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 .
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 3/9/26 has been entered.
Claim Rejections - 35 USC § 103
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.
Claim(s) 1, 2, 4, 8, 9, 11, 15, 16 and 18 is/are rejected under 35 U.S.C. 103 as being unpatentable over Harkness (10,885,343) in view of Feng (CN114095754A).
Regarding claim 1, Harkness discloses an apparatus comprising: at least one memory; and at least one processor coupled to the at least one memory, the at least one processor configured to:
receive a set of video frames (205);
identify a discontinuity in the set of video frames (215);
generate one or more replacement frames associated with the discontinuity based on at least one video frame selected from among the set of video frames (220); and
provide the one or more replacement frames to a user (225).
However, Harkness does not disclose that wherein the one or more replacement frames are generated by a machine-learning model trained to create replacement frames based on contextual information, the contextual information including at least event data as now claimed.
Feng, from the similar field of endeavor, teaches that the one or more replacement frames are generated by a machine-learning model trained to create replacement frames based on contextual information, the contextual information including at least event data (note page 8-14 of the translated text).
Harkness teaches that the replacement frame can be generated by conventional trained neural network (note col. 4, lines 24-27).
Thus, in view of Feng, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to include machine learning model of Feng into Harkness to perform the well known replacement frame generation based on the contextual information including the event data as claimed.
In addition, Feng does not explicitly disclose the newly included limitation: “behaviors occurring before or after an identified discontinuity”.
However, in page 9 of the translated text, Feng discloses the following:
example, if the video section to be parsed comprises a video frame C and a video frame D. the video frame C and the video frame D are the image of the human riding, the multi-mode information extracted from the video frame C by the electronic device comprises: Scene information: grassland; Role information: people and horses; The position information of the role: a pixel position in the video frame; The posture information of the role: Two water chestnut (“water chestnut” is incorrection translation; it should be “horseshoes”) falls on the ground, and two horseshoes are not on the ground. The multi- modality information extracted from the video frame D by the electronic device includes: Scene information: grassland; Role information: people and horses; The position information of the role: a pixel position in the video frame; The posture information of the role: The four water chestnut (horseshoes) falls on the ground. the electronic device according to the multi-mode information of the video frame C and the video frame D from the plurality of scenario events matching the target scenario event is used for representing character movement event. The semantic information of the video segment to be parsed generated according to the target scenario event is: People ride on the grassland and run on the grassland.
step 1043, according to the semantic information, determining the second number of video frame to be supplemented between two adjacent display of the second video frame.
According to above passages, the people whose riding horses and running on the grassland can be interpreted as behavior of the people because riding and running are being performed by the people. Thus, above quoted passages meet the “behavior” limitation as claimed.
Regarding claim 2, Harkness discloses to generate the one or more replacement frames, the at least one processor is configured to: provide at least one video frame selected from among the set of video frames to a generative machine-learning model; and receive the one or more replacement frames from the generative machine-learning model (note col. 5, lines 42-50).
Regarding claim 4, Harkness discloses that the generative machine-learning model is camera-specific (205 and 210).
Regarding claims 8, 9, 11, 15, 16 and 18, see similar rejections as set forth above.
Claim(s) 3, 7, 10, 14, and 17 is/are rejected under 35 U.S.C. 103 as being unpatentable over Harkness (10,885,343) in view of Feng (CN114095754A), further in view of Klein et al. (2021/0049409).
Regarding claim 3, Harkness does not disclose that the generative machine-learning model is trained using video frames collected by two or more imaging devices that have an overlapping field of view; however, Harkness does teach that the replacement frames may be generated by a generative network (col. 5, lines 48-50).
Klein, from the similar field of endeavor, teaches that when more than two or more cameras shares an overlapping field of view, video frames and any information or attributes generated from one camera can be provided and utilized by other cameras to train and/or improve a machine-learning model to classify and detect objects in image data or video frames (par. 16, 24, 29 and 31). In other words, missing frames of one camera can be generated based on information from another camera by using the machine-learning model (par. 51). This is more precise than by just using one camera alone.
Thus, in view of Klein, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to include Klein into the combination of Harkness and Feng so that the machine learning model could be improved by utilizing overlapping video frames generated from two or more cameras.
Regarding claim 7, Harkness does not disclose that the at least one processor is further configured to: receive an input from the user, the input providing a quality indication for the one or more replacement frames; however, Harkness does teach that the replacement frames may be generated by a generative network (col. 5, lines 48-50).
Klein teaches a labeled training data, such as image and/or video data annotated by humans (par. 16). The annotated data can be used to indicate quality of the video frames generated. Thus, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to include Klein into Harkness to perform the well known functions as claimed.
Regarding claims 10, 14, and 17, see similar rejections as set forth above.
Claim(s) 5, 6, 12, 13, 19 and 20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Harkness (10,885,343) in view of Feng (CN114095754A), further in view of Reda et al. (2021/0067735).
Regarding claim 5, Harkness does not disclose that to generate the one or more replacement frames, the at least one processor is configured to: provide audio data to a generative machine-learning model. Reda, from the similar field of endeavor, teaches that the audio data is provided to the generative machine-learning model (note par. 80).
Therefore, knowing that audio is required in the missing frames of Harkness, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to include Reda into Harkness so that the audio data could be generated through the machine-learning model.
Regarding claims 6, 13 and 20, Harkness or Feng does not disclose that the machine-learning model is trained by applying a loss function to compare predicted output values with target output values as claimed. Reda, teaches such training (note par. 66 and steps 508, 510 and 514). By applying a loss function on the ML model training, dropped or missing frames can be replaced properly (note par. 66). Thus, in view of Harkness, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to include Reda into the combination of Harkness and Feng so that the quality of replacement frame(s) could be further improved.
Regarding claims 12 and 19, see rejection to claim 5.
Response to Arguments
Applicant's arguments filed 3/9/26 have been fully considered but they are not persuasive.
In view of the additional interpretations of Feng as set forth to claim 1, the new “behavior” limitation is still met by Feng. As a result, the rejection is maintained.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to MICHAEL LEE whose telephone number 571-272-7349. The examiner can normally be reached on Monday through Thursday from 9:00 am to 6:00 pm.
If attempts to reach the examiner by telephone are unsuccessful, the examiner's supervisor, John Miller, can be reached on 571-272-7353. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free).
/MICHAEL LEE/ Primary Examiner,
Art Unit 2422
Prosecution Insights