GOOGLE LLC

Patent Trials and Appeals Board

Mar 29, 2022

2020005168 (P.T.A.B. Mar. 29, 2022)

UNITED STATES PATENT AND TRADEMARK OFFICE UNITED STATES DEPARTMENT OF COMMERCE United States Patent and Trademark Office Address: COMMISSIONER FOR PATENTS P.O. Box 1450 Alexandria, Virginia 22313-1450 www.uspto.gov APPLICATION NO. FILING DATE FIRST NAMED INVENTOR ATTORNEY DOCKET NO. CONFIRMATION NO. 15/668,745 08/04/2017 Ruijie Xu GOGL-1241-A 3466 97818 7590 03/29/2022 Google LLC c/o Young Basile Hanlon & MacFarlane, P.C. 3001 West Big Beaver Rd., Ste. 624 Troy, MI 48084-3107 EXAMINER HAGHANI, SHADAN E ART UNIT PAPER NUMBER 2485 NOTIFICATION DATE DELIVERY MODE 03/29/2022 ELECTRONIC Please find below and/or attached an Office communication concerning this application or proceeding. The time period for reply, if any, is set in the attached communication. Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the following e-mail address(es): audit@youngbasile.com docketing@youngbasile.com PTOL-90A (Rev. 04/07) UNITED STATES PATENT AND TRADEMARK OFFICE BEFORE THE PATENT TRIAL AND APPEAL BOARD Ex parte RUIJIE XU and DAKE HE Appeal 2020-005168 Application 15/668,745 Technology Center 2400 Before ROBERT E. NAPPI, ST. JOHN COURTENAY III, JUSTIN BUSCH, Administrative Patent Judges. BUSCH, Administrative Patent Judge. DECISION ON APPEAL STATEMENT OF THE CASE Appellant1 appeals under 35 U.S.C. § 134(a) from a Final Rejection of claims 1-10 and 13-22. We have jurisdiction over the remaining pending claims under 35 U.S.C. § 6(b). We REVERSE. 1 Throughout this Decision, we use the word “Appellant” to refer to “applicant” as defined in 37 C.F.R. § 1.42 (2018). Appellant identifies Google LLC as the real party in interest. Appeal Br. 3. Appeal 2020-005168 Application 15/668,745 2 CLAIMED SUBJECT MATTER Appellant’s invention generally relates to video coding and, more particularly, encoding or decoding data to obtain a first subset of elements in a block of video data using a context selected based on a category that is determined based on a second subset of decoded elements in the block. See, e.g., Spec. ¶¶ 4-8. Claims 1, 14, and 16 are independent. Claim 1 is representative of the subject matter on appeal and is reproduced below: 1. A system for decoding video, comprising: a memory; and a processor, wherein the memory stores instructions executable by the processor to cause the system to: partition a block of video data into a plurality of groups of elements; decode, using an entropy decoder, data from an encoded bitstream to obtain elements of a first group from the plurality of groups of elements; determine a category based on the elements of the first group; based on the category, select a context for an element from a second group from the plurality of groups of elements; and decode, using the entropy decoder using the context, data from the encoded bitstream to obtain the element of the second group from the plurality of groups of elements. REJECTIONS Claims 1-8, 10, and 14-22 stand rejected under 35 U.S.C. § 103 as being obvious over Zheng (US 9,445,093 B2; Sept. 13, 2016), Rojals (US 9,197,890 B2; Nov. 24, 2015), and Tu (US 9,386,306 B2; July 5, 2016). Final Act. 2-11. Appeal 2020-005168 Application 15/668,745 3 Claim 9 stands rejected under 35 U.S.C. § 103 as being obvious over Zheng, Rojals, Tu, and Fu (US 2016/0277746 A1; Sept. 22, 2016). Final Act. 12-13. Claim 13 stands rejected under 35 U.S.C. § 103 as being obvious over Zheng, Rojals, Tu, and He (US 2016/0080749 A1; Mar. 17, 2016). Final Act. 13-14. ANALYSIS2 REJECTION OF CLAIMS 1-8, 10, AND 14-22 The Examiner finds the combination of Zheng, Rojals, and Tu teaches or suggests every limitation recited in independent claim 1. Final Act. 3-6. In general, the Examiner finds Zheng teaches a video decoding system with a processor and a memory storing instructions that cause the system to partition a video block into multiple groups of elements, decode data from a bitstream to obtain elements of a first group, determine a category, select a context for an element in a second group, and, using the context, decode data from the bitstream to obtain the element in the second group. Final Act. 3-4 (citing Zheng 5:25-38, 18:15-19:35, 12:65-14:36, 15:14-22, Figs. 3, 5-7). The Examiner finds Zheng does not teach that determining a context is based on the category, as recited in claim 1, but that Rojals teaches that 2 We note Appellant did not identify any related matters to this Appeal. However, we are aware of at least Appeal 2020-002872 of a rejection in Application No. 15/671,595, which has a specification that is nearly identical to the Specification in this matter. The anticipation rejection over a different reference was affirmed in that case. We remind Appellant of the duty of candor under 37 C.F.R. § 1.56 and the duty under 37 C.F.R. § 41.37(c)(ii) to identify applications and appeals that are owned by Appellant and that “may be related to, directly affect or be directly affected by or have a bearing on the Board's decision in the pending appeal.” Appeal 2020-005168 Application 15/668,745 4 selecting a context, which used for decoding the data to obtain a second element, is based on a category. Final Act. 4 (citing Rojals 20:40-60, 22:42-23:6, 31:5-15, 31:55-65, Figs. 10, 11, 14, 16-18). The Examiner finds Tu also teaches selecting a context, which is used to decode data to obtain a second element, based on a category. Final Act. 4 (citing Tu 21:40- 50, 22:1-23). The Examiner further finds Tu teaches or suggests that determining a category is based on the elements of the first group. Final Act. 5 (citing Tu 19:45-21:17). Our analysis focuses on whether the rejection demonstrates that the prior art teaches “determin[ing] a category based on the elements of the first group,” as recited in claim 1. The Examiner finds Zheng and Tu teach this limitation and Rojals “is not relied upon to teach this limitation.” Ans. 20. More specifically, the Examiner finds Zheng’s disclosure of determining a scan order from a signaled prediction mode teaches or suggests determining a category, but Zheng does not teach that the category is determined based on the elements in the first group. Final Act. 3; Ans. 20 (citing Zheng 5:25- 38, 15:14-22); see also Ans. 21-22 (finding that scan orders correspond to categories and, therefore, Zheng determines a category because it determines a category from the prediction mode). The Examiner finds Tu cures this deficiency because Tu discloses using a reference block to determine a scan order and context for a current block. Final Act. 4; Ans. 20, 22-28. The Examiner determines that an ordinarily skilled artisan would have modified Zheng with Tu’s teachings because Tu teaches that using texture information improves coding efficiency and Zheng already contemplates adaptively adjusting coding “based on coefficients that are already coded.” Final Act. 5-6; Ans. 26-27 (quoting Zheng 5:38-42). Of note, the Examiner Appeal 2020-005168 Application 15/668,745 5 determines that, “[b]ecause a current block is similar to its reference block, classification of the reference block accurately reflects the current block” and applying the technique to “the current block itself, the results would accurately reflect the current block itself.” Ans. 23-24, 41. Zheng discloses breaking a block of residual transform coefficients into a plurality of zones and applies a multiple scanning order. Zheng 12:65-67. One example of multiple zones in a block is depicted in Figure 5, reproduced here: Appeal 2020-005168 Application 15/668,745 6 Zheng, Fig. 5 (depicting “a multiple zone scanning process used in video encoding” 2:40-41). As depicted in Figure 5, Zheng divides block of coefficients 61 into frequency zones 0, 1, and 2. Zheng 13:2-4. Zheng then scans “[a]ll transform coefficients in a particular frequency zone . . . before scanning transform coefficients in any other frequency zone.” Zheng 13:19-21. Zheng scans the coefficients from each zone together into a corresponding section of a one-dimensional array-for example, Zheng may scan the coefficients from Zone 0 into section 63, the coefficients from Zone 1 into section 64, and the coefficients from Zone 2 into section 65. Zheng 13:29-39. Tu relates to “determin[ing] a value of a current video unit associated with the reference block based on, at least in part, a classification of the reference block and a scan order selected by the processor based upon the classification.” Tu, Abstract; see also Tu 19:36-40 (“FIG. 5 describes one embodiment of a method to determine a scan order of an enhancement layer block based upon a classification of a reference block (e.g., from base layer or reference layer) associated with the enhancement layer block.”). Tu explains that its technique for determining a scan order can replace traditional techniques for determining scan order. Tu 2:66-3:6. Generally, Tu classifies a reference block into, for example, one of “vertical texture dominated,” “horizontal texture dominated,” or neither vertical nor horizontal texture dominated. Tu 19:45-54. Tu may classify the reference block in the pixel domain, see Tu 19:62-20:37, or in the transform domain, see Tu 20:38-21:16. To determine the horizontal and vertical strengths of the reference block in the transform domain, Tu may sum the coefficients in the reference block’s first row and sum the coefficients in the first column Appeal 2020-005168 Application 15/668,745 7 and compare the sums to threshold values. Tu 20:38-21:16. Tu derives a scan order and context model for a current block based on how it classifies the reference block associated with the current block. Tu 21:17-46. Appellant argues, among other things, that a person of ordinary skill in the art would not have modified Zheng with Tu to arrive at the claimed limitation of determining a category of a current block based on decoded elements of that same block. See, e.g., Appeal Br. 10; Reply Br. 4-5. More specifically, Appellant argues that the Examiner has not demonstrated why an ordinarily skilled artisan, without the benefit of improper hindsight, would have applied Tu’s teachings-i.e., determining, based on a reference block, a category used to determine a scan order for an entire video block- to modify Zheng so that the system determines, based on one portion of the current block, a category used to determine a scan order for a different portion of the same current block. Reply Br. 4-5. The Examiner finds an ordinarily skilled artisan would have incorporated Tu’s teachings into Zheng to determine an optimal scanning order and improve coding efficiency. Final Act. 5-6. The Examiner further states that “the reference block accurately reflects the current block” because they are similar to each other and when applying the technique “on the current block itself, the results would accurately reflect the current block.” Ans. 24. The Examiner finds Zheng’s Zone 1 and Zone 2 “are ripe for improvement with adaptive scanning based on the classification of Tu.” Ans. 26-27. To the extent the Examiner finds that a person of ordinary skill in the art would have understood Tu’s technique could provide a better scan order and more efficient coding for a block as compared to traditional techniques Appeal 2020-005168 Application 15/668,745 8 for determining scan order, we agree. See Tu 2:66-3:6. We also agree that Zheng teaches separately coding each of its frequency zones and further teaches that a scan order can be adjusted based on already-coded coefficients. See, e.g., Zheng 5:26-42, 13:19-21. However, Zheng goes on to explain that the already-coded coefficients may be “in similar types of video blocks previous[ly] coded.” Zheng 5:42-53. Furthermore, because Zheng discloses that scan order may be based on already-coded coefficients in similar types of video blocks and appears to rely on known techniques for determining scan order, a person of ordinary skill in the art could have used Tu’s technique to determine scan order. It is possible that a person of ordinary skill in the art would have used the coded information from one frequency zone of a video block to categorize that video block and determine a scan order for the other frequency zone(s) in that block based in light of (1) Tu’s disclosure of categorizing a reference block using decoded information from a first row and a first column of the block and determining a scan order for a block associated with the reference block based on that category and (2) Zheng’s disclosure of dividing a block into frequency zones and coding one zone before moving on to the next zone. However, without further reasoning and explanation, we agree with Appellant that the combination supported by the record would merely incorporate Tu’s technique that uses information from a reference block to categorize the reference block and determine a scan order for the current block and that the Examiner’s finding of motivation relies on impermissible hindsight. Appeal Br. 9; Reply Br. 5. Accordingly, on this record and absent impermissible hindsight, the Examiner’s finding of Appeal 2020-005168 Application 15/668,745 9 motivation to combine Zheng and Tu to arrive at Appellant’s claimed invention is unsupported by a preponderance of the evidence. Therefore, constrained by the record, we do not sustain the rejection of claim 1 as obvious over Zheng, Rojals, and Tu. Independent claims 14 and 16 include a limitation of similar scope and claims 2-8, 10, 15, and 17-22 ultimately depend from and incorporate the limitations of one of claims 1, 14, and 16. Accordingly, for the same reasons as discussed above, we also do not sustain the rejection of claims 2-8, 10, 15, and 17-22 as obvious over Zheng, Rojals, and Tu. Claims 9 and 13 each depend directly from and incorporate the limitations of claim 1, and the Examiner does not find the additionally recited references-Fu with respect to claim 9 and He with respect to claim 13-cure the deficiencies identified with respect to independent claim 1. Accordingly, for the same reasons, we do not sustain the rejection of claim 9 as obvious over Zheng, Rojals, Tu, and Fu, and we do not sustain the rejection of claim 13 as obvious over Zheng, Rojals, Tu, and He. CONCLUSION Under 35 U.S.C. § 103, Appellant has shown that the Examiner erred in rejecting claims 1-10 and 13-22 over the cited combinations of references. Appeal 2020-005168 Application 15/668,745 10 DECISION SUMMARY Claim(s) Rejected 35 U.S.C. § Reference(s)/ Basis Affirmed Reversed 1-8, 10, 14-22 103 Zheng, Rojals, Tu 1-8, 10, 14-22 9 103 Zheng, Rojals, Tu, Fu 9 13 103 Zheng, Rojals, Tu, He 13 Overall Outcome 1-10, 13-22 REVERSED