US 8351351

I'll search for the specific patent number and any related litigation, then summarize based on the authoritative full text provided.

Let me run additional targeted searches for litigation and USPTO records.

US Patent 8,351,351 — Summary

Important verification note: I searched for the literal identifier "8351351" and did not retrieve any USPTO litigation record or Federal Circuit (CAFC) 2026 docket entry for this patent. Searches for this number returned only patent-document sources (Google Patents, Justia, FreePatentsOnline, USPTO publication PDFs) — no court docket. I flag this explicitly rather than inferring litigation. A few queries did surface numerically similar but legally distinct patents (e.g., U.S. 8,867,351 in IPR2018-00709, and U.S. 8,073,161 in Maxell v. Samsung) — these are different patents and must not be conflated with 8,351,351.

The details below are drawn from the authoritative full text you supplied (Google Patents record, fetched 2026-09-28).

Bibliographic data

Abstract

The patent describes a multimedia-data generation method that (1) allocates one or more pieces of image-object information to each of a plurality of nodes, (2) generates a scene tree by linking the nodes according to correlation between the object information allocated to each node, and (3) adds an "event execution level attribute" to each node, where that attribute determines whether to execute an event using the node's image-object information. The resulting multimedia data represents the added attribute in a scene description language recorded in a scene-descriptor area.

Technical gist

The invention addresses a problem in object-based (e.g., MPEG-4) multimedia: conventionally, a decoding terminal had to decode/analyze/reproduce all nodes equally, regardless of whether it had the processing or reproduction capacity to do so, and distinguishing "basic profile" (AV) from "enhanced profile" (subtitle, related-article) data was not defined (e.g., in BIFS for DMB). The patent tags scene-tree nodes with hierarchical "event execution levels," so a receiving terminal can compare the node's level against an "allowable level" derived from its own capability (or a user request) and skip objects/events — and even avoid receiving their payload data — above that threshold.

Independent claims — plain-language overview

There are 20 claims total, with six independent claims (1, 6, 12, 15, 19, 20).

Claim 1 — Method of generating multimedia data
Stores object information at nodes, builds a scene tree by linking nodes per their correlation, attaches to every node an "event execution level attribute" that governs whether an event is executed using that node's object information, and outputs multimedia data where this attribute is written in a scene description language into a scene-descriptor area.

Claim 6 — Method of reconstructing multimedia data
After translating the scene description into a scene tree, the method: reads a node's event-execution-level attribute and extracts its level; determines an "allowable level" from the receiving end's status; compares extracted vs. allowable level; and, based on that comparison, decides both whether to execute the event for the current object and whether to descend to analyze a lower (child) node.

Claim 12 — Generation apparatus
The device counterpart of claim 1, comprising three units: a scene-tree generation unit (allocates object info, links nodes), an event-execution-level attribute addition unit (tags each node), and a multimedia data generation unit (writes the attribute into the scene descriptor via a scene description language).

Claim 15 — Reconstruction apparatus
The device counterpart of claim 6, comprising: an event-execution-level extraction unit (analyzes the current node's attribute and extracts its level), a level comparison unit (determines allowable level per receiving-end status and compares), and an analysis determination unit (decides whether to execute the current object's event and whether to analyze a lower node).

Claim 19 — Non-transitory computer-readable medium (generation)
A non-transitory storage medium bearing a program that, when executed, performs the generation method of claim 1 (allocate object info to nodes → build scene tree → add per-node event execution level attribute → output multimedia data with the attribute in a scene-descriptor area).

Claim 20 — Non-transitory computer-readable medium (reconstruction)
A non-transitory storage medium bearing a program that performs the reconstruction method of claim 6 (extract current node's level → determine allowable level → compare → decide event execution and whether to analyze a lower node).

Key dependent-claim concepts

• Claims 2–4 / 13: Hierarchical structure — an upper node holds a basic object; lower nodes hold the basic object plus related additional objects; lower nodes get a "more enhanced" (higher) event execution level than upper nodes; the attribute is placed at the uppermost node of a set of nodes sharing the same level.
• Claim 5 / 14: The attribute may additionally be fully defined in a descriptor other than the scene descriptor (e.g., an object-descriptor area).
• Claim 7: The allowable level reflects the receiving terminal's data-processing capability and/or reproduction capability, and/or an external (user) request level.
• Claim 8 / 16: If extracted level ≤ allowable level → execute the current object's event and descend to the lower node. If extracted level > allowable level → return to a higher node without executing the event (and without analyzing the lower node).
• Claims 9–10 (and 17): When the event is executed, the corresponding object's data is received, and in some embodiments received and decoded, to carry out the event.
• Claim 11 / 18: The event execution level may be determined from a descriptor other than the scene descriptor (e.g., from an object descriptor).

Caveats / uncertainty

• No litigation found: I could not locate any CAFC 2026 docket, district-court case, ITC action, or PTAB proceeding naming US 8,351,351. If such a matter exists, it is not reflected in the sources returned by my searches; I do not have authoritative information confirming or denying litigation activity for this specific number.
• Prosecution/assignment details (e.g., the Feb. 4, 2008 assignment effective date, the Dec. 2012 fee-payment/entity-status entries) are taken from the patent record and are stated as recorded, not independently verified against USPTO PatentCenter.
• The patent is indicated as expired for non-payment of maintenance fees; this is a status listed in the record and not a legal conclusion.

I'll search for litigation involving US patent 8351351 across patent litigation sources.

Let me search more targeted litigation databases.

Let me check specific litigation databases directly.

Litigation Search Results — U.S. Patent No. 8,351,351

Bottom line

I found no litigation involving U.S. Patent No. 8,351,351 ("the '351 patent"). My searches of patent‑litigation‑oriented sources returned no district court, ITC, or Federal Circuit case in which the '351 patent was asserted, challenged, or otherwise at issue. Based on the information available to me, no known litigation exists for this patent.

I want to be explicit about the limits of this answer: I was unable to obtain direct, docket‑level confirmation from PACER, Docket Navigator, or the Unified Patents litigation portal (the portal is subscription/dynamic and does not render results to a general search). The conclusion below rests on the searches I ran, which did not surface any case tied to this patent number.

What I checked and what I found

Important caution — likely sources of confusion (false positives)

Several results that looked close are not this patent. I flag them so they aren't mistaken for '351 litigation:

• ITC Investigation No. 337‑TA‑1351 (Certain Active Matrix OLED Display Panels…, Samsung Display as complainant). The "1351" here is an investigation number, not a patent number. Unrelated.
• ITC Investigation No. 337‑TA‑1308 / Arigna Technology v. Samsung et al. — involves U.S. Patent No. 7,183,835 ("the '835 patent"), a different patent from a different number family. Not the '351 patent.
• Imperium IP Holdings v. Samsung — involves U.S. Patent No. 6,271,884. Unrelated.
• Telcom Ventures v. Samsung (E.D. Tex.) — asserts a different set of patents (e.g., 11,937,172; 9,462,411; etc.). Unrelated.

Patent status context (for completeness, not litigation)

From the authoritative full‑text record you provided:

• Patent: US 8,351,351 B2 — "Method and apparatus for generating multimedia data having decoding level, and method and apparatus for reconstructing multimedia data by using the decoding level"
• Inventor: Sun‑hee Youm; Assignee: Samsung Electronics Co., Ltd.
• Priority: 2007‑07‑31 (KR 10‑2007‑0077166); Filed: 2008‑06‑02; Granted: 2013‑01‑08
• Continuation: US 9,185,378 B2 (from application 13/736,549), granted 2015‑11‑10
• Legal status: Expired – Fee Related. Maintenance fees lapsed; patent expired for failure to pay maintenance fees (event dated 2021‑02‑15; effective 2021‑01‑08), with adjusted expiration listed as 2031‑11‑09 (absent the lapse).
• Citations of record: six references (e.g., US 2002/0083032 A1 "Optimized BIFS encoder"; US 6,496,217 B1; KR 20050121412 A; etc.). None of these citations correspond to litigation.

Recommended next steps to confirm

Because I could not reach PACER/Docket Navigator/Unified Patents directly, if you need a definitive negative:

1. Docket Navigator — search 8,351,351 and 9185378 under "Patents Asserted" and "Patents in Suit."
2. Unified Patents litigation portal → https://portal.unifiedpatents.com/litigation/caselist — filter by patent number.
3. PACER / CourtListener RECAP — full‑text search for "8,351,351" and "8351351".
4. USPTO Patent Trial and Appeal Board — confirm no IPR/PGR filed against the '351 patent (relevant given it is an MPEG‑4/BIFS‑adjacent Samsung asset).

If you'd like, I can draft a short verification memo or a search‑log declaration memorializing these queries and their null results.

Note on adherence to your instruction: I interpreted the patent number literally as 8,351,351 and did not substitute or "correct" any similar‑looking identifier. The '351 patent is a Samsung multimedia/scene‑description patent that appears, on the record available to me, never to have been litigated.

I'll start by verifying the structured-data baseline against any web-searchable PTAB activity.

Let me confirm whether any of those hits actually reference this patent, and check for litigation-adjacent PTAB activity.

Let me run two final checks against PTAB docket aggregators and the patent's own Google Patents/PTAB linkage.

Proceedings overview

Total AIA trial proceedings on file for US 8,351,351: 0. The USPTO Open Data Portal structured block returned no AIA trial proceedings (no IPR, no PGR, no CBM) as of the most recent ingest. My independent web searches — PTAB petition documents, Docket Alarm PTAB dockets, and Google Patents — surfaced no petition, institution decision, final written decision, or appeal naming this patent. Because there are no proceedings, there is no breakdown by status: nothing was invalidated, nothing was sustained, nothing settled, and no institution was denied. The bottom line for a defendant is unusual and stronger than a "hardened patent" posture: this patent is not merely un-attacked, it is expired and unenforceable. Per the structured data it is "Expired – Fee Related" with a lapse for failure to pay maintenance fees on 2021-02-15, effective 2021-01-08, and an adjusted expiration of 2031-11-09 that was never reached. A demand letter citing this patent is citing a dead right, with no PTAB record to litigate over.

No proceedings to report

There is no proceeding number, petitioner, panel, ground, institution decision, or FWD to summarize. I am stating this explicitly rather than padding the template, because fabricating a proceeding number or a claim-level disposition would be worse than an empty section. Two verification notes:

• The structured block is canonical and I found nothing contradicting it. The ODP entry shows only maintenance-fee legal events (2016-06-29 fee payment; 2020-08-31 reminder; 2021-02-15 lapse and discontinuation). No APPR, IPR, PGR, CBM, or trial-number event appears.
• Lookalike "'351" patents dominate search results — do not confuse them with this one. Every apparent hit was a different patent: US 8,867,351 (Wi-LAN, IPR2018-00709, LG as petitioner), US 8,278,351 (Neptune Technologies krill oil, IPR2014-00003, Enzymotec), US 8,296,351, US 8,935,351, and US 8,223,117 (X-Mobile's brightness-adjustment patent). None of these is US 8,351,351 (Samsung Electronics, "Method and apparatus for generating multimedia data having decoding level"). In particular, a complaint snippet repeatedly featuring a "'351 Patent" whose asserted claim 1 concerns automatic brightness settings cannot be this patent — US 8,351,351 is Samsung-assigned and claims a multimedia scene-description method, so it could not be asserted by a third party against Samsung.

Why zero proceedings is plausible here (not an indication of a hardened patent):

• PGR was never available. The patent's effective filing date is 2007-07-31 (KR priority) / filed 2008-06-02, well before the 2013-03-16 AIA cutoff. Only IPR and (theoretically) CBM were open.
• CBM was never available. Under § 18 of the AIA, CBM review reaches only patents claiming a financial product or service. This patent claims object-oriented multimedia scene description and event-execution levels — not a covered business method.
• IPR alone was available, and no one filed one. The patent's own file shows why the incentive was low: it is a narrow MPEG-4/BIFS scene-description improvement that appears never to have been commercially asserted before it lapsed for unpaid fees.

Strategic summary

Claim status. Because no AIA trial ever reached a final written decision, no claim of US 8,351,351 has been canceled, and no claim has been adjudicated sustained. Claims 1–5 (the generation-method family, with independent claim 1 and dependents 2–5), claims 6–11 (the reconstruction-method family, independent claim 6), claims 12–14 (generation apparatus), claims 15–18 (reconstruction apparatus), and claims 19–20 (the two non-transitory computer-readable-medium claims) are all untested at the PTAB. That is a not-proven status, not a validated one — the distinction matters enormously here, because the reason no one tested them is that the patent's economic life ended early. The continuation US 9,185,378 (from application 13/736,549, filed 2013-01-08 as a continuation of this application) shows the same "Expired – Fee Related" status, and the Korean counterpart KR101370290B1 likewise lapsed. So the entire asserted family is dead.

Estoppel landscape. There is no § 315(e)(2) estoppel on anyone. Estoppel attaches only to a petitioner that loses on a claim in a final written decision under § 318(a). No IPR here was instituted, so no FWD exists, so no petitioner, real party in interest, or privy is estopped. For a defendant being asserted against today, every prior-art ground is technically still available — no reference is off the table by estoppel. That said, the more efficient answer is not an invalidity fight at all: the patent is expired, so the practical battleground is § 286 damages, not validity. If a validity fight were ever needed (e.g., over the tail end of the term), the closest known art sits in the patent's own face: US 2002/0083032 A1 (Bourges-Sevenier, "Optimized BIFS encoder"), US 6,496,217 B1 (Koninklijke Philips), US 2004/0030694 A1 (Tokumo), and KR 2005-0121412 A (ETRI, parametric scene description) — plus the two Korean family citations, KR 100505492 (MPEG-4 player object priority compositor) and KR 100622645 (MPEG-4 scene-rendering object replacement on embedded systems). The ETRI and KR 100622645 references are particularly on-point for the "differential object level per terminal capability" concept. Any such challenge would now have to go through ex parte or inter partes reexamination (available for this pre-AIA patent), not IPR-as-of-right in the ordinary commercial sense.

Pattern signals. No pattern exists: no petitioner filed on this patent at all, let alone serially; no patent owner appeal to the Federal Circuit exists; and there is no sign of a defensive aggregator such as Unified Patents in the chain. Samsung Electronics is the original and current assignee. The absence of any PTAB activity across the patent's entire ~13-year enforceable life is itself the signal — this is a patent that was never asserted hard enough to attract an IPR, and it was then abandoned when maintenance fees came due in January 2021.

Recommended next steps

• Lead with expiration, not invalidity. US 8,351,351 lapsed for nonpayment of maintenance fees with an effective date of 2021-01-08 (recorded as patent expiration/discontinuation on 2021-02-15). There is no PTAB final written decision to quote, so your strongest citation is the USPTO's own legal-status record rather than an FWD. Link the record: Google Patents US8351351B2, and verify status directly in USPTO Patent Center and PTAB E2E.
• Quantify the residual damages exposure. With the patent dead as of 2021-01-08 and a six-year § 286 lookback, a suit filed today reaches back only to roughly September 2020 — an exposure window of about four months of pre-expiration conduct, before accounting for marking (§ 287) and laches. That window is the entire case. Do not let a demand letter framed around "the patent's claims" obscure this.
• Check the family before you stop. Confirm the status of the continuation US 9,185,378 and the Korean counterpart KR101370290B1 — both show as not-active/expired-fee-related. If a plaintiff asserts either, the same expiration defense applies.
• If a plaintiff asserts claims 1–20 anyway, note that no PTAB tribunal has ever construed a single term of this patent. There is no institution decision, no FWD, and no claim construction to inherit. Your invalidity case starts from a clean slate, using the face-of-the-patent art listed above (most notably the BIFS-encoder and MPEG-4 object-priority references).
• No PTAB milestones to track. There is no active proceeding, so no institution deadline, no oral hearing, and no statutory one-year FWD clock is running. If you later believe a proceeding has been filed — and ODP ingest can lag — re-verify before assuming, but as of 2026-09-28 the correct statement to a client is: zero PTAB proceedings on file.

Confidence and caveats. I have high confidence in the "zero proceedings" conclusion because it comes from the canonical ODP block and my searches found no contrary primary document. I have not verified whether any district court litigation ever named this specific patent, and I found no Federal Circuit docket or CourtListener entry for it — so I am not characterizing its litigation history, only its PTAB history. I could not confirm whether the "'351 patent" mentioned in third-party complaint snippets is a numerically distinct patent; the subject matter mismatch strongly suggests it is, and I have treated it as such.

I'll verify the assignment record for US 8,351,351 before writing up the ownership chain.

Let me verify the assignment record details and check the continuation's ownership chain.

Ownership Analysis — US 8,351,351 B2

"Method and apparatus for generating multimedia data having decoding level, and method and apparatus for reconstructing multimedia data by using the decoding level"

Verification caveat up front: I could not open a live, docket-level view of USPTO Assignment Center (the search API at assignmentcenter.uspto.gov is not reachable from general search, and my proxy query for the reel/frame returned unrelated Samsung recordings from the 2001–2006 era). The ownership chain below is built from the legal-events block and bibliographic record in the authoritative full text you supplied, which reproduces the USPTO reassignment entry including the reel/frame. No post-issuance transfer appears anywhere in that record. Where a data point (the recording correspondent) is not exposed by these sources, I say so rather than guess.

Inventors

Patterns checked:

• Sole inventor, no co-inventors. Both the '351 patent and its continuation US 9,185,378 list Youm alone. No co-inventor join/severance dynamics to analyze.
• No inventor-departure signal. Youm is credited with 15 Samsung-assigned US patents, first granted 2009 and most recently 2021-12-28 (US 11,210,944). An inventor who is still filing under the same assignee in December 2021 — i.e., after this patent lapsed for unpaid fees in January 2021 — did not leave the original assignee within 12 months of filing or at any point relevant to a fire-sale. The classic "all inventors bolted" precursor is absent.
• No small-entity / micro-entity restructuring. All assignment and fee entries carry "ENTITY STATUS OF PATENT OWNER: LARGE ENTITY" (reel events dated 2012-12-05).

Original assignee

Samsung Electronics Co., Ltd. (Korea, Republic of) — assignee on the granted patent and, per the record, still the owner.

• Primary line of business: global consumer-electronics and semiconductor manufacturer — mobile devices, displays, memory, appliances, imaging.
• Product embodiment: the claims are directed to MPEG-4-style scene-description / BIFS event-execution levels for terminal-capability-adaptive decoding, a technology squarely within the DMB (Digital Multimedia Broadcasting) handset and set-top ecosystem Samsung build for in that period — the specification's own stated problem is the undefined BIFS basic-vs-enhanced profile split in DMB transport streams. Samsung shipped DMB-capable handsets in the relevant era, so it is plausible that commercial devices read on the claims. I cannot confirm that any specific shipping product practices the claims from the sources available, so I am not asserting product-embodiment as a fact.
• Current status: operating. No bankruptcy, dissolution, merger-into-shell, or divestiture of this asset appears in the record. Samsung remains a going concern and the named current assignee in the Google Patents "Current Assignee" field.

Assignment timeline

2008-02-04 (executed) / recorded 2008-06-02 — Reel 021027 / Frame 0930

• Conveyance: Assignment ("ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS)")
• Assignor: YOUM, SUN-HEE (individual; sole inventor)
• Assignee: SAMSUNG ELECTRONICS CO., LTD., Korea, Republic of
• Correspondent: Not exposed by the sources retrievable to me. The Google Patents legal-events block carries the reel/frame and the assignor/assignee fields but not the recording attorney or agent. I could not retrieve the Assignment Center cover sheet or the legacy legacy-assignments.uspto.gov PDF for reel 021027/0930 to read the correspondent of record. I am flagging this as a gap, not a null finding — it is the one field in the prescribed template I cannot populate. (For contrast, the legacy-assignment PDFs I did surface for other Samsung matters — e.g., reel 011745/0181, reel 017824/0844 — show that correspondents are documented on Samsung recordings from that era, so the omission here is a retrieval limitation, not an absence of record.)
• Context: In-house prosecution assignment. Routine pre-filing transfer of inventor rights to the corporate employer, executed ~5 months before the US filing date and contemporaneous with the Korean priority filing (KR 10-2007-0077166, 2007-07-31). Not an acquisition, reorg, securitization, or asserter transfer.

No other recorded assignment exists in the chain. The record contains no post-issuance conveyance of any type — no Assignment, Security Agreement, Merger, Change of Name, License, Release, or Correction recorded against US 8,351,351 after the 2008 entry. I checked the continuation and the Korean counterpart too:

• US 9,185,378 B2 (application 13/736,549, filed 2013-01-08 as a continuation of this application; issued 2015-11-10) — family record shows the same "Expired – Fee Related" status and no separate transfer of record.
• KR 10-1370290 B1 — Korean counterpart, likewise indicated not-active / expired-fee-related.

What follows the 2008 assignment in the record is maintenance-fee administration only, not title movement:

Timeline diagram

timeline
    title Ownership of US 8351351
    2007 : Korean priority filed
    2008 : Inventor assigns rights to Samsung
         : US application filed
    2013 : US patent issued
    2015 : Continuation issued as US 9185378
    2021 : Lapsed for unpaid maintenance fees

NPE / troll-pattern signals

Verdict

Insufficient data — on the rubric's express definition ("no records, or only the original assignment").

This is the correct label notwithstanding that the single record present is informative: the chain consists of exactly one conveyance, the 2008-02-04 in-house inventor-to-employer assignment to Samsung Electronics, reel 021027/0930, recorded 2008-06-02, and then nothing but maintenance-fee events terminating in the 2021-01-08 lapse for nonpayment. There is no downstream chain to score for NPE conduct — every one of the eight signals is not present because no transfer off the operating assignee ever occurred. Samsung, an operating company, still holds the patent; the asset was abandoned, not asserted and not sold, which is the decisively non-NPE fact pattern (no shell entities, no known asserter, no cascading LLCs, no pre-suit transfer, no privateering, no defensive-aggregator buyout). The one residual weakness in the analysis is the unretrieved recording correspondent for reel 021027/0930, which is a retrieval gap in my sources rather than evidence of a clean or dirty chain.

Verify at: USPTO Assignment Center (search patent number 8351351) · Google Patents US8351351B2 · USPTO Patent Center for the fee-lapse record.

Cross-reference status: the RPX and Unified Patents asserter directories are not implicated — there is no assertion activity to index. Per the earlier sections of this analysis, no district-court, ITC, Federal Circuit, or PTAB proceeding naming US 8,351,351 was found, so there is no Maxell v. Samsung-style or IPR-driven reassignment history to reconcile against this chain.

I'll research each cited reference to build the prior-art analysis. Let me run targeted searches.

Let me research the remaining cited references.

Prior-Art Analysis — US Patent 8,351,351

Scope and method. This section builds on the summary, litigation, PTAB, and strategic work already produced for the '351 patent. The patent number is used literally as 8,351,351; no numerically similar patent (e.g., US 8,867,351, US 8,278,351, US 8,296,351) is substituted or conflated. I analyzed the six patent citations of record from the Google Patents/USPTO citation list, plus the two "Family Cites Families" references and one non-patent-literature item, against the 20 claims of the '351 patent.

Three framing points before the reference-by-reference analysis:

1. Anticipation vs. obviousness. A § 102 anticipation requires every element of a claim, arranged as in the claim, disclosed in a single reference. On the record available, none of these references is a clean, complete § 102 anticipation of independent claims 1, 6, 12, 15, 19, or 20, because none discloses the full combination of (a) an "event execution level attribute" that determines whether to execute an event using the node's image-object information, (b) recorded in a scene-descriptor area via a scene description language, with (c) receiver-side allowable-level comparison. The references are stronger as § 103 obviousness material and as partial § 102 art against dependent claims. I state the closest § 102 mapping for each and flag where it falls short.

2. Date qualification (priority 2007-07-31 KR / filed 2008-06-02 US). References published before 2007-07-31 are available under § 102(a)/(b). References published after that date but filed earlier may be available under § 102(e). One citation of record (US 2009/0295805 A1) shares the '351 filing date and is therefore questionable as § 102 art — flagged below.

3. Two references are currently only partially retrievable. I retrieved full text for the U.S. references but could not retrieve the actual disclosure of KR 2005-0121412 A or the two Korean family citations; for those I rely on the recorded titles, dates, and assignees, and I mark the § 102 mapping as provisional pending full-text review. Per your rules, I am not fabricating claim-level content I could not verify.

1. Cited references of record — summary table

(Reference 5's asterisk in the USPTO list marks "cited by examiner"; the date columns are reproduced as recorded.)

2. Reference-by-reference analysis

Reference 1 — US 2002/0083032 A1 / US 6,693,645 B2 ("Optimized BIFS encoder")

• Full citation: US 2002/0083032 A1 (published application; granted as US 6,693,645 B2), Optimized BIFS encoder, inventor Mikael Bourges-Sevenier, assignee iVast, Inc. (Santa Clara, CA). Application 09/727,800; provisional 60/168,778.
• Dates: Provisional filed 1999-12-01; application filed 2000-11-30; published 2002-06-27; patent issued 2004-02-17.
• Brief description: Processes a hierarchy of nodes (a scene graph) specifying a VRML/XML scene and inserts quantization parameters per node, then subsumes child-node quantization parameters into the parent node where cost-effective, and encodes the hierarchy into a BIFS binary file (MPEG-4 Binary Format for Scenes). Explicitly discusses the scene graph, parent/child node levels, node "parameters," and hierarchical inheritance where a parent node's parameter governs children unless overridden at the child.
• Potential § 102 anticipation: This is the closest reference on the structural half of the invention:
  • Claim 1 / claim 12 (partial): discloses allocating object/parameter data to nodes, generating a linked node hierarchy (scene graph), and recording node attributes in a scene-descriptor format (BIFS). It anticipates the scene-tree-generation and scene-descriptor-recording elements but does not disclose an "event execution level" that determines whether to execute an event, so it cannot alone anticipate claim 1/12.
  • Claim 3 (partial): the parent-default / child-override mechanism, and the deliberate subsumption of child parameters into the parent, is structurally analogous to the claim's "more enhanced level at the lower node than the upper node" and claim 4's placement of an attribute at the uppermost node of a group sharing the same value — but the reference's parameter is a quantization attribute, not an event-execution level.
  • Claim 5 (partial): supports the notion that an attribute may be carried in the scene description itself (BIFS) rather than a separate descriptor.
• Bottom line: Strong § 102(b) art for the scene-tree/BIFS scaffold of claims 1 and 12, and for the hierarchical-inheritance and "uppermost-node attribution" concepts of claims 3–4; not a full anticipation of any independent claim because the "event execution" determination is absent.

Reference 2 — US 6,496,217 B1 ("Video communication system using model-based coding and prioritization techniques")

• Full citation: US 6,496,217 B1, Video communication system using model-based coding and prioritization techniques, assignee Koninklijke Philips Electronics N.V. (family publication US 2002/0196328 A1; JP 2004-529597 A).
• Dates: Filed/priority 2001-06-12; issued 2002-12-17.
• Brief description: Identifies at least two objects in an image/video, assigns object models, determines a priority for each object according to a predetermined priority rule, assigns higher-priority objects to higher-reliability channels, and masks lower-priority objects while encoding higher-priority objects (and vice versa). Priority may change dynamically during a session.
• Potential § 102 anticipation:
  • Claim 3 (partial): discloses differential treatment of objects by a priority/level ordering and the idea that additional/lower-priority objects are handled separately from the base object.
  • Claim 1 / claim 12 (partial): object identification and prioritization, but there is no scene tree, no scene descriptor, and no per-node "event execution" attribute — it operates at the channel-coding layer, not the scene-description layer.
• Bottom line: Best characterized as § 102(b)/§ 103 art for the "objects carry differential priority levels" premise underlying claims 3–4. Not anticipatory of any independent claim.

Reference 3 — US 2004/0030694 A1 ("Search information transmitting apparatus")

• Full citation: US 2004/0030694 A1, Search information transmitting apparatus, inventor Yasuaki Tokumo.
• Dates: Priority 2000-10-20; published 2004-02-12.
• Brief description: Transmits/receives search information represented as a tree structure for locating moving-image content. A search-information analyzer, an extraction unit that extracts only the necessary search-information elements according to a prescribed condition, and a regenerator that builds a partial tree; a condition-setting unit variably sets the condition (e.g., by element type). The stated object is to transmit only the necessary portion of the tree.
• Potential § 102 anticipation: This is the closest reference on the receiver-side traversal/selection half of the invention:
  • Claims 6 / 15 (partial): discloses traversing a tree structure, applying a condition to select which parts to process/transmit, and generating a partial tree — conceptually analogous to the claim's "compare extracted level to allowable level and decide whether to analyze a lower node." The condition-setting unit parallels the claim's "allowable level … according to a status of a receiving end."
  • Claims 8 / 16 (partial): the extraction/regeneration of a partial tree (skipping portions not meeting the condition) maps onto "proceed without processing the node if the level is exceeded."
• Bottom line: Potentially § 102(a)/(b) art directed at the selective tree traversal / partial-tree concept of claims 6, 8, 15, and 16 — but its "condition" governs search-information selection, not event execution of image objects in a scene description, and it has no "event execution level attribute" per node. Not a full anticipation.

Reference 4 — KR 2005-0121412 A ("Apparatus for coding/decoding interactive multimedia contents using parametric scene description")

• Full citation: KR 2005-0121412 A, applicant 한국전자통신연구원 (Electronics and Telecommunications Research Institute, ETRI).
• Dates: Filed 2004-06-22; published 2005-12-27.
• Brief description (from recorded title/metadata only): an apparatus for coding/decoding interactive multimedia contents using a parametric scene description. On its face this is the most thematically on-point citation — it sits squarely in the same MPEG-4 scene-description/"scene descriptor" space as the '351 patent, and the earlier strategic section already flagged it as "particularly on-point for the 'differential object level per terminal capability' concept."
• Potential § 102 anticipation (provisional): the recorded subject matter could implicate the generation and reconstruction claims generally:
  • Claims 1 / 12 (generation): coding multimedia using a scene description — potentially relevant to the scene-descriptor-recording element.
  • Claims 6 / 15 (reconstruction): decoding interactive multimedia from a scene description — potentially relevant to the scene-descriptor-translation element.
  • Claims 5 / 11 (attribute defined by a descriptor other than the scene descriptor): a "parametric" scene-description approach may separate parameters from the base scene description.
• Caveat — I could not retrieve the full text. I therefore do not assert that it discloses the "event execution level attribute," the allowable-level comparison, or the differential lower/upper node handling. Verify against the Korean full text before relying on this reference. On the title alone, this is the citation most likely to overlap the '351 invention's core, and it should be the first reference a challenger develops.

Reference 5 — US 2009/0295805 A1 ("Hierarchical based 3D image processor, method, and medium")

• Full citation: US 2009/0295805 A1, assignee [Samsung Electronics Co.](/litigations/by-defendant/Samsung%20Electronics%20Co.), Ltd. (the '351 patent's own assignee).
• Dates (as recorded): priority listed 2008-06-02; published 2009-12-03.
• Brief description: a hierarchical-based 3D image processor/method/medium — i.e., level-of-detail or hierarchical processing of 3D imagery.
• Potential § 102 anticipation — likely unavailable on dates:
  • Because it published (2009-12-03) after the '351 priority date (2007-07-31), it is not § 102(a)/(b) art.
  • Its recorded priority/filing date (2008-06-02) is the same day as the '351 filing (2008-06-02). § 102(e) requires the reference's effective filing date to be before the applicant's, so a same-day filing does not qualify as § 102(e) art absent an earlier benefit claim. This citation therefore appears date-blocked as § 102 art and was likely cited as co-pending/related art (same assignee, same filing date) or for § 103 context.
  • Bottom line: If its true effective date is earlier than 2008-06-02 (e.g., via a foreign priority not shown), it could support § 102(e) against claims 1/12 on generic hierarchical-image-processing grounds only. As recorded, treat it as not § 102-qualifying and not an anticipation of any claim.

Reference 6 — US 2012/0084839 A1 ("Surveillance network system")

• Full citation: US 2012/0084839 A1, assignee The Boeing Company.
• Dates: priority 2005-12-22; published 2012-04-05.
• Brief description: a surveillance network system — distributed sensing/processing across a network of nodes.
• Potential § 102 anticipation:
  • Published after the '351 priority date, so not § 102(a)/(b). Its 2005-12-22 priority predates the '351 filing, so it could be § 102(e) art if it has a U.S. filing/benefit date before 2008-06-02 — but the subject matter (surveillance networking) does not disclose a scene tree, event execution levels, or scene-descriptor attributes.
  • Bottom line: No direct § 102 anticipation of any claim. At most a § 103 background reference on distributed/networked processing; it does not touch the claim elements.

Family citations (cited in the family — Korean counterpart prosecution)

F1 — KR 100505492 B1, The object priority order compositor for MPEG-4 player (KETI / 전자부품연구원)

• Dates: filed 2002-09-30; granted 2005-08-03. Availability: § 102(b).
• Brief description: a compositor that assigns/handles a priority order among objects in an MPEG-4 player.
• Potential § 102 anticipation: Compares directly to the '351 patent's core differentiator — "different objects carry different levels":
  • Claim 3 (partial): priority-order assignment to MPEG-4 objects resembles the "more enhanced event execution level to the lower node than the upper node."
  • Claim 4 (partial): priority-order grouping resembles placing a shared attribute at one node of a group.
  • Claim 1 (partial): MPEG-4 object compositing/coding context.
• Caveat: Title-level only; full text not retrieved. This is priority/leveling art in the same MPEG-4 scene domain and should be developed alongside Reference 4.

F2 — KR 100622645 B1, Object replacement and attribute conversion device and method in MPEG-4 scene rendering on embedded system (KETI)

• Dates: filed 2004-12-14; granted 2006-09-19. Availability: § 102(b).
• Brief description: device/method for object replacement and attribute conversion during MPEG-4 scene rendering on an embedded (resource-constrained) system.
• Potential § 102 anticipation: This is the closest family citation to the receiver-side half of the invention:
  • Claim 7 (partial): terminal-capability-based adaptation of an MPEG-4 scene on an embedded system parallels "allowable level … according to a data-processing/reproduction capability of a terminal."
  • Claims 6 / 15 (partial): scene rendering/reconstruction with attribute conversion on a constrained device.
• Caveat: Title-level only. Not a full anticipation of claims 6/15, but materially relevant to the capability-based rendering limitation of claim 7 and dependent reconstruction claims.

Non-patent literature (similar-document listing, not a citation of record)

• Chon et al. (2006), "An approach of dynamically forming an acceptable transcoding path by a search algorithm based on a context-free grammar in a MPEG-21 DIA." Appears in the "Similar Documents" list. Relevant thematically (dynamically selecting a path based on terminal/context conditions in an MPEG-21 Digital Item Adaptation setting) and potentially useful as § 103 background on capability-driven processing-path selection. It is not a citation of record on the '351 patent, so I flag it separately.

3. Claim-by-claim § 102 mapping (best available)

4. Ranking and recommended focus

Most relevant prior art, in order:

1. KR 2005-0121412 A (ETRI) — same MPEG-4 scene-description space; highest thematic overlap, full text needed. Prioritize this.
2. KR 100505492 B1 and KR 100622645 B1 (KETI) — object priority ordering in MPEG-4 players and attribute conversion on embedded systems; together they map closely onto the "differential object levels" and "terminal-capability-based rendering" concepts (claims 3, 4, 7, 15).
3. US 2002/0083032 A1 / US 6,693,645 B2 (Optimized BIFS encoder) — strongest U.S. reference; supplies the scene-tree + per-node-attribute + BIFS scene-descriptor scaffold (claims 1, 3, 4, 12).
4. US 2004/0030694 A1 (Tokumo) — supplies selective/partial tree traversal under a settable condition (claims 6, 8, 15, 16).
5. US 6,496,217 B1 (Philips) — object prioritization + hierarchical coding (claims 1, 3).
6. US 2012/0084839 A1 (Boeing) and US 2009/0295805 A1 (Samsung) — background/§ 103 only; the latter appears date-blocked as § 102 art.

Key takeaway for any challenge. The '351 patent's asserted novelty is the combination of a per-node "event execution level attribute" governing event execution, its recording in a scene descriptor, and receiver-side comparison to an allowable level to skip node traversal and object reception. The cited art broadly teaches objects with differential priority/parameters in a scene tree (Refs. 1, 2, F1), selective tree traversal under a condition (Ref. 3), and terminal-capability-based MPEG-4 rendering (F2, Ref. 4). The most credible attack is therefore a § 103 combination (e.g., Ref. 1's per-node scene-tree attributes + Ref. 3's condition-driven partial-tree traversal + F2/Ref. 4's capability-based rendering), not any single-reference § 102 anticipation.

5. Caveats and verification notes

• No auto-correction applied. All identifiers are reproduced literally. Numerically similar patents (US 8,867,351; US 8,278,351; US 8,296,351; US 8,935,351; US 8,223,117) are different patents and were excluded.
• KR 2005-0121412 A, KR 100505492 B1, and KR 100622645 B1 were not retrieved in full. Their § 102 mappings are provisional and based on recorded titles, dates, and assignees; I did not fabricate disclosure content. Full-text review (KIPRIS/Korean originals) is required before asserting anticipation.
• US 2009/0295805 A1 shares the '351 filing date (2008-06-02) as recorded and post-dates the priority date; treat its § 102(e)/§ 102(a)/(b) availability as unconfirmed and likely unavailable unless an earlier benefit date is shown.
• Anticipation is a high bar and, on the record, unmet by any single reference. This analysis identifies § 102 candidates and their limitations; it is not a legal conclusion and does not replace a full claim chart against the Korean-language originals.

I'll build the § 103 obviousness analysis on the prior art of record. Let me first pull details on the cited references so I ground each teaching accurately rather than paraphrasing only the titles.

Obviousness Analysis — U.S. Patent No. 8,351,351 under 35 U.S.C. § 103

Analysis date context: The '351 patent is a pre-AIA patent (effective filing date 2007-07-31 via KR 10-2007-0077166; U.S. filed 2008-06-02). Section 103 therefore applies with pre-AIA § 102 prior-art definitions, but the KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007) framework governs the obviousness inquiry. The patent lapsed for non-payment of maintenance fees effective 2021-01-08, so this analysis is retrospective.

Scope note on the prior art considered: Per instruction I used the Prior Art / Citations section of the patent page as the universe of references, i.e. the six patent citations and two family citations of record. I did not retrieve full texts for every reference; where I relied on title/abstract-level information only, I flag it. Critically, I note that the two principal references — US 2002/0083032 A1 (Bourges-Sevenier) and US 6,496,217 B1 (Koninklijke Philips) — were also cited during prosecution, meaning the Examiner considered them and still allowed the claims. That fact is a genuine (though not dispositive) obstacle to an obviousness theory, and I address it candidly in Part IX.

I. The prior art of record — what each reference actually teaches

Two quick observations. (1) Bourges-Sevenier supplies the mechanism claim 1 needs — a per-node attribute recorded in the scene description of an MPEG-4/BIFS file — and even supplies claim 4's "uppermost node" consolidation almost verbatim. (2) Philips and KETI '492/'645 supply the semantic the claims need — a per-object priority/level used, based on the capability of the receiver, to decide which objects to process/present and which to skip. What no single reference supplies is the two combined under the label "event execution level."

II. Level of ordinary skill in the art (POSITA)

A POSITA here would be a software/systems engineer with (a) a bachelor's in EE/CS or equivalent, (b) 2–3 years' experience with MPEG-4 Systems / BIFS scene description and object-based multimedia coding, and (c) familiarity with scalable/layered coding and terminal-capability adaptation. This is the level the patent's own specification presumes — it treats VRML/BIFS scene trees, object descriptors, and payload/scene-descriptor partitioning as background knowledge. The prior art references are all at or below this level (several are themselves player/encoder implementations), which supports that the references are analogous art.

III. Claim construction of the pivotal terms

These constructions drive the analysis:

• "scene tree … nodes … linked according to correlation" — the MPEG-4/VRML/BIFS scene graph of parent/child nodes describing objects and their spatial/temporal relations. Bourges-Sevenier and KETI '492 both disclose exactly this structure.
• "event execution level attribute … determines whether to execute an event by using the … object information" — a per-node value that gates whether the object's associated processing/reproduction ("event") is carried out. This is the point of novelty over the cited art; Bourges' quantization parameter and Philips/KETI's priority are the closest analogues.
• "scene description language … recorded on an area of a scene descriptor" — BIFS/VRML/BIFS-Command text or binary written into the scene-description portion of the MPEG-4 stream. Bourges-Sevenier is squarely on point (its output is a BIFS scene description).
• "allowable level" (claim 6) — a receiver-side threshold derived from terminal capability and/or user request. Bourges' user-supplied "performance parameter" and KETI '645's limited-resource embedded systems are the two natural sources for this term.

IV. Obviousness of independent claim 1 (generation method)

Claim 1 requires: (a) allocating object information to nodes; (b) linking nodes into a scene tree by correlation; (c) adding to each node an event execution level attribute that determines whether to execute an event using that node's object information; and (d) generating multimedia data in which that attribute is represented in a scene description language and recorded in a scene-descriptor area.

Primary combination: Bourges-Sevenier + Philips (+ KETI '492)

Why the combination is proper. All three are in the same field of endeavor (MPEG-4 object-based scene description / players / encoders) and are reasonably pertinent to the problem the '351 specification itself states — that a terminal must "manually decode, analyze, and reproduce all available data, while equally treating each node." Bourges-Sevenier demonstrates that a per-node processing attribute can be authored into the scene description; Philips and KETI '492 demonstrate that objects in such a scene can be ranked and selectively processed. Substituting Bourges' quantization semantics for a priority/level semantics — i.e., "tag each node with an attribute that says whether it's in-scope for this receiver" — is the kind of substitution of one known parameter type for another that KSR holds obvious.

Motivation to combine — articulated rationales (KSR/MPEP)

1. Known problem, known solution set. The patent admits the problem (uniform treatment of all nodes) was pre-existing. Philips identifies the same class of problem for constrained channels ("low bitrate channels … the received resulting image quality is generally not acceptable") and solves it by prioritizing objects; KETI '645 targets "embedded systems with limited resources." Motivating combination to solve this recognized problem is a classic KSR rationale.
2. Predictable result from combining known elements. Placing a priority/level flag on scene-graph nodes (Philips/KETI) inside the scene-description stream (Bourges) yields the expected result: encoders emit the flag, decoders read it. No unpredictable behavior.
3. Improvement of a similar device in the same way. KETI '492 is an MPEG-4 player module; Bourges is an MPEG-4 encoder. A POSITA adapting a player-side priority scheme to be signaled from the encoder through the existing scene description is routine.
4. Design incentive / market force. DMB/BIFS terminals spanned low-capability PDAs to notebooks (the patent's own FIG. 6 framing). Providing capability-adaptive scene signaling was an evident design requirement.

Claim 1 conclusion: Obvious under § 103 over Bourges-Sevenier in view of Philips and KETI '492. I would put this at a moderate-to-strong obviousness position — diminished somewhat by the Examiner having had Bourges and Philips of record (Part IX).

V. Obviousness of independent claim 6 (reconstruction method)

Claim 6 requires: translating the scene description into a scene tree; reading a current node's event execution level attribute and extracting its level; determining an "allowable level" according to a status of the receiving end; comparing the two; and, per the comparison, deciding whether to execute the current object's event and whether to analyze a lower node.

Primary combination: KETI '492 + Philips + Bourges-Sevenier (+ KETI '645 for "receiving-end status")

Why the combination is proper and motivated.

• The traversal-gating idea is expressly in KETI '492. It traverses the scene tree child-by-child (depth-first) and re-presents only the objects that require reconstruction, skipping the rest. That is the same functional operation as claim 6's "determine whether to analyze a lower node of the current node" keyed to a per-object priority.
• The "allowable level" threshold is taught by Bourges and Philips. Bourges discloses a user-supplied or predetermined performance parameter against which node parameters are evaluated — structurally identical to the "allowable level." Philips discloses that the priority rule may be set manually / by criteria / dynamically — i.e., receipt-end status drives the level.
• The "status of the receiving end" is taught by KETI '645, which frames the whole optimization around an embedded system with limited resources and avoids parsing unnecessary nodes.
• Motivation: the references share the same objective as the '351 patent — reduce decoding/rendering work by not processing objects a receiver cannot or need not handle. Philips' and KETI's explicit statements about presentation speed, blinking/overhead, bandwidth, and limited resources in resource-constrained terminals supply the articulated reason to combine with Bourges' BIFS-level parameter signaling.

Claim 6 conclusion: Obvious under § 103 over KETI '492 and KETI '645 in view of Philips and Bourges-Sevenier. Position: moderate. The weakest link is that no reference literally gates "execution of an event" on the level; the references gate presentation/transmission. A POSITA would, however, read "execute an event using the object information" as encompassing presentation/rendering, and the claim does not appear to require anything more.

VI. Apparatus claims 12 and 15, and CRM claims 19 and 20

• Claim 12 (generation apparatus): maps unit-for-unit onto the Part IV combination — the "scene tree generation unit" ≈ Bourges' scene-graph construction / KETI's parser-manager + scene tree; the "event execution level attribute addition unit" ≈ Bourges' quantization-parameter insertion engine (with Philips/KETI supplying the priority semantics); the "multimedia data generation unit" ≈ Bourges' BIFS encoding engine. Obvious for the same reasons.
• Claim 15 (reconstruction apparatus): maps onto the Part V combination — "event execution level extraction unit" ≈ KETI '492's priority-order extraction; "level comparison unit" ≈ Bourges' parameter-vs-performance-parameter comparison plus Philips' priority rules; "analysis determination unit" ≈ KETI '492's decision to (re)present only required objects / KETI '645's selective node handling. Obvious for the same reasons.
• Claims 19–20 (non-transitory CRM): these recite the claim 1 / claim 6 methods embodied as program code. Where the underlying method claims are obvious, the CRM claims add only the non-transitory medium — and the cited art already discloses storing such logic as "computer readable code … stored in a memory … or read/downloaded from a … CD-ROM or floppy disk" (Philips, US 6,496,217, describing its coder/decoder applications). No separate patentable weight.

VII. Dependent claims

Assessment: Claims 4 and 7 are the most clearly obvious (claims 4 and 7 are strongly, almost literally, met). Claims 5/14 and 11/18 (attribute in a non-scene descriptor) are the most vulnerable to a counterargument that the cited art locates all attributes in the scene description, not the object descriptor — though MPEG-4's OD layer is squarely within the POSITA's knowledge and the specification itself treats the OD option as an obvious variant.

VIII. Consolidated motivation-to-combine statement

The single-sentence articulation I would use against every independent claim:

It would have been obvious to a POSITA, motivated by the recognized need to serve heterogeneous, resource-constrained terminals (Philips; KETI '645) and the known technique of ranking objects in an MPEG-4 scene (Philips; KETI '492), to add to the node hierarchy of a BIFS scene description (Bourges-Sevenier) a per-node level/priority attribute — recorded in the scene-descriptor area exactly as Bourges records quantization parameters — and, at the receiver, to compare that attribute against a receiver-derived threshold (Bourges' user-supplied performance parameter; KETI '645's resource status) to decide whether to process the node's object and whether to traverse into its children (KETI '492's selective, depth-first (re)presentation of only required objects). Each reference is in the same field of MPEG-4 object-based scene description/rendering and addresses the same problems of bandwidth, capability, and processing overhead, so no non-analogous-art or teaching-away barrier applies.

IX. Counterarguments, weaknesses, and secondary considerations — candid assessment

1. The Examiner already had the two strongest references and allowed. Bourges-Sevenier and Philips are of record and were considered. Their express listing in the "Patent Citations" table is evidence (weak, non-binding) that the claimed "event execution level attribute" was viewed as distinct from a quantization parameter and from a transmission-priority/channel-assignment value. Any obviousness challenge must therefore explain the distinction, and the patent's own framing — gating execution and analysis of child nodes, not channel protection or bit-depth — is where the nonobviousness argument lives.
2. Philips is arguably a different field of endeavor. US 6,496,217 is model-based (facial) video coding for conferencing, with prioritization aimed at channel reliability, not scene-description authoring. It is still reasonably pertinent (it discusses scene composition and layered coding), but its priority teaching is about transport robustness, which is a different purpose than gating a decoder's event execution. KETI '492/'645 are the stronger references on that point because they operate directly on the MPEG-4 scene tree.
3. No reference literally recites an "event execution level" that suppresses event execution and child-node analysis. KETI '492's "re-present only objects requiring reconstruction" is the closest functional analog, but a patent owner could argue it is a rendering-order optimization, not a decode/analysis gate keyed to a level attribute. This is the single most contestable limitation.
4. Claims 5/14 and 11/18 (attribute defined outside the scene descriptor) are the least well-supported, because the on-point references place attributes in the scene description itself.
5. No secondary considerations are evidenced. There is no record of commercial success, long-felt need, copying, or industry praise tied to this patent, and there is no PTAB final written decision to cite (the prior analysis confirms zero AIA proceedings on the '351 patent). Nexus is therefore not an issue — but that also means the patent owner could try to establish secondary considerations afresh.
6. The Samsung '805 reference must be excluded (same-day filing, same assignee — not prior art). Relying on it would be an error.
7. I could not verify the full technical content of ETRI '412 and Tokumo, and Boeing '4839 is remote in subject matter. I would not lean on these as primary references; they are corroborating at best.

X. Bottom line

Overall: On this record, the independent claims of US 8,351,351 are reasonably, but not conclusively, obvious under § 103. The combination of Bourges-Sevenier (per-node attribute recorded in a BIFS scene description, with uppermost-node consolidation) + Philips (per-object priority by predetermined rules, optional bases for setting it) + KETI '492/'645 (priority-driven, resource-aware selective processing of an MPEG-4 scene tree) supplies every limitation. The residual strength of the patent lies entirely in the proposition that none of the references gates execution of an event and descent into child nodes on a receiver-derived level — a distinction that is narrow, functionally small, and, given KSR's skepticism of such narrow functional differences, likely insufficient to preserve patentability against a well-constructed challenge. That challenge would now have to proceed by ex parte/inter partes reexamination (the patent being pre-AIA), since no IPR path was ever used and the patent is expired.

Caveats: This analysis is grounded in the patent's own citation record plus the retrieved text of Bourges-Sevenier, Philips, KETI '492, and KETI '645. I did not obtain full texts of ETRI '412 or Tokumo, and I have flagged where I relied on title/abstract-level information. No PTAB or court has ever construed any term of this patent (zero AIA proceedings on file), so all claim constructions above are my own and unadjudicated.