Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Detailed Action
Claims 1-37 are pending in this Office Action.
Claims 1, 8, 17, 21, and 28 are amended.
Claims 35-37 are new.
Response to Arguments
Applicant’s arguments filed in the amendment filed 3/20/2026, have been fully considered but are not persuasive. The reasons set forth below.
In responding the applicant’s remarks, on page 10, applicant notes the status of the rejections on the claims. The examiner appreciates the response and will clarify the rejection in the instant action.
Regarding claims 8-16 and 21-27, as discussed in the interview 1/15/26, the language of the claims has been searched but it is noted that the further amendments towards the training of neural networks may necessitate a need for a restriction.
On page 11 of the arguments, applicant argues the Kale reference does not teach the amended steps of claim 1.
The examiner disagrees.
Kale discloses detecting illicit imagery before display in col. 3, lines 17-34. Streamed media is buffered and then ‘analytics capability can be configured in the memory device to intelligently identify and filter out unwanted images/audio in real time so kids and/or other passengers in a vehicle or other uses of the media player, are not exposed to the unwanted images/audio.’ Clearly, the art shows filtering out before display of content.
Kale then goes on to show different artificial neural networks are trained to detect and classify unwanted content (col. 3, lines 35-51).
The claim then goes to claim both a ‘feature map’ which is merely described as a ‘representation’ of a likelihood that a specific pixel or region of the one or more gameplay images depicts ‘illicit information’ and ‘classifying’ an image.
Kale teaches both as they are extremely broadly described. Kale teaches (in col. 6, lines 8-35), ‘video frames may contain regions of interest for classification’ and that video frames are scored to a threshold of confidence, that areas are unwanted or objectionable content.
Kale teaches that we are both looking at the unwanted objects but also the classification of these unwanted objects (Kale: col. 4, lines 40-65) “the video processing device has configurable preference that identifies unwanted content based on classifications. Based on the classification and the preferences, the video processing device can adjust the output video frames generated by the process to prevent the presentation of unwanted content.”
Kale is only silent about the claimed descriptor ‘gameplay images.’ This term persists after the practitioner amended the claim language to change from claiming ‘cheating information’ to the broader ‘illicit’ information. The Pardeshi reference is introduced to address this phrasing.
Applicant’s invention as claimed:
Claim Rejections - 35 USC § 112
The following is a quotation of the second paragraph of 35 U.S.C. 112:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter, which the applicant regards as his invention.
Claim 36 rejected under 35 U.S.C. 112, second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which applicant regards as the invention.
Claim 36 recites the limitation " at least one of input nodes or output nodes of the one or more neural networks that are associated with a number of pixels in the one or more gameplay images" in lines 2-3. The language is unclear and indefinite. Is the at least one of the input or output or the at least one of the input and output and the number of pixels. Some clarifying language is needed.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or non-obviousness.
Claims 1-9; 17-21; 28-32; 34-35 are rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent No. 11636339 by Kale et al in view of 20210232907 by Pardeshi et al.
Regarding claim 1, (Currently Amended) Kale teaches:
one or more processors comprising:
circuitry to:
intercept one or more gameplay images of a computer game before display (Kale: col. 3, lines 17-34; gameplay images are consumed like video streams);
prior to display of the one or more gameplay images, use one or more neural networks to generate:
a feature map based on the one or more gameplay images of the computer game (Kale: col. 3, lines 35-51), the feature map including a representation of a likelihood that a specific pixel or region of the one or more gameplay images depicts illicit information (Kale: col. 6, lines 21-35); and
a classification of the one or more gameplay images (Kale: col. 4, lines 40-50); and
use the one or more neural networks to detect the illicit information used by one or more users of the computer game based, at least in part, on the feature map and the classification of the gameplay images (Kale: col. 4, lines 40-65).
Kale fails to teach gameplay media but instead mentions audio, video and image data that is similar.
However, in analogous art, the Pardeshi reference teaches using neural networks to analyze gameplay imagery (Pardeshi: page 3, para 55) in order to use artificial intelligence to ensure fair gameplay (Pardeshi: page 2, para 48).
It would have been obvious to one of ordinary skill in the art before the effectively filed date to include the gameplay analysis of Pardeshi with the feature map and image classification of Kale in order to use artificial intelligence to ensure fair gameplay (Pardeshi: page 2, para 48).
2. The one or more processors of claim 1, wherein the circuitry is further to:
store the one or more images in a buffer, wherein the stored one or more images are to be provided as input to the one or more neural networks, and are to be rendered on a display unit (Kale: col. 3, lines 18-29).
Kale fails to teach gameplay images but instead mentions audio, video and image data that is similar.
However, in analogous art, the Pardeshi reference teaches using neural networks to analyze gameplay imagery (Pardeshi: page 3, para 55) in order to use artificial intelligence to ensure fair gameplay (Pardeshi: page 2, para 48).
It would have been obvious to one of ordinary skill in the art before the effectively filed date to include the gameplay analysis of Pardeshi with the feature map and image classification of Kale in order to use artificial intelligence to ensure fair gameplay (Pardeshi: page 2, para 48).
3. The one or more processors of claim 2, wherein the circuitry is further to:
generate an indication of detection of illicit information; and communicate the indication to a server (Kale: col. 11, lines 57 -col. 12, line 13).
4. The one or more processors of claim 3, wherein the circuitry is further to:
generate, using the one or more neural networks, a confidence level characterizing confidence that the one or more gameplay images comprise the illicit information (Kale: col. 6, lines 21-35).
5. The one or more processors processor of claim 4,
wherein the report is communicated to the game server responsive to determining that the confidence level is at or above a threshold value (Kale: col. 6, lines 21-35).
6. The one or more processors of claim 3, wherein the circuitry is further to:
receive, from the game server, updated parameters for the one or more neural networks, wherein the updated parameters are generated based on a set of retraining images (Kale: col. 18, lines 27-55).
7. The one or more processors of claim 1, wherein the circuitry is to generate a certification signal to a server, wherein the certification signal is to certify to the server that the circuitry is one or more circuits are capable of confidence in scoring (Kale: col. 6, lines 21-51; confidential threshold scores are filtered and manual input for retraining; col. 11, lines 57- col. 12, line 8).
Claim 8 is substantially similar to claim 1.
Regarding claim 9, The Kale reference teaches detecting illicit information.
The Kale reference fails to teach cheating software associated with a computer game.
The Pardeshi reference teaches finding cheating images are generated using a cheating software associated with the computer game (Pardeshi: page 3, para 55-59).
It would have been obvious to one of ordinary skill in the art, before the effectively filed date, to include the cheating detection as taught by Pardeshi with the detection of Kale in order to classify behaviors by players and keep players motivated (Liu: page 1, para 4, 10-11).
Regarding 34, the Kale reference teaches, the one or more processors of claim 1, wherein the illicit information comprises at least one of information that provides an unfair advantage or unauthorized information (Pardeshi: page 3, para 55-59).
35. (New) The one or more processors of claim 1, wherein the circuitry is further to cause a graphics processing unit to render the one or more gameplay images for display and use by the one or more neural networks (Pardeshi: page 3, para 54).
Claim 17 is substantially similar to claim 1.
Claim 18 is substantially similar to claim 2.
Claim 18 is substantially similar to claim 3.
Claim 19 is substantially similar to claim 5.
Claim 20 is substantially similar to claim 4.
Claim 21 is substantially similar to claim 1.
Claims 28-31 are rejected as being substantially similar to above claims 1-4.
Claims 10-14; 22-26; 32 are rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent No. 11636339 by Kale et al. in view of 20210232907 by Pardeshi et al. in further view of 20190130218 by Albright et al.
Regarding claim 10, modified Kale teaches a system of detecting illicit information in images.
The modified Kale fails to teach details of training the neural network with image data but explains how its trained based on detection of certain image information (Kale: col. 2, lines 20-27 and col. 6, lines 21-51).
The Albright reference teaches wherein training of the one or more neural networks is further based, at least in part, on one or more positive outcome images, wherein each of the one or more positive outcome images is devoid of the negative information (Albright: page 1, para 2; page 2, para 16).
It would have been obvious to one of ordinary skill in the art, before the effectively filed date, to include the classification training of images as taught by Albright with the detection of Kale in order to classify images properly (Albright: page 1, para 3-4).
Claim 22 is rejected as being substantially similar to claims 10 above.
Claim 32 is rejected as being substantially similar to claim 10 above.
Regarding claim 11,
The modified Kale reference teaches one or more processors of claim 10, wherein at least a subset of the one or more images comprises images comprising the illicit information augmented with information from non-illicit information that are generated by a gaming software associated with the computer game (Kale: col. 7, lines 63- col. 8, lines 10; transformed content; Pardeshi: page 3, para 55-59; overlay).
12. The one or more processors of claim 11, wherein each of the subset of the one or more gameplay images comprising the illicit information comprises a part that is replaced with a part of a non-illicit image. (Kale: col. 7, lines 63- col. 8, lines 10; transformed content; Albright: page 2, para 12).
Regarding claim 13, (Currently Amended) The modified Kale reference teaches the one or more processors of claim 10, wherein at least a subset of the one or more non-illicit images comprises non-illicit images augmented with information from the one or more gameplay images comprising the illicit information generated by a gaming software associated with the computer game (Pardeshi: page 3, para 55-59; overlaps and unshown; para 64, audio data).
Regarding claim 14, (Currently Amended) The one or more processors of claim 13.
The modified Kale reference fails to teach replacing part of images.
However, in analogous art, the Albright reference teaches
each of the subset of the one or more non-illict images comprises a part replaced with a part of an image comprising the illicit information (Albright: page 2, para 19-20 teaches replacing parts of images to make recognition training with positive and negative evaluations more accurate).
It would have been obvious to one of ordinary skill in the art, before the effectively filed date, to include the image manipulation as taught by Albright with the image detection of Kale in order to more accurately train the neural network and detect irregularities (Albright: page 1, para 3-4).
Claims 23 is rejected as being substantially similar to claims 11 above.
Claims 24 is rejected as being substantially similar to claims 12 above.
Claims 25 is rejected as being substantially similar to claims 13 above.
Claims 26 is rejected as being substantially similar to claims 14 above.
Claims 15-16, 27; 33 are rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent No. 11636339 by Kale et al. in view of 20210232907 by Pardeshi et al. in further view of US Patent Publication No 20210051162 by Taylor et al.
Regarding claims 15, the modified Kale reference teaches using neural networks to detect illicit images.
The modified Kale reference focuses on abnormality detection but fails to explicitly state adversarial attacks.
The Taylor reference teaches using neural networks to train against adversarial attacks (Taylor: Figure 2, page 4, para 31-32) in order to more easily detect malware signatures (Taylor: page 1, para 2).
It would have been obvious to one of ordinary skill in the art, before the effectively filed date, to include the adversarial detection as taught by Taylor with the neural network of modified Kale in order to more easily detect malware signatures (Taylor: page 1, para 2).
16. (Currently Amended) The one or more processors of claim 15, wherein the training of the one or more neural networks against adversarial attacks is based, at least in part, on a subset of the one or more cheating images modified with adversarial perturbations (Kale teaches detecting irregularities in images, such abnormalities are likened to the malicious attempts of Taylor; Taylor: Figure 2, page 4, para 31-3).
Claim 27 is rejected as being substantially similar to claims 15 above.
Claim 33 is rejected as being substantially similar to claim 15 above.
Claim 36 is rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent No. 11636339 by Kale et al. in view of 20210232907 by Pardeshi et al. in further view of 6028956 by Shustorovich et al.
Regarding claim 36, the modified Kale teaches (New) The one or more processors of claim 1.
The modified fail reference fails to detail the feature map with nodes.
However, in analogous art, the Shustorovich reference teaches a feature map is generated using a number of at least one of input nodes or output nodes of the one or more neural networks that are associated with a number of pixels in the one or more gameplay images (Shustorovich: col. 15, lines 13-50) in order to provide a simple and fast yet accurate way of recognizing images and objects by a neural network (Shustorovich: col. 5, lines 42-59).
It would have been obvious to one of ordinary skill in the art, before the effectively filed date, to include the node size driven representation as taught by Shustorovich with the neural network of modified Kale in order to provide a simple and fast yet accurate way of recognizing images and objects by a neural network (Shustorovich: col. 5, lines 42-59).
Claim 37 is rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent No. 11636339 by Kale et al. in view of 20210232907 by Pardeshi et al. in further view of US Patent Publication No. 20190340462 by Pao et al
Regarding claim 37, the modified Kale teaches (New) The one or more processors of claim 1, wherein the representation of the likelihood is scored.
The modified Kale reference teaches pixel level analysis but not a score for each pixel.
However, in analogous art, the Pao reference teaches a representation comprises at least one of a score for the specific pixel or scores for each pixel in the region of pixels (Pao: page 4, para 34) in order to help identify pixels that make up objects in images (Pao: page 1, para 1-3).
It would have been obvious to one of ordinary skill in the art, before the effectively filed date, to include the pixel scoring as taught by Pao with the neural network of modified Kale in order to help identify pixels that make up objects in images (Pao: page 1, para 1-3).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BENJAMIN R BRUCKART whose telephone number is (571)272-3982. The examiner can normally be reached M-TH: 7-6p.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
BENJAMIN R. BRUCKART
Supervisory Patent Examiner
Art Unit 2424
/BENJAMIN R BRUCKART/Supervisory Patent Examiner, Art Unit 2424
Prosecution Insights