DCT
8:18-cv-02058
Uniloc 2017 LLC v. Hulu LLC
Key Events
Complaint
I. Executive Summary and Procedural Information
- Parties & Counsel:
- Plaintiff: Uniloc 2017 LLC (Delaware)
- Defendant: Hulu, LLC (Delaware)
- Plaintiff’s Counsel: Feinberg Day Alberti Lim & Belloli LLP
- Case Identification: 8:18-cv-02058, C.D. Cal., 11/17/2018
- Venue Allegations: Plaintiff alleges venue is proper because Defendant has committed acts of infringement in the district and maintains a regular and established place of business in the Central District of California.
- Core Dispute: Plaintiff alleges that Defendant’s video streaming service and underlying video encoding technologies infringe three patents related to video compression and online content aggregation.
- Technical Context: The technologies at issue concern methods for efficient digital video compression, particularly within the H.264 standard, and systems for aggregating disparate sources of online media for presentation to a user, both of which are central to the operation of modern streaming platforms.
- Key Procedural History: Post-filing Inter Partes Review (IPR) proceedings have significantly impacted this case. The asserted claims of U.S. Patent No. 6,519,005 (IPR2019-01126, IPR2019-01270) and U.S. Patent No. 9,721,273 (IPR2019-01363) were found unpatentable and cancelled by the U.S. Patent and Trademark Office. These cancellations occurred after the filing of the complaint and are likely to be dispositive for the infringement counts related to those two patents.
Case Timeline
| Date | Event |
|---|---|
| 1999-04-30 | U.S. Patent 6,519,005 Priority Date |
| 2001-03-06 | U.S. Patent 6,895,118 Priority Date |
| 2003-02-11 | U.S. Patent 6,519,005 Issue Date |
| 2005-05-17 | U.S. Patent 6,895,118 Issue Date |
| 2008-08-21 | U.S. Patent 9,721,273 Priority Date |
| 2017-08-01 | U.S. Patent 9,721,273 Issue Date |
| 2018-11-17 | Complaint Filing Date |
| 2021-08-31 | Cancellation of claims 1-16 of U.S. Patent 6,519,005 |
| 2025-01-31 | Cancellation of claims 1-3 of U.S. Patent 9,721,273 |
II. Technology and Patent(s)-in-Suit Analysis
U.S. Patent No. 6,519,005, "Method of Concurrent Multiple-Mode Motion Estimation For Digital Video," issued February 11, 2003
The Invention Explained
- Problem Addressed: The patent describes that prior art video compression techniques, such as those used in the MPEG standard, relied on motion estimation procedures that were computationally intensive and expensive, creating a need for a more efficient process (Compl. ¶12; ’005 Patent, col. 3:25-42).
- The Patented Solution: The invention provides a method to perform motion estimation for multiple different prediction modes (e.g., frame prediction vs. field prediction) concurrently within a single search operation. This is achieved by comparing portions of a macroblock against an anchor picture and combining the resulting error metrics to simultaneously determine the optimal mode and motion vector, which is described as simpler, faster, and less expensive than prior methods (Compl. ¶13; ’005 Patent, col. 1:6-11, col. 7:14-30).
- Technical Importance: This approach aimed to improve the efficiency of video encoding, a critical step for the widespread adoption of digital video technologies like digital television and DVDs that rely on standards like MPEG-2 (Compl. ¶12; ’005 Patent, col. 1:31-38).
Key Claims at a Glance
- The complaint asserts independent claim 1 ('005 Patent, col. 15:8-22).
- The essential elements of claim 1 include:
- A method for motion coding an uncompressed digital video data stream, including steps of:
- comparing pixels of a first pixel array with pixels of a plurality of second pixel arrays in a reference picture;
- concurrently performing motion estimation for each of a plurality of different prediction modes to determine an optimum prediction mode;
- determining a best match for the first pixel array for the optimum prediction mode; and
- generating a motion vector for the first pixel array in response.
- The complaint's infringement analysis focuses exclusively on claim 1 (Compl. ¶17).
U.S. Patent No. 6,895,118, "Method Of Coding Digital Image Based on Error Concealment," issued May 17, 2005
The Invention Explained
- Problem Addressed: The patent acknowledges that while dropping data blocks from a video stream can increase compression, it can cause synchronization problems for the decoder. Existing standards like MPEG-4 used resynchronization markers to recover from data loss, but there was a need to more intelligently combine block dropping with these recovery mechanisms (Compl. ¶34; ’118 Patent, col. 1:35-42).
- The Patented Solution: The invention proposes a method where an encoder first estimates a macroblock's "capacity to be reconstructed" by a decoder using error concealment techniques. Based on this estimation, a decision is made whether to exclude (i.e., not code) the macroblock. Following the exclusion, a resynchronization marker is inserted into the data stream, a process that involves evaluating the data savings of the exclusion against the overhead cost of the marker (Compl. ¶34, ¶36; ’118 Patent, col. 2:8-27, Abstract).
- Technical Importance: This method enables more efficient compression by allowing an encoder to strategically omit data that can be reliably reconstructed by the decoder, which is valuable for transmission over constrained networks (Compl. ¶35).
Key Claims at a Glance
- The complaint asserts independent claim 1 (’118 Patent, col. 8:1-12).
- The essential elements of claim 1 include:
- A method of coding a digital image comprising macroblocks in a binary data stream, including:
- an estimation step, for macroblocks, of a capacity to be reconstructed via an error concealment method;
- a decision step to exclude a macroblock from coding based on that capacity; and
- inserting a resynchronization marker into the binary data stream after the exclusion of one or more macroblocks.
- The complaint’s infringement analysis focuses exclusively on claim 1 (Compl. ¶38).
U.S. Patent No. 9,721,273, "System and Method For Aggregating And Providing Audio And Visual Presentations Via A Computer Network,” issued August 1, 2017
- Technology Synopsis: The patent addresses the problem of information overload on the internet, where conventional search engines may return non-meaningful results (Compl. ¶51; ’273 Patent, col. 1:51-2:2). The invention provides a system and method to aggregate presentations from multiple sources—including locally stored content and content from external feeds that are periodically accessed—and deliver them for presentation on a common web page (Compl. ¶52; ’273 Patent, col. 2:39-55).
- Asserted Claims: The complaint asserts independent claim 1 (’273 Patent, col. 12:40-13:20).
- Accused Features: The complaint alleges that Hulu's service, which presents users with aggregated collections of content on its homepage and category pages (e.g., in "Keep Watching," "Line Up," and "Live" rows), infringes the patent's method of aggregating and providing presentations (Compl. ¶¶57-69).
III. The Accused Instrumentality
Product Identification
- The complaint identifies the accused instrumentalities as Hulu's H.264 encoders and the Hulu video streaming service and computer system (Compl. ¶¶16, 22, 37, 55).
Functionality and Market Context
- The complaint alleges that Hulu's streaming service operates over the internet, delivering TV shows and movies to subscribers (Compl. ¶57). The service utilizes the H.264/AVC video compression standard to encode and transcode video streams (Compl. ¶¶18, 40).
- Evidence cited in the complaint indicates that Hulu streams content using the MPEG-DASH format, and that the associated manifest files reference the "avc1" (H.264) video codec (Compl. ¶19, ¶41). The complaint provides a screenshot of a DASH manifest file showing multiple representations with codecs identified as "avc1" (Compl. ¶19, p. 7).
- The service is alleged to aggregate content from various sources and present it to users in distinct collections. The complaint includes screenshots of the Hulu user interface showing rows of content labeled "Keep Watching," "Line Up," "For You," and "New Episodes," which represent different collections of presentations (Compl. ¶¶58, 60, 61, 67). One screenshot shows the "Keep Watching" row, which displays content a user has started but not finished (Compl. ¶58, p. 27).
IV. Analysis of Infringement Allegations
6,519,005 Infringement Allegations
| Claim Element (from Independent Claim 1) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| comparing pixels of a first pixel array... with pixels of a plurality of second pixel arrays in at least one reference picture and | The Accused Infringing Devices' H.264 encoders compare pixels of a macroblock in a picture being coded with pixels of pixel arrays in at least one reference picture (Compl. ¶20). | ¶20 | col. 2:50-54 |
| concurrently performing motion estimation for each of a plurality of different prediction modes in order to determine which... is an optimum prediction mode; | Hulu's H.264 encoders use different motion estimation modes (inter-modes) and concurrently perform motion estimation for these modes to determine the optimum one. The complaint includes a "Mode Decision" diagram from a technical presentation illustrating this process (Compl. ¶21, p. 9). | ¶21 | col. 3:10-15 |
| determining which of the second pixel arrays constitutes a best match with respect to the first pixel array for the optimum prediction mode; and, | The encoder determines the best match by selecting the prediction mode with the least rate-distortion cost (Compl. ¶25). A diagram shows the encoder finding the "best matching block" in a reference frame (Compl. ¶24, p. 13). | ¶24-25 | col. 2:65-3:2 |
| generating a motion vector for the first pixel array in response to the determining step. | The encoder calculates and generates motion vectors for the macroblock, which are then represented in the video data stream (Compl. ¶26). | ¶26 | col. 7:5-8 |
Identified Points of Contention
- Scope Questions: A central question is whether the term "concurrently performing motion estimation", as described in the ’005 Patent for MPEG-2 frame/field modes, can be construed to read on the H.264 "Mode Decision" process for different macroblock partition sizes as alleged in the complaint.
- Technical Questions: What evidence does the complaint provide that Hulu's encoders perform the "concurrent" estimation using the specific parallel architecture taught in the patent's embodiments (e.g., Fig. 3), as opposed to simply evaluating multiple modes as part of a standardized, albeit computationally intensive, decision process? The complaint relies on high-level diagrams of the H.264 standard, not specific details of Hulu's implementation.
6,895,118 Infringement Allegations
| Claim Element (from Independent Claim 1) | Alleged Infringing Functionality | Complaint Citation | Patent Citation |
|---|---|---|---|
| an estimation step, for macroblocks, of a capacity to be reconstructed via an error concealment method, | Before a macroblock is coded in H.264, an estimation is made as to whether it can be reconstructed via error concealment by examining its motion characteristics and checking if the resulting prediction contains no non-zero quantized transform coefficients (Compl. ¶42). | ¶42 | col. 1:7-12 |
| a decision... to exclude a macroblock from coding being made on the basis of the capacity of such macroblock to be reconstructed, | H.264 encoders perform a decision to "skip" a macroblock, thereby excluding it from coding, based on its capacity to be reconstructed by inferring its properties from neighboring macroblocks (Compl. ¶43). | ¶43 | col. 4:13-19 |
| ...inserting a resynchronization marker into the binary data stream after the exclusion of one or more macroblocks. | Skipped macroblocks in H.264 are communicated using an mb_skip_flag. The complaint alleges this flag is a "resynchronization marker" inserted at the point where the macroblock is not coded (Compl. ¶44). A diagram from an H.264 technical document shows a "Skip indication" within the slice layer (Compl. ¶44, p. 22). |
¶44 | col. 2:8-12 |
Identified Points of Contention
- Scope Questions: The primary dispute will likely concern the construction of "resynchronization marker". The infringement theory equates the H.264
mb_skip_flag—a single bit indicating a macroblock is skipped—with this term. The defense may argue that the patent's specification describes a "resynchronization marker" as a more substantial element, like a packet header, used to recover from transmission errors over a larger segment of data, not a flag for a single macroblock.
V. Key Claim Terms for Construction
Patent 6,519,005
- The Term: "concurrently performing motion estimation for each of a plurality of different prediction modes"
- Context and Importance: This term is the central inventive concept. The infringement case rests on whether the H.264 "mode decision" process falls within this definition. Practitioners may focus on this term because the patent’s specific examples relate to MPEG-2, while the accused product uses the later H.264 standard.
- Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The specification states the invention can be used with "any digital video transmission or storage system that employs a video compression scheme... in which motion coding with multiple prediction modes is used," suggesting applicability beyond the specific embodiments (’005 Patent, col. 4:3-9).
- Evidence for a Narrower Interpretation: The detailed description and Figure 3 depict a specific hardware architecture with four parallel search engines whose error metric outputs are combined by adders (’005 Patent, Fig. 3; col. 7:50-8:2). This may support a narrower construction tied to this specific method of parallel processing and metric combination.
Patent 6,895,118
- The Term: "resynchronization marker"
- Context and Importance: The viability of the infringement allegation hinges on this term's definition, specifically whether an
mb_skip_flagin H.264 qualifies. If it does not, the claim is not infringed. - Intrinsic Evidence for Interpretation:
- Evidence for a Broader Interpretation: The abstract and claim language broadly recite "inserting a resynchronization marker" after excluding macroblocks, which a plaintiff might argue covers any bit or signal that logically marks the position of the excluded data for the decoder (’118 Patent, Abstract; col. 8:10-12).
- Evidence for a Narrower Interpretation: The background section discusses "resynchronization markers" in the context of MPEG-4 as tools that are "periodically inserted" to recover from data loss and are part of a larger "video packet" structure that includes a header (’118 Patent, col. 1:35-51). This suggests a more substantial data structure than a single bit flag within a slice.
VI. Other Allegations
- Indirect Infringement: The complaint focuses on allegations of direct infringement under 35 U.S.C. § 271(a) by "making, using, offering for sale, selling and/or importing" the accused instrumentalities (Compl. ¶¶ 2, 27, 45, 70). It does not plead specific facts to support claims of induced or contributory infringement.
- Willful Infringement: The complaint includes a prayer for relief seeking enhanced damages for willful infringement (Compl., p. 37, ¶e). However, the body of the complaint does not allege facts typically used to support willfulness, such as pre-suit knowledge of the asserted patents.
VII. Analyst’s Conclusion: Key Questions for the Case
- Patent Validity Post-Filing: The most significant issue is that the asserted claims for both the ’005 and ’273 patents were cancelled in IPR proceedings that concluded after the complaint was filed. This development raises the fundamental question of whether any viable case remains for these two patents, as infringement of a cancelled claim is not possible.
- Claim Scope and Technical Equivalence: For the remaining ’118 patent, a core issue will be one of definitional scope: can the term "resynchronization marker", described in the patent as a mechanism for error recovery within a video packet, be construed to cover the
mb_skip_flagin the H.264 standard, which functions as a simple indicator that a single macroblock was not transmitted? - Evidentiary Burden: Should the claim against the '005 patent proceed despite its cancellation, a key evidentiary question would be one of functional operation: does the accused H.264 mode decision process operate in the specific "concurrent" manner taught in the '005 patent’s embodiments, or do the complaint's allegations, based on high-level technical standards, fail to show a true technical overlap with the patented method?