so rather than fill the gap.

Bottom line

Yes — US 8,035,644 has been litigated. Google Patents' own record for the patent carries five U.S. district court litigation links, and third-party databases (Stanford NPE Litigation Database, Unified Patents' PATROLL program) independently corroborate assertion activity. Two distinct assertion "waves" are visible:

1. 2017 — inventor-led suit by Douglas G. Richardson (the named inventor) against Samsung.
2. 2021–2022 — NPE wave by Cine Graphic Solutions LLC (an NPE; described by Unified Patents as a subsidiary of Strategic Intellectual Solutions, LLC) against Fujifilm, Motorola, and Samsung, asserting the '644 patent together with its sibling patents US 7,388,587 and US 7,629,977.

Known cases involving US 8,035,644

Sources for the five cases: Google Patents' litigation links for US 8,035,644 (patents.google.com/patent/US8035644/en) point to Unified Patents' litigation portal for each: .../Texas Eastern District Court/case/6:17-cv-00428, .../New York Southern District Court/case/1:21-cv-07917, .../New York Southern District Court/case/1:22-cv-04963, .../New York Southern District Court/case/1:22-cv-07114, and .../Illinois Northern District Court/case/1:21-cv-01840. Case #1's caption (Richardson v. Samsung) is corroborated by the Stanford NPE Litigation Database (npe.law.stanford.edu/patent/8035644), which also classifies the asserter as an "Individual." Case #2's caption, court, judge, and early docket entries are corroborated by a UniCourt docket record (unicourt.com, Cine Graphic Solutions LLC v. FUJIFILM Holdings America Corporation et al.).

Important context on the status field

I was not able to confirm the final outcome of any of the five cases within my tool budget. In particular:

• Whether case #1 (2017, E.D. Tex.) was dismissed, settled, or otherwise resolved — unconfirmed.
• Whether the 2021–2022 Cine Graphic Solutions cases settled, were dismissed, or went to judgment — unconfirmed.
• The party identities for cases #4 and #5 (both S.D.N.Y., 2022) — unconfirmed.

I would treat the "outcome/current status" column as open pending verification on PACER and the Unified Patents litigation caselist (https://portal.unifiedpatents.com/litigation/caselist). I am not going to assert a specific outcome I cannot ground in a source.

Related enforcement activity (not litigation, but relevant)

• Unified Patents PATROLL prior-art contest. On Dec 8, 2021, Unified Patents opened three contests seeking prior art on claim 1 of US 7,388,587, US 7,629,977, and US 8,035,644 — all owned by Cine Graphic Solutions. The '644 contest offered a $2,000 prize and ran through March 31, 2022; a winner was announced July 24, 2023. Unified's own announcement states the patents "have been asserted against Motorola, Samsung, and Fujifilm." (Sources: patroll.unifiedpatents.com/contests/HcLEbgtYaYeC863Bz; unifiedpatents.com/insights/2021/12/8/6000-for-cine-graphic-solutions-prior-art; unifiedpatents.com/insights/2023/7/24/...)
• Ownership trail matters for identifying current plaintiffs: the patent was assigned to EmailFilm Technology, Inc. (2012), then back to Douglas G. Richardson (2017), and Google Patents lists the current assignee as "Individual." Yet the 2021–2022 suits were brought by Cine Graphic Solutions LLC. So the case captions reflect different owners over time — which is corroborated by the "et al"/different-plaintiff pattern above.

Caveats and where to verify

• I could not directly pull the Unified Patents caselist page for patent 8035644; the five case links above come from the patent record and search snippets, not from my own retrieval of that portal.
• Case #3's defendant (Motorola) is a reasonable inference from the Illinois venue plus Unified Patents' statement that Motorola was sued — not something I confirmed for that specific docket number. Treat it as unverified.
• Because of the shared specification family, some litigation may have been brought on the sibling patents (US 7,388,587, US 7,629,977) or continuations (e.g., US 8,487,939) rather than on the '644 patent itself. I have listed only what is tied to 8,035,644.

Recommended authoritative sources to close the gaps: PACER dockets for the five case numbers above; the Unified Patents litigation caselist filtered to patent 8035644; and the USPTO Patent Center / assignment records for any transfers after 2017.

I'll verify the structured "no proceedings" baseline against public sources before finalizing.

Proceedings overview

Zero AIA trial proceedings. The structured ODP block reports no AIA trial proceedings on file for US 8,035,644, and independent web searching (PTAB E2E petition pages, Unified Patents litigation/PATROLL pages, NPE litigation databases, Law360/Docket Alarm coverage) surfaced no IPR, PGR, or CBM on the '644 patent — no institution decision, no FWD, no settlement, no appeal. Breakdown: 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denied. Bottom line for a defendant: all 21 claims stand untested at the PTAB. There is no IPR-based invalidity shortcut and, critically, no § 315(e)(2) estoppel against anyone — the entire § 102/§ 103 prior-art universe is still open to you in district court. Because the patent's adjusted expiration is 2027-05-04, an IPR (or ex parte reexam) filed now can still plausibly reach an FWD before expiry, so the absence of prior PTAB activity is not a reason to forgo filing one.

Proceedings

None. There is no proceeding number to report, and I will not synthesize one. Per the operating rule, the absence of proceedings is itself the finding.

Claim-level status of US 8,035,644

Because no claim has been canceled, there is nothing to quote from an FWD. Any statement that claims 1–5 (or any other claims) have been invalidated in an IPR would be fabricated.

Adjacent signals you should not mistake for PTAB proceedings

These are not AIA trials, but they are the reason the patent shows up in PTAB-adjacent searches:

1. Unified Patents PATROLL prior-art contest (defensive aggregator, 2021-12-08 → 2022-03-31). Unified ran a crowdsourced prior-art bounty on at least claim 1 of the '644 patent, alongside '587 and '977, for a total of $6,000 — Unified Patents announcement, 2021-12-08. The winner was announced 2023-07-24 (result); contest page: patroll.unifiedpatents.com/contests/HcLEbgtYaYeC863Bz. A PATROLL contest is prior-art sourcing, not a filed petition. The bounty closed with no IPR on the '644 docket — a notable signal that an aggregator scouted the patent and did not (or has not yet) pulled the trigger.

2. "Richardson '493" appearing in PTAB petition text is a false positive. Web results include PTAB petition excerpts (e.g., IPR2020-00869 re U.S. 8,531,834) citing "Richardson '493" (US 2006/0274493 A1) as prior art. That is an unrelated proceeding against a different patent. Do not cite it as a challenge to the '644 patent.

3. Litigation docket (context for why an IPR is still available to you). The '644 patent has been asserted repeatedly, but always in district court:

   • Richardson v. [[Samsung Electronics Co.](/litigations/by-defendant/Samsung%20Electronics%20Co.), Ltd.](/litigations/by-plaintiff/Samsung%20Electronics%20Co.%2C%20Ltd.) et al, No. 6:17-cv-00428 (E.D. Tex.) — Unified litigation record; Stanford NPE Database entry.
   • Cine Graphic Solutions LLC v. Fujifilm Holdings America Corp. et al. (S.D.N.Y.) — complaint filed 2021-09-27, asserting '587, '977 and '644.
   • Further S.D.N.Y. cases 1:21-cv-07917, 1:22-cv-04963, 1:22-cv-07114, and N.D. Ill. 1:21-cv-01840 (all per the Google Patents litigation links in the authoritative text).
   • The patent was also asserted against Motorola, Samsung, and Fujifilm according to Unified Patents.

4. Ownership churn. Inventor Douglas G. Richardson → EmailFilm Technology, Inc. (assignment recorded 2012-01-12, corrected 2017-02-21) → back to Richardson (2017-03-29 and 2017-06-15) — and Unified describes the '644 patent as having been "formerly owned by Cine Graphic Solutions, an NPE and a subsidiary of Strategic Intellectual Solutions, LLC." Google Patents currently lists the assignee as Individual. Ownership identity matters for IPR real-party-in-interest and privity analysis, and it also bears on whether a defendant can negotiate directly.

5. Prosecution/status quirks worth noting for a challenger. The patent lapsed for failure to pay maintenance fees and was reinstated by petition (2014–2015), then maintenance fees were paid through the 12th year (2023-06-28). Adjusted expiration 2027-05-04. A patent with a lapse/reinstatement history is a common target for § 112/§ 101-adjacent attacks but, more relevantly, it remains live and enforceable — there is no invalid-claim shield.

Strategic summary

Claim status. Every one of claims 1–21 of US 8,035,644 is untested rather than sustained: no AIA tribunal has ever construed a term, ruled on a ground, or canceled a claim. Independent claim 1 (method), independent claim 8 (electronic message), and independent claim 15 (computer programming product) — and all 18 dependents — are all fair game. Contrast this with the usual "hardened patent" posture after surviving two IPRs: that does not describe this patent. This patent is un-hardened because untested, which is materially better for a defendant than a patent with a clean PTAB win record.

Estoppel landscape. Because there has never been an IPR/PGR/CBM, § 315(e)(2) estoppel is a null set — no petitioner, no privy, and no "grounds raised or reasonably could have been raised." There is also no preclusion flowing from any ex parte reexamination that I could locate; the structured ODP block covers AIA trials, and my searches turned up no reexam certificate. Practical consequence: you may raise any § 102/§ 103 ground in district court and simultaneously file an IPR on the same or overlapping art without estoppel risk to your litigation defenses (subject to § 315(b)'s one-year bar from service of an infringement complaint, which runs per-defendant and is worth checking immediately — the older 2017 and 2021–22 cases make timing urgent for any defendant named in those suits).

Pattern signals. (a) No repeat petitioner exists, because there is no petitioner at all. (b) The patent owner has never appeared before the PTAB — no appeals to the Federal Circuit from any PTAB case, and no IPR defense posture has been tested; you have no read on how Patent Owner litigates claim construction at the Board. (c) A defensive aggregator did scout the patent: Unified Patents ran a $2,000 PATROLL bounty on at least claim 1 and announced a winner, but did not file. That means crowdsourced invalidity art and claim charts already exist and are public (APEX Standards claim-charting, Limestone validity report, Ambercite/Amplified leads, all linked from the PATROLL page). That is a head start on an IPR petition — but note PATROLL rules excluded art previously cited during prosecution, so the crowdsourced leads are additive to, not a substitute for, the examination record. (d) The family is large and continuing: US 7,388,587, US 7,629,977, US 8,487,939, US 11,232,768, US 11,263,998, US 11,893,965, US 12,051,391, plus pending US 2025/0054465. Assertion of the '644 patent has historically come with the '587 and '977 siblings, so your freedom-to-operate and invalidity analysis should cover the family, not just the '644.

Recommended next steps

• Treat the "no PTAB activity" result as verified-but-verify. I found no proceedings in public sources; I could not query USPTO PTAB E2E or Docket Alarm directly from this environment. Before relying on this for a filing decision, confirm against PTAB E2E and the USPTO Patent Trial and Appeal Board End-to-End docket for the '644 patent, and re-check for any recently filed petition the ODP ingest may not yet reflect. If a petition has been filed since this ingest, everything in the estoppel section changes.
• Calendar the real deadline, not a PTAB one. There is no institution deadline, oral hearing, or 1-year FWD clock to track because no trial exists. The binding dates are (i) § 315(b)'s one-year bar running from the date your client was served with an infringement complaint on the '644 patent, and (ii) the patent's 2027-05-04 adjusted expiration — an IPR filed after roughly mid-2026 risks not reaching an FWD before the patent dies, which may moot the practical benefit.
• Build the petition from the PATROLL work product. The Unified/APEX/Limestone/Ambercite charting is public and targeted at claim 1. Pair it with art not cited during prosecution (the '644 file history cites only the two Richardson family patents, per the "Patent Citations" list in the authoritative text — a remarkably thin record for a 2011 patent, which is itself an invalidity opportunity).
• Consider a § 101/§ 112 flank as well. Claims 1/8/15 are directed to the familiar "capture, clip, layer, save as web-enabled file" workflow; the written description's own emphasis on file-extension stripping and browser-save behavior (see the '644 description at the "Save Mod. Image" block) may support indefiniteness or enablement attacks that an IPR cannot reach.
• Do not tell your client claims 1–5 are dead. They are not. Any demand letter citing the '644 patent must be answered on the merits (invalidity, non-infringement, damages), not by pointing to a PTAB cancellation that does not exist.

Confidence statement: I have high confidence that no AIA trial proceeding has been indexed for US 8,035,644 as of the ODP ingest, and that my searches found none. I cannot rule out a very recent, not-yet-indexed filing. I found no Federal Circuit appeal, no FWD, no settlement, and no judge panel to report — and I have deliberately not invented any.

I'll start by trying to pull the assignment records and the litigation context from live sources.

Key leads surfaced. Let me dig into the Cine Graphic Solutions chain and the litigation record.

I could not reach the USPTO Assignment Center API directly from this session, so the chain below is reconstructed from the USPTO-derived legal-events record reproduced on the Google Patents family page for US 8,035,644 (which carries recording dates, effective dates and reel/frame numbers keyed to USPTO assignment records) plus third-party verification. Full reel/frame verification in Assignment Center is still required for two legs of the chain — flagged below.

Inventors

• The '644 patent is a continuation-in-part of US 11/586,016 (filed 2006-10-25, issued as US 7,388,587), which descends from US 11/403,374 (filed 2006-04-12, issued as US 7,629,977), which claims priority to provisional US 60/670,402 (filed 2005-04-12). All four filings name Richardson alone, and none names a corporate assignee at filing. Google Patents lists "Original Assignee: Individual" for US 8,035,644.
• Unusual pattern worth noting: the patent was filed 2008-06-16 and did not leave Richardson's hands until 2012-01-12 — a 3.5-year gap. The company that eventually took title, EmailFilm Technologies, Inc. (Austin, TX), was founded in 2006 (i.e., after the 2005 provisional). This is the classic "founder holds IP personally, then papers it into the funded entity" sequence, which is a common precursor to a later portfolio monetization. There is no inventor-departure signal here because there is only one inventor and he is the founder — so the usual "all inventors left within 12 months" screen does not apply.
• The same inventor is the named inventor on the entire family: US 7,388,587; US 7,629,977; US 8,035,644; US 8,487,939; US 11,232,768; US 11,263,998; US 11,893,965; US 12,051,391; US 2025/0054465.

Original assignee

Douglas G. Richardson, an individual (Dripping Springs, TX). US 8,035,644 issued 2011-10-11 with no corporate assignee of record. The first corporate owner — EmailFilm Technology, Inc. / EmailFilm Technologies, Inc. (Austin, TX) — did not take title until the 2012-01-12 assignment (reel 027525/0231), i.e. after issuance.

• Line of business: software enabling full-content video to be HTML-compatible and embedded in email. Founded 2006. Per the Austin Business Journal (Sept. 2013), EmailFilm Technologies raised $234,483 of a planned $500,500 from six investors per an SEC filing, and its listed CEO was Graham McFarland, simultaneously CEO of Austin-based Cinegif. (Austin Business Journal)
• Product embodying the claims: the company publicly described a product concept (HTML-compatible video-in-email), but I found no evidence of a commercially shipped product in any source reviewed. Small-entity maintenance fees were paid at the 4th, 8th and 12th years — with a lapse-and-reinstatement at the 4th year (2015-10-11 lapse; reinstated 2015-11-17) — which is consistent with a thinly capitalized owner, not a product company with revenue.
• Current status: not determinable from the sources reviewed. The corporate entity still appears as applicant/assignee on later family members (US 8,487,939; US 11,232,768 lists "Applicant: Emailfilm Technology, Inc., Austin, TX"), which means it was still filing as an entity years later. I found no bankruptcy, dissolution or acquisition record. Do not treat "defunct" as established — it is not.
• Family split (important): the 2016/2017 recordings moved the early-generation patents (the '587 / '977 / '644 set) back out of the company, while the later continuations stayed with the company. That asymmetry is the single most informative structural fact in this file.

Assignment timeline

Correspondent of record: NOT AVAILABLE. The USPTO-derived legal-events data I retrieved exposes assignor, assignee, reel/frame, execution date, recording date and conveyance text — but not the correspondent/recording attorney. I am not going to invent one. The only attorney name connected to this family in any source I found is Gregory K. Goshorn (Greg Goshorn, P.C.), listed as attorney, agent or firm on the face of US 11,232,768 — that is prosecution counsel, not assignment correspondent, and a single appearance is not the recurrence signal. Verify the correspondent fields in Assignment Center.

1. Executed 2012-01-12 / recorded 2012-01-12 — Reel 027525/0231

• Conveyance: Assignment (of assignors' interest)
• Assignor: RICHARDSON, DOUGLAS G.
• Assignee: EMAILFILM TECHNOLOGY, INC. (Texas)
• Correspondent: not exposed in the record I retrieved
• Context: Founder contributes personally-held patent rights into his own newly-funded operating company (the 3.5-year delay between filing and this recording is the tell).

2. Executed 2012-01-01 / recorded 2017-02-21 — Reel 041759/0442

• Conveyance: Corrective Assignment ("to correct the name of assignee previously recorded on reel 027525 frame 0232")
• Assignor: RICHARDSON, DOUGLAS G.
• Assignee: EMAILFILM TECHNOLOGIES, INC. (Texas)
• Correspondent: not exposed in the record I retrieved
• Context: Change of name only / record correction — the "Technology" → "Technologies" fix. Note it was filed five years after the original, in the same five-month burst as the reversions below.
• ⚠️ Literal-reading discrepancy to preserve: the original was recorded at 027525/0231, but the corrective assignment recites the earlier record as 027525/0232. One frame off. Do not auto-correct — verify in Assignment Center.

3. Executed 2016-12-19 / recorded 2017-03-29 — Reel 041778/0309

• Conveyance: Assignment (of assignors' interest)
• Assignor: EMAILFILM TECHNOLOGIES, INC.
• Assignee: RICHARDSON, DOUGLAS G. (Texas)
• Correspondent: not exposed in the record I retrieved
• Context: Reversion — company transfers rights back to the individual founder.

4. Executed 2017-06-14 / recorded 2017-06-15 — Reel 042721/0425

• Conveyance: Assignment (of assignors' interest)
• Assignor: EMAILFILM TECHNOLOGIES, INC.
• Assignee: RICHARDSON, DOUGLAS G. (Texas)
• Correspondent: not exposed in the record I retrieved
• Context: Second reversion — a second, separate EmailFilm → Richardson transfer on a different effective date, recorded the day after execution.

5. Unrecorded-on-this-page leg — Richardson → CINE GRAPHIC SOLUTIONS (by 2021)

• No reel/frame found. This leg does not appear in the legal-events record for US 8,035,644 as reproduced on Google Patents (which runs through 2023-06-28). Yet Unified Patents states, twice, that Cine Graphic Solutions owned the '644 patent — "[o]wned by Cine Graphic Solutions, an NPE and a subsidiary of Strategic Intellectual Solutions, LLC" (2021-12-08) and "formerly owned by Cine Graphic Solutions, an NPE and a subsidiary of Strategic Intellectual Solutions, LLC" (2023-07-24). (Unified, 2021-12-08; Unified, 2023-07-24)
• Treat this leg as confirmed-as-fact but not confirmed-as-recorded. Possible explanations: (a) an assignment recorded against a different family member (e.g. the '587 or '977), (b) an exclusive license / co-plaintiff arrangement rather than a title transfer, or (c) a title transfer recorded in a reel the Google Patents legal-events feed does not carry. Resolve this in Assignment Center before relying on it.

6. Apparent unwinding — back to Richardson (by 2023)

• Google Patents lists "Current Assignee: Individual" for US 8,035,644, and Unified's July 2023 post describes the patent as currently "owned by Douglas G. Richardson." So the litigation vehicle appears to have been unwound after the 2021–2022 campaign. Again, no reel/frame located.

Bottom line on completeness: four assignments are firmly documented with reel/frame (027525/0231, 041759/0442, 041778/0309, 042721/0425). Two additional legs — the transfer into Cine Graphic Solutions and the transfer back out — are each supported by third-party (Unified Patents) statements and by the litigation record, but I could not locate a recording for either.

Timeline diagram

timeline
    title Ownership of US 8035644
    2005 : Provisional filed by Richardson
    2008 : CIP application filed by Richardson
    2011 : Patent issued with no corporate assignee
    2012 : Assigned to EmailFilm Technology Inc
    2016 : Reassigned back to Douglas G Richardson
    2017 : Second reversion to Richardson
         : Suit filed against Samsung in E D Tex
    2021 : Cine Graphic Solutions asserts the patent
         : Suits filed against Fujifilm and others
    2023 : Unified lists patent as owned by Richardson

NPE / troll-pattern signals

1. Shell-entity transfer — PRESENT.
The terminal asserting entity is Cine Graphic Solutions, expressly characterized by Unified Patents as "an NPE and a subsidiary of Strategic Intellectual Solutions, LLC" (2021-12-08 and 2023-07-24 posts). The parent name carries the "Intellectual" holding-company marker and a "Solutions, LLC" suffix. Critically, this is not a naming inference — Unified Patents is a defensive aggregator that designates asserters after reviewing the entity, and it ran a PATROLL prior-art bounty against three Cine Graphic patents ('587, '977, '644) on 2021-12-08. The reversion cluster at reels 041778/0309 and 042721/0425 — pulling the early-generation patents out of the company and back to the individual — is the transfer that precedes the asserter vehicle. I found no products in commerce attributable to Cine Graphic Solutions.

2. Known asserter in the chain — PRESENT (via Unified Patents discovery, not the enumerated list).
Cine Graphic Solutions is not on the enumerated list supplied (Acacia, Marathon, IV, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, Document Generation Corp, Spangenberg entities). But the instruction extends the check to "any entity surfaced by Unified Patents or RPX as a high-frequency plaintiff." Unified surfaced it, as an NPE, in a paid prior-art contest. Strategic Intellectual Solutions, LLC is the named parent and should be run through the Unified and RPX directories as a separate asserter.

3. Repeat correspondent across the chain — UNCLEAR / INSUFFICIENT DATA.
I could not retrieve the recorded correspondent for any of the four reel/frame entries. Without that field I cannot run the recurrence test, and I will not substitute a prosecution attorney for it. The only name available — Gregory K. Goshorn, Greg Goshorn, P.C., attorney of record on US 11,232,768 — is prosecution counsel on a family member, appears once, and therefore is not a finding under the stated rule. Action item: pull correspondent fields for 027525/0231, 041759/0442, 041778/0309, 042721/0425 and for the Cine Graphic leg.

4. Cascading transfers — PRESENT in a weak, non-LLC form.
Three recordings landed in a five-month window: 2017-02-21 (reel 041759/0442), 2017-03-29 (reel 041778/0309) and 2017-06-15 (reel 042721/0425) — all between the same two parties, including a name-correction five years after the fact and two separate reversions on different effective dates (2016-12-19 and 2017-06-14). That is a paperwork burst, not a chain of single-purpose LLCs, so the canonical "chained LLCs in <24 months" form is not present. The candidate cascade is instead the June 2017 reversion followed within weeks by the Cine Graphic campaign launch years later.

5. Pre-litigation transfer — PRESENT.
The second reversion executed 2017-06-14 and was recorded 2017-06-15 (reel 042721/0425). The first plaintiff-side suit on this patent family is Richardson v. [[Samsung Electronics Co.](/litigations/by-defendant/Samsung%20Electronics%20Co.), Ltd.](/litigations/by-plaintiff/Samsung%20Electronics%20Co.%2C%20Ltd.) et al, 6:17-cv-00428 (E.D. Tex.), listed in the Stanford NPE Litigation Database under patent 8035644 with asserter category "Individual." A 6:17-cv-00428 docket number implies a mid-2017 E.D. Tex. filing — i.e. weeks after the reversion. I could not confirm the exact complaint date, so treat the sub-6-month window as strongly indicated but unverified. The 2021 Cine Graphic suits (e.g. Cine Graphic Solutions LLC v. FUJIFILM Holdings America Corp. et al, S.D.N.Y., complaint filed 2021-09-28, asserting the '587, '977 and '644 patents) fit the same template at the second hop. (NPE Litigation Database; UniCourt docket)

6. Bankruptcy fire-sale — NOT PRESENT.
No Chapter 7/11 proceeding, no trustee sale, no secured-party or security-interest recording appears anywhere in the chain. EmailFilm Technologies' 2013 raise was small ($234,483 of $500,500), but a small raise is not a bankruptcy, and I will not characterize it as one.

7. Privateering — NOT PRESENT.
The transfer runs inventor/individual → NPE, not operating-company → NPE. There is no operating company with market competitors handing patents to an asserter to sue on its behalf; the asserted defendants (Motorola, Samsung, Fujifilm) are third-party consumer-electronics makers with no relationship to the patentee.

8. Defensive aggregator — NOT PRESENT.
No RPX, AST, LOT Network, Unified Patents or OIN acquisition anywhere in the chain. The opposite is true: Unified Patents attacked this patent via PATROLL and paid $2,000 for winning prior art (awarded 2023-07-24), which is the inverse of the defensive-aggregation signal.

Data-quality flag: the Google Patents page for US 8,035,644 lists three "Family Cites Families" entries — US 4,984,302, IL 105930 A and US 5,392,773 — which are nose-worn air filters and therapeutic respiration devices. These are unrelated to this family and are a feed artifact, not a licensing or corporate trail. Do not read them as evidence of any transfer.

Verdict

NPE — high confidence.

Two independent strong signals, both grounded in specific records. First, the patent's title was deliberately pulled out of the operating company and back into the individual inventor by the assignments executed 2016-12-19 (recorded 2017-03-29, reel 041778/0309) and 2017-06-14 (recorded 2017-06-15, reel 042721/0425) — a reversion that lands inside the six-month window before the 2017 E.D. Tex. assertion campaign Richardson v. Samsung, 6:17-cv-00428 (per the Stanford NPE Litigation Database, pat. 8035644). Second, the entity that then asserted the patent against Motorola, Samsung and Fujifilm — Cine Graphic Solutions — is expressly identified by Unified Patents as an NPE and a subsidiary of Strategic Intellectual Solutions, LLC (Unified posts of 2021-12-08 and 2023-07-24, the latter after a $2,000 PATROLL prior-art bounty on this patent). No products, no defensive aggregator, and no bankruptcy anywhere in the chain; instead the assertion ran against third-party consumer-electronics manufacturers with no relationship to the patentee.

Caveat on completeness: the transfer into Cine Graphic Solutions and the apparent later transfer back out to Richardson are each supported only by Unified Patents' statements and the litigation record — I found no reel/frame for either leg, and no correspondent field for any recorded entry. Verify both in the USPTO Assignment Center (search by patent number 8035644, and separately by assignor "Richardson, Douglas G." and assignee "Cine Graphic Solutions" / "Strategic Intellectual Solutions") before relying on this chain for standing or priority purposes.

I'll ground this in the authoritative record and verify the cited references before analyzing anticipation.

Prior‑Art Analysis — US 8,035,644 B2

Scope note. Google Patents lists this patent as being cited against exactly two earlier US patents, both naming the same sole inventor, Douglas G. Richardson. I treat that list as authoritative for this patent. Below I map each citation, then explain why neither is available as § 102 anticipation art in the ordinary sense, and finally list the third‑party art that the Patent Office did not cite but that third‑party searchers rank as most relevant.

Sources used: the authoritative full text supplied in the prompt (https://patents.google.com/patent/US8035644/en), the citation list corroborated at Justia (https://companyprofiles.justatic.com/patent/[8035644](/patent/8035644)) and Patexia (https://patexia.com/us/publication/20080259167); the Google Patents pages for the two cited patents; the Unified Patents/Cine Graphic Solutions Patroll contest page (https://patroll.unifiedpatents.com/contests/HcLEbgtYaYeC863Bz); and the Limestone validity report linked from that contest.

1. The cited references — the complete list on the face of US 8,035,644

Google Patents shows "Patent Citations (2)" and "Citations (1)" for US 8,035,644. The Justia record independently lists the same two "Referenced Cited" documents:

That is the entire citation list relevant to this patent. No third‑party patent or publication was cited by the examiner on this patent.

2. Reference 1 — US 7,388,587 B1

Full citation. Douglas G. Richardson, "Method for Embedding Animation in Electronic Mail and Websites," U.S. Patent 7,388,587 B1, application No. 11/586,016, granted Jun. 17, 2008.

Dates. Application filing date Oct. 25, 2006 per the patent's own file‑wrapper data (Google Patents lists 2006‑10‑25; the cross‑reference paragraph inside US 8,035,644 recites "filed Nov. 25, 2006," and Unified Patents shows 2006‑10‑24 — a three‑way discrepancy in the record that I am flagging rather than resolving). Grant date Jun. 17, 2008. Earliest priority Apr. 12, 2005 (provisional 60/670,402).

Brief description. This is the near‑identical disclosure to 8,035,644 minus the "pixel‑block sensitivity" wording later added to the claims of 8,035,644. It describes capturing a sequence of photographs from a fixed camera position; designating the first as the "naked photo"; editing subsequent photos to clip the moving element; pasting the clipped images into the naked photo as layers positioned where the element appeared; and saving the modified naked photo (base image + layers) as a web‑enabled graphic file embedded in an email or web page. Deletion of the file extension so that a browser can only save a snapshot is described in the § 5 specification text shared with 8,035,644.

Claim mapping / what it could be used against. The reference discloses or closely tracks the subject matter of claims 1, 3, 6, 8, 10, 13, 15 and the email/website dependents (4, 5, 7, 11, 12, 14, 19, 20, 21). It is the single closest document to the entire claim set, because 8,035,644 is by definition a CIP of it and the two share figures 1–7 and the Animation Setup / Display Animation / Edit Image flowcharts.

§ 102 assessment. As a matter of law, US 7,388,587 B1 cannot anticipate any claim of US 8,035,644 under pre‑AIA § 102(a) or § 102(e). Both patents name the same sole inventive entity, Douglas G. Richardson. Pre‑AIA § 102(a) requires the invention to have been "known or used by others," and pre‑AIA § 102(e) requires a patent "granted on an application for patent by another." Same‑inventor work is not "another." The cited reference is therefore best understood as being on the face of the patent for continuity/disclosure reasons (it is the CIP parent whose disclosure is carried forward), or as potential support for an obviousness‑type double patenting concern between family members — not as § 102 anticipation art. If one insisted on charting it, the only claim limitation the parent's text arguably does not carry in the same terms is claim 1's "defining a sensitivity level corresponding to pixel blocks," which is the limitation added to 8,035,644 relative to its own pre‑grant publication US 2008/0259167 A1 (see § 5 below).

3. Reference 2 — US 7,629,977 B1

Full citation. Douglas G. Richardson, "Embedding Animation in Electronic Mail and Websites," U.S. Patent 7,629,977 B1, application No. 11/403,374, granted Dec. 8, 2009.

Dates. Application filing date Apr. 12, 2006; grant date Dec. 8, 2009; earliest priority Apr. 12, 2005 (provisional 60/670,402).

Brief description. The original non‑provisional in the chain. It discloses the same core technique — fixed‑position photograph sequence, "naked photo" first frame, clipping the element that moves between frames, layering the clips onto the naked photo at their original coordinates, and transmitting the composite as a web‑enabled graphic file in email or on a website — with the layering/positioning and display‑and‑remove teaching that recurs in all later family members. Its own front page carries non‑patent‑literature citations in the same technical space (e.g., Schödl et al., "Video Textures," SIGGRAPH 2000; Freeman et al., "Motion Without Movement," 1991; Bhat et al., "Flow‑Based Video Synthesis and Editing," ACM 2004; Sun et al., "Video Input Driven Animation (VIDA)," ICCV 2003; Shade et al., "Layered Depth Images," SIGGRAPH 98; Wallace, "Merging and Transformation of Raster Images for Cartoon Animation," 1981) — these are the technical antecedents of the "layer a moving element over a static base frame" idea.

Claim mapping / what it could be used against. Discloses the foundational elements of claims 1, 8 and 15 — sequential capture from a particular location, differing portions, cutting, layering in corresponding position, time sequence, and saving as a web‑enabled graphic file — and the email/website/HTML dependents (4, 5, 7, 11, 12, 14, 19, 20, 21). Less complete than 7,388,587 on the "naked photo"/sensitivity discussion because the later filings added that text.

§ 102 assessment. Same answer as Reference 1, and for the same reason: identical sole inventive entity, so not "by another" under pre‑AIA § 102(a)/(e). It is a family member (grandparent), not third‑party prior art.

Common point for both references: US 8,035,644 has an effective filing date (Apr. 12, 2005 priority; Jun. 16, 2008 actual filing) before the AIA's Mar. 16, 2013 first‑inventor‑to‑file cutoff, so pre‑AIA § 102 governs. Under pre‑AIA law the examiner citations cannot be anticipatory art against this patent.

4. References that appear on the record but are not prior art — flag as erroneous data

Google Patents shows a "Family Cites Families (3)" group containing:

• US 4,984,302 A — Robert A. Lincoln, "Nose‑worn air filter," priority 1987‑03‑20, published 1991‑01‑15
• IL 105930 A — Natan Oren, "Therapeutic respiration device," 1993‑06‑07
• US 5,392,773 A — Archie A. Bertrand, "Respiratory particulate filter," priority 1994‑04‑13, published 1995‑02‑28

These three documents are respiratory‑device/face‑mask patents with no subject‑matter relationship whatsoever to image animation, and none is cited against any claim of 8,035,644. I read this as a data‑attribution error in the rendered record (a mis‑joined family/classification identifier), not as a real citation group, and I am not treating any of them as prior art. Flagging explicitly because the instruction is to be literal about the record: they are literally present under that heading, yet they cannot be prior art on their face.

Likewise I discarded the International Search Report fragments that surfaced in search (WO2017105742A1 citing US 2009/0077326 A1, and the JP 2015‑510848 ISR) — those belong to unrelated applications and are not citations of US 8,035,644.

5. Internal record that bears on the citation history (not prior art, but important)

The pre‑grant publication US 2008/0259167 A1 (Oct. 23, 2008) states claim 1 as "defining a sensitivity level; identifying portions of the sequential images that differ from the first image to a degree corresponding to the sensitivity level." The granted claim 1 reads "defining a sensitivity level corresponding to pixel blocks of the plurality of images." The "pixel blocks" limitation was therefore added during prosecution. The specification text explaining that parameter (Set Parameters block 303: "the size of pixel blocks that are compared, i.e. the granularity, the degree of difference in compared pixel blocks…") is shared with the parent 7,388,587. Inference (flagged as such): the parent was likely cited as the closest disclosure of the pixel‑block sensitivity concept, which is consistent with claim 1 being amended to tie the sensitivity level to pixel blocks. The record I have does not show the examiner's stated grounds.

6. The art that is actually being used against this patent family — third‑party leads (not examiner citations)

Because the only examiner citations are the inventor's own family members, the substantive invalidity art comes from outside the record. The Unified Patents Patroll contest for US 8,035,644 (Cine Graphic Solutions) generated third‑party searches, and the Limestone validity report linked from that page ranks candidates. Highest‑scoring items (score = algorithmic claim‑relevance, not a verified anticipation finding):

Report URL: http://s3.amazonaws.com/report.limestoneip.com/k2jK2Dqh1-6rSNE1AVo63h.full.html
Contest URL (which also links APEX Standards claim charts, Amplified results and Ambercite similar‑art lists): https://patroll.unifiedpatents.com/contests/HcLEbgtYaYeC863Bz

These are leads for a validity chart, not confirmed § 102 references. In particular, the "sensitivity level corresponding to pixel blocks" limitation (claim 1) and the "single, web‑enabled graphic file … without a file extension" limitation (claims 9/17) are the elements most likely to require combination rather than single‑reference anticipation.

7. Bottom line

1. The only two citations on US 8,035,644 are US 7,388,587 B1 and US 7,629,977 B1, both Richardson's own earlier patents in the same CIP chain. No third‑party art was cited by the examiner.
2. Neither cited reference can anticipate any claim under pre‑AIA § 102, because both are the same sole inventive entity as the patent‑in‑suit and therefore are not "by another" (§ 102(a): "by others"; § 102(e): "by another"). Their proper role is continuity/disclosure support (and potentially obviousness‑type double patenting between family members).
3. If a party needs § 102 art against 8,035,644, it must come from third‑party references such as those in the Limestone/Patroll set (§ 6), which would need to be independently verified against the granted claim language — particularly the pixel‑block sensitivity limitation of claims 1, 7 and 15 and the no‑extension file‑name limitation of claims 9 and 17.
4. Contradiction flagged against the earlier summary: that summary listed the cited prior art as "US 7,388,587 B1; US 7,629,977 B1." This full re‑check confirms it, and adds that the "Family Cites Families" trio (US 4,984,302, IL 105930, US 5,392,773 — nose‑worn air filter, therapeutic respiration device, respiratory particulate filter) should not be carried forward as prior art; those are unrelated respiratory‑device documents appearing in the rendered record under an apparent data‑mapping error.

Uncertainty flags. (a) I could not run a direct USPTO PatentCenter/Examiner's‑search‑strategy query; the citation list is taken from the authoritative Google Patents full text corroborated by Justia and Patexia. (b) The filing date of 11/586,016 is recorded inconsistently (2006‑10‑24, 2006‑10‑25, or "Nov. 25, 2006") across sources; I have not resolved it. (c) The Limestone scores are relevance rankings, not legal conclusions.

I'll research the prior art landscape for this patent, starting with the references actually listed on the record page and then the broader field of animation/image-compositing art.

Now let me research the specific prior-art references cited in this patent family and the invalidity work already done on it.

Obviousness Analysis — US 8,035,644 B2 under 35 U.S.C. § 103

0. A note on the "Prior Art" section of this page, and how I handled it

The record for US 8,035,644 lists a very thin backward-citation set. Literally, the "Prior Art"/Citations area shows only:

• US 7,388,587 B1 (Richardson) — cited by examiner
• US 7,629,977 B1 (Richardson) — cited by examiner
• Forward-only citations: WO 2012/038924 A2 (NDS/Cisco) and JP 2013-162221 A (Sony) — both post-date the 2005 priority and are therefore not prior art.

That list cannot support a real § 103 analysis, so — per your instruction to use the page's prior-art results but flag gaps — I supplement it with (a) the examiner-cited art of record on this patent family's later continuations and (b) well-known art that a POSITA would have known. Every reference below is given with its source and a confidence flag. I was unable to retrieve the specific winning reference from the Unified Patents PATROLL contest on this patent (contest ended 2022‑03‑31; winner announced 2023‑07‑24) — I do not know what that art was, and I won't guess.

Identifiers are reproduced literally as they appear; I have not "corrected" any patent number.

1. Governing framework

Statute: Because the application (12/140,060, filed 2008‑06‑16) has an effective filing date before 16 March 2013, pre‑AIA § 103 applies. Priority date: 2005‑04‑12 (provisional 60/670,402). Art must therefore predate 2005‑04‑12, or qualify under pre‑AIA § 102(e).

POSITA: A person with a bachelor's degree in CS/EE (or equivalent) and ~2–3 years' experience in digital imaging, image editing/compositing, and web/email content delivery — i.e., someone versed in Photoshop-style layered raster editing, motion/change detection, and web image formats (GIF/JPEG/HTML). This is a low-skill bar; the specification itself concedes that "those with skill in the computing arts should be familiar with various graphic and web‑enabling formats for images such as gif, jpeg and html."

Standard: KSR Int'l v. Teleflex, 550 U.S. 398 (2007) — a claimed combination is obvious where the elements were individually known and combined in a predictable manner with no unexpected result; "a court must ask whether the improvement is more than the predictable use of prior art elements according to their established functions." Also Graham v. John Deere.

Critical threshold point — the Richardson references are not valid § 103 art. US 7,388,587 B1 and US 7,629,977 B1 name the same sole inventor, Douglas G. Richardson, and claim the same 2005‑04‑12 priority. Under pre‑AIA § 102(e), a reference must be "by another"; same inventive entity is not "another." They also cannot be § 102(a)/(b) art because their public dates (2008, 2009) postdate 2005. And even if one argued the '644 claims contain new matter (e.g., the "sensitivity level" limitation) pushing their effective date later, § 103(c) common-inventorship/ownership would still defeat their use in an obviousness combination. Conclusion: the only two references the page actually lists carry essentially zero § 103 weight against these claims. Any viable invalidity case must rest on third-party art — which is exactly why the real-world invalidity effort (the PATROLL contest) targeted third-party art.

2. The prior-art corpus (with sources and confidence)

3. Element-by-element mapping — Claim 1

Only two limitations even arguably lack a single-reference hit: (c) the pixel-block "sensitivity level" and (h) the "single, web-enabled graphic file." Both are supplied by the secondary references above, and both are supplied with a clean motivation (see §5).

4. Proposed § 103 grounds

Ground 1 (primary): Epshteyn '833 in view of Qian/Elgammal, further in view of GIF89a (and/or Chen '278).
Epshteyn discloses every step of the process except the express "pixel-block sensitivity level" formulation and the "single web-enabled graphic file" wrapper. Qian (incorporated into Epshteyn by reference) and Elgammal supply thresholded pixel-block differencing; GIF89a (or Chen's single network-transmitted animation object) supplies one browser-native file that both stores and plays base + changing portions. Rationale under KSR: combining is "the mere application of a known technique (thresholded background subtraction) to a known structure (an Epshteyn-style animated photograph), with a predictable result," and GIF is simply the known web-native container.

Ground 2: Chen '278 in view of GIF89a/GIF Movie Gear and a segmentation reference.
Chen discloses the file architecture — a single animation object containing a background track plus object tracks at multiple resolutions, explicitly for low-bandwidth (Internet) delivery — i.e., claim 1(f)–(h). Combining with pixel-block change detection (Elgammal/Kondo/Matsushita) supplies (c)–(e). Combined result is the predictable union of the two references' functions.

Ground 3: Animated-GIF practice (GIF89a + GIF Movie Gear) in view of Epshteyn/Chen.
This ground emphasizes that the "web-enabled graphic file" limitation was, by the mid-1990s, an off-the-shelf, plug-in-free, in-email playable format, and that composing sequentials into one GIF by cutting/pasting regions over a base frame was a routine Photoshop/Movie-Gear workflow. This is also the ground that most directly kills dependent claims 6 and 13 (GIF89a's "restore to background" disposal method literally "remove[s] each layer from the first image prior to displaying a next layer").*

Ground 4 (for the file-name limitation, claims 9/17): any of the above in view of the routine practice of renaming/deleting a file extension.
No reference needed beyond general knowledge: stripping or altering a filename extension is a predictable, purely conventional variation with no unexpected result — the express KSR "predictable variation of a known element" rationale. The specification's stated purpose (to prevent a recipient saving the animation as a .bmp snapshot) is a desired-result argument with no structural or functional change to the claimed file, which is fatal to patentability.

5. Motivation to combine (why a POSITA would have done this)

1. Same field, same problem. Epshteyn, Chen, the segmentation literature, and GIF authoring tools all address small-file transmission/playback of animation — the sole problem the '644 Background identifies ("What is needed is an efficient method for transmitting dynamic images … in small files such as email messages so that download times are minimized").
2. Known, finite technique set. By 2005 the field offered a finite menu: (i) full motion video — bandwidth-heavy; (ii) Flash — needs a plug-in (the spec's own admitted drawback); (iii) layered/composited "moving object over static background" animation — Epshteyn/Chen; (iv) format = GIF/JPEG/HTML. KSR: choosing among a finite number of identified, predictable solutions.
3. Express design incentive. Epshteyn's own text states the base image should be high-resolution "since most of the time this image will be displayed" and the motion recorded at inferior quality — the very "high-res base + lower-res change layers" architecture the '644 specification claims as an embodiment. This is a direct teaching of the "smaller file" advantage.
4. GIF is the plug-in-free container that solves the stated Flash problem. The spec's stated motivation (avoid Flash's plug-in requirement) points straight at GIF89a, which needed no plug-in. Where the motivation is supplied by the reference's own disclosure of the problem, that is textbook KSR rationale (a).
5. No teaching away. Nothing in Epshteyn or Chen disparages single-file web GIFs; Chen affirmatively targets Internet delivery.

6. Claim-by-claim disposition

7. Where the patent owner has its best (and weakest) arguments

Best surviving arguments (limited):

• "Sensitivity level corresponding to pixel blocks." Patent Owner may argue the art discloses segmentation generically but not the configured triple parameterization (block granularity + degree-of-difference + degree-of-movement calc). This would turn on whether the limitation is given patentable weight or treated as an inessential design choice — the reference to "pixels corresponding to automobile 172" at "low sensitivity" suggests it is a routine tunable, which is exactly the sort of "result-effective variable" optimization that is obvious. Likely to fail, but this is the narrowest point.
• "Single, web-enabled graphic file." If construed to require the base image and the change layers flattened into one GIF (as the '644 describes), GIF89a/Movie Gear defeat it directly. Likely to fail.
• Claim differentiation via the Adobe construction. In Richardson v. Adobe (S.D.N.Y.), the court held an "electronic message" (claim 8) must be something distinct from the "web-enabled graphic file." That construction actually helps an obviousness attack on claim 8: the art needs only (i) an email/web page and (ii) a web graphic file inside it — both bone-standard.

Weakest counterargument defense: the "reverse engineering / protect against saving as .bmp" rationale (claims 9/17). Purpose-based limitations without a structural difference do not confer patentability.

What I cannot currently evaluate: (i) the specific winning PATROLL art — if it exists, it may be stronger than any ground above; (ii) the exact claim construction of "sensitivity level" and "corresponding to the timing," which govern whether Ground 1 requires the Qian/Elgammal addition; (iii) the contents of US 6,335,988 / 6,801,663 / 6,591,006 / 6,944,357 / 7,039,245, which I did not individually verify.

8. Bottom line

Claims 1, 3–7, 9–14, and 17–21 appear highly vulnerable to a § 103 challenge built on Epshteyn (US 7,034,833) + Qian/Elgammal + GIF89a, with Chen (US 6,081,278) as the principal alternative/additional reference for the "single web-enabled graphic file" limitation. The combination relies on predictable arrangements of known elements with known functions, is supported by an express design incentive in Epshteyn itself, and is reinforced by the '644 specification's own admissions that GIF/JPEG/HTML, layering in Photoshop, and the desire to avoid Flash plug-ins were all known. The independent claims' only genuinely distinguishing limitations — the pixel-block sensitivity level and the single web-enabled graphic file — are each addressed by well-dated prior art, so the patent's defensibility rests narrowly on claim construction rather than on any demonstrated technical advance. This is an analyst's prima facie assessment, not a legal opinion, and it is not a substitute for a claim chart built from the granted-claim text in USPTO PatentCenter.