DETAILED ACTION
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and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114.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 May 26, 2026, has been entered.
Information Disclosure Statement
The IDS(s) filed May 26, 2026, and June 18, 2026, has/have been considered and placed in the application file.
Response to Arguments
This office action is in response to the amendment filed May 26, 2026. Claims 1-21 are pending in this application and have been considered below. Claims 1, 6, 11, and 18 are amended. Claim 21 is new. No claims are canceled.
Applicant’s arguments with respect to claims 1, 6, 11, 18, and 21 have been considered but are moot in view of new ground(s) of rejection because of the amendments.
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.
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 nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claims 1-7 and 10-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Friedman et al. (US 2022/0292705 A1 – hereinafter “Friedman”) in view of Podkamien et al. (US 2023/0168364 A1 – hereinafter “Podkamien”) in view of Gronau (US 20240265554 A1 – hereinafter “Gronau”) in view of in view of Luo et al. (US 2005/0185845 A1 – hereinafter “Luo”).
Claim 1, 11, and 18.
Friedman discloses a processor comprising: one or more processing units to (FIG. 2A "Processor 152"; ¶106 discloses "The control board 150 may comprise one or more of processors 152, memory 154 and communication circuitry 156.")
generate, based at least on applying a representation of sensor data to one or more machine learning models, different types of predictions of age or presence of one or more detected occupants, the one or more machine learning models configured to (Fig. 8, 820 discloses “Apply a pose detection algorithm on each of the obtained sequence of 2D images to yield a 2D skeleton representation of said one or more occupants”; ¶8 discloses “combine one or more 3D image of said sequence of 3D images with said one or more skeleton representations … to yield at least one skeleton model for each one or more occupants … analyze the one or more skeleton models to extract one or more features of each of the one or more occupants”; ¶¶ 23, 160 discloses “an OpenPose algorithm on the images and/or other DNN algorithms such as DensePose configured to extract body pose.”);
generate a representation of whether a child is present based at least on a combined assessment of the different types of predictions of the age or the presence of the one or more detected occupants (¶134 discloses “the integration process includes computationally combining the formed skeleton (2D skeleton) and the depth maps representation to yield the skeleton model which includes data for each key-point in the skeleton model in an (x,y,z) coordinate system.; ¶165 discloses “a mass classification (282) for each identified object in the scene, such as objects 254 and 255 may be determined in accordance with a number of pre-determined mass categories, e.g. child; teenager; adult.”) (Friedman ¶87 discloses “a face detector sensor and/or face detection and/or face recognition software module for analyzing the captured 2D and/or 3D images.”; Classifying a face into an age range is a well-understood application of facial analysis. Podkamien ¶111 discloses “a neural network may be trained to identify and categorize passengers, mapping them to classifications such as age category.”); and
execute one or more operations based at least on the representation of whether the child is present (¶¶15-16 discloses “the output signals are associated with an operation of one or more of the vehicle's units. In an embodiment, the vehicle's units are selected from the group consisting of: airbag; Electronic Stabilization Control (ESC) Unit; safety belt.”).
Friedman discloses all of the subject matter as described above except for specifically teaching “different types” of predictions, “compute a confidence level for individual predictions,” and “wherein the combined assessment is based at least on selective inclusion of the different types of predictions of age or presence based at least on the confidence level for the individual predictions.” However, Podkamien in the same field of endeavor teaches different types of predictions (Podkamien discloses radar-based system for monitoring a vehicle cabin: Abstract discloses “Vehicle cabin monitoring using a radar unit centrally positioned within the cabin to obtain image data of the vehicle cabin and a processor to generate detect occupancy of seats within the vehicle cabin, categorize occupants, detect posture, determine seatbelt status and monitor life signs of the occupants.” ¶84 discloses “Embodiments of the invention use a single sensor to track both occupancy and movements within the cabin of a vehicle.” Determining and categorizing movements, including heartbeat and breathing: ¶149 discloses “The system is also able to determine and categorize movements of each occupant, including heart beat and breathing, for example.” ¶22 discloses “In addition, the data detected which includes both macro and minor movements over time, can monitor posture, hand gestures, breathing and heart rate.” Monitoring posture and classifying occupants: ¶21 discloses “The radar sensor array is configured to monitor the cabin and the objects and passengers within the cabin, and can differentiate between different kinds of passengers, such as adults and children, babies, pets and inanimate objects.” ¶148 discloses “the signal may be analysed or compared with a signal for various targets 610, such as adults, children, pets, babies, and inanimate objects”). Neither Friedman nor Podkamien details the mathematical mechanism for combining different types of age predictions based on confidence levels. However, Gronau teaches and in-cabin vehicle monitoring system that predicts occupant characteristics, including age (Grouau ¶125), that fuses more than one detection method (Grouau ¶ 225 “the present invention methods and systems which include fusing multiple detection methods such as DNN and face detection methods”), and that weighs and discards inputs according to their uncertainty (Grouau ¶ 219 “inputs of the predicted state and the current state (e.g. weighted by their uncertainties) into an updated state” and ¶ 220 discloses “the updating algorithm 613 is configured to discard input data which includes high uncertainty as eventually this data should not affect and change the current state.”). Gronau does not include or exclude those prediction types on the basis of age or presence predictions. However, Luo in the same field of endeavor teaches camera-based classification of a vehicle occupant for restraint control into classes including a child class and a rearward facing infant seat class (Luo ¶49, ¶54 discloses “the output classes can represent potential occupants of a passenger seat. such as a child class, an adult class, a rearward facing infant seat class. an empty seat class, and similar useful classes.”), in which the classifiers may operate on prediction specific inputs (Luo ¶ 53 discloses “The plurality of classifiers 420, 422, 424 can receive a common set of features from the feature extraction assembly 410, or the feature extraction assembly can provide feature sets specific to each classifier.”), each classifier is a machine learning model that computes its own confidence (Luo ¶ 56 discloses “the classifier 54 can utilize one of a Support Vector Machine ("SVM") algorithm or an artificial neural network ("ANN") learning algorithm to classify the image into one of a plurality of output classes.”; ¶ 58 discloses “A final layer of nodes provide the confidence values for the output classes of the ANN, with each node having an associated value representing a confidence for one of the associated output classes of the classifier.”), and an arbitrator that evaluates individual classifier outputs to determine if they are included in or excluded from the combination on the basis of those confidence values (¶¶ 60-62).
Therefore, it would have been obvious to one of ordinary skill in the art to combine Friedman and Podkamien before the effective filing date of the claimed invention. The motivation for this combination of references would have been to a predictable improvement to enhance Friedman’s pose-based classification system by integrating the different, complementary prediction types as taught by Podkamien (e.g., heartbeat and breathing detection). This combination would yield a more robust and reliable system, capable of confirming the presence of a living occupant and improving classification accuracy, thus rendering the claim obvious.
It would also have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the multi-modal occupant monitoring system of Friedman, Podkamien and Gronau so that each age or presence prediction carries its own confidence level and the combined assessment includes or excludes individual predictions on the basis of that confidence level, as taught by Luo, because Luo states the purpose of its arbitration as follows (¶ 48 discloses “the process illustrated in FIGS. 6A-6D utilizes the best available data, in the form of the classification with the largest confidence value. at each stage in the process. increasing the overall accuracy of the classification system.”), whereas Friedman, Podkamien and Gronau carry every age or presence prediction into the combination regardless of how reliable each one is. This is the use of a known technique to improve similar devices in the same way. See MPEP § 2143(I)(C). Luo computes its confidence values inside the same SVM and neural network classifiers that produce the predictions. Friedman, Podkamien and Gronau already run neural network models that produce age and presence predictions. No new model architecture is needed, only reading the confidence each model already produces and gating the combination on it. Luo is analogous art since it deals with camera based classification of the occupant of a vehicle seat into age-related classes for air bag control (¶49).
Claims 2, 12, and 19.
The combination of Friedman, Podkamien, Gronau, and Luo discloses the processor of claim 1, the one or more processing units further to generate a first prediction of the different types of predictions based at least on using a first machine learning model of the one or more machine learning models to detect a pose (Friedman Fig. 8, 820 discloses “Apply a pose detection algorithm on each of the obtained sequence of 2D images to yield a 2D skeleton representation of said one or more occupants”; ¶8 discloses “combine one or more 3D image of said sequence of 3D images with said one or more skeleton representations … to yield at least one skeleton model for each one or more occupants … analyze the one or more skeleton models to extract one or more features of each of the one or more occupants”; ¶¶ 23, 160 discloses “In an embodiment, the pose detection algorithm is an OpenPose algorithm.”), estimating limb length based at least on the detected pose, and using a second machine learning model of the one or more machine learning models to regress age based at least on the limb length.
Claims 3 and 13.
The combination of Friedman, Podkamien, Gronau, and Luo discloses the processor of claim 1, the one or more processing units further to generate a first prediction of the different types of predictions based at least on using a first machine learning model of the one or more machine learning models to detect a pose (Friedman ¶160 discloses “an OpenPose algorithm on the images and/or other DNN algorithms such as DensePose configured to extract body pose.”), estimating limb length based at least on the detected pose, and using a mapping that associates the limb length with a corresponding age range (Friedman Fig’s 4C-4G determines occupant size by first detecting features (key-points ¶133) in a 2D image, generating a 3D pose estimate (¶132), then correlating it with depth data (e.g., from a point cloud ¶134) to scale the pose estimate to real world dimensions. Limb length can be determined from the scaled 3D pose.).
Claims 4 and 14.
The combination of Friedman, Podkamien, Gronau, and Luo discloses the processor of claim 1, the one or more processing units further to generate a first prediction of the different types of predictions based at least on classifying a detected face of an occupant of the one or more detected occupants into one of a plurality of age ranges (Friedman ¶87 discloses “a face detector sensor and/or face detection and/or face recognition software module for analyzing the captured 2D and/or 3D images.”; Podkamien ¶ 111 discloses “a neural network may be trained to identify and categorize passengers, mapping them to classifications such as age category and in-position/out-of position states.”).
Claims 5 and 15.
The combination of Friedman, Podkamien, Gronau, and Luo discloses the processor of claim 1, wherein the different types of predictions of the age of an occupant of the one or more detected occupants comprise a first estimated age predicted based at least on a detected face of the occupant (Friedman ¶87 discloses “a face detector sensor and/or face detection and/or face recognition software module for analyzing the captured 2D and/or 3D images.”; Podkamien ¶ 111 discloses “a neural network may be trained to identify and categorize passengers, mapping them to classifications such as age category and in-position/out-of position states.”), a second estimated age predicted based at least on an estimated size of the occupant (Friedman ¶165 discloses “a mass classification (282) for each identified object in the scene, such as objects 254 and 255 may be determined in accordance with a number of pre-determined mass categories, e.g. child; teenager; adult.”), and a third estimated age predicted based at least on a RADAR classification of the occupant (Podkamien ¶111 discloses “The radar receiver array 310 is coupled to a memory 326 which stores the signals received by receiver 310 … a neural network may be trained to identify and categorize passengers, mapping them to classifications such as age category and in-position/out-of position states.”). The motivation to combine is the same as for the independent claims.
Claims 6 and 16.
The combination of Friedman, Podkamien, Gronau, and Luo discloses the processor of claim 1, wherein the different types of predictions of the presence of an occupant of the one or more detected occupants comprise a classification of the occupant as a child predicted based at least on detecting a child seat in a first slot and classifying the slot as being occupied based at least on RADAR data (Podkamien ¶111 discloses “The radar receiver array 310 is coupled to a memory 326 which stores the signals received by receiver 310 … a neural network may be trained to identify and categorize passengers, mapping them to classifications such as age category and in-position/out-of position states.”). The motivation to combine is the same as for the independent claims.
Claim 7.
The combination of Friedman, Podkamien, Gronau, and Luo discloses the processor of claim 1, the one or more processing units further to generate the representation of whether the child is present based at least on applying a representation of the different types of predictions of the age or the presence of the one or more detected occupants to one or more subsequent machine learning models to generate one or more predicted values representative of the age of the one or more detected occupants (Friedman ¶142 discloses “The mass prediction module 224 obtains the valid images of the objects from the skeleton model data … the pre-trained regression module to provide the most accurate mass prediction (e.g. estimation prediction) for each captured object (e.g. persons).).
Claims 10, 17, and 20.
The combination of Friedman, Podkamien, Gronau, and Luo discloses the processor of claim 1, wherein the processor is comprised in at least one of: a control system for an autonomous or semi-autonomous machine; a perception system for an autonomous or semi-autonomous machine; a system for performing simulation operations; a system for performing digital twin operations; a system for performing light transport simulation; a system for performing collaborative content creation for 3D assets; a system for performing deep learning operations; a system for performing real-time streaming; a system implemented using an edge device; a system implemented using a robot; a system for performing conversational AI operations; a system for generating synthetic data; a system incorporating one or more virtual machines (VMs); a system implemented at least partially in a data center; or a system implemented at least partially using cloud computing resources (Friedman ¶224 discloses “In further embodiments, software modules are hosted on cloud computing platforms.”).
Claim(s) 8 and 21 is/are rejected under 35 U.S.C. 103 as being unpatentable over Friedman, Podkamien, Gronau, and Luo as applied to claim 1 above, and further in view of Weng et al. (US 11713600 B1 – hereinafter “Weng”).
Claim 8.
The combination of Friedman, Podkamien, Gronau, and Luo discloses the processor of claim 1, wherein the sensor data comprises one or more RGB images and one or more infrared images, the one or more processing units further to generate the different types of predictions of the age or the presence of the one or more detected occupants (Friedman Fig. 8, 820, ¶¶ 8, 23, 160; Podkamien Abstract, ¶¶ 21-22, 148-149);
(Friedman Fig. 8, 820, ¶¶ 8, 23, 160).
Friedman, Podkamien, Gronau, and Luo discloses all of the subject matter as described above except for specifically teaching “based at least on applying a combined representation of the one or more RGB images and the one or more infrared images.” However, Weng in the same field of endeavor teaches “based at least on applying a combined representation of the one or more RGB images and the one or more infrared images” (C8:L20-25 discloses “the sensor 140a may implement an RGB-InfraRed (RGB-IR) sensor.”).
Therefore, it would have been obvious to one of ordinary skill in the art to Friedman, Podkamien, Gronau, Luo, and Weng before the effective filing date of the claimed invention. The motivation for this combination of references would have been to incorporate both RGB and IR sensors to provide robust image analysis across all lighting conditions.
Claim 21. (NEW)
Friedman, Podkamien, Gronau, Luo, and Weng disclose the processor of claim 1, the one or more processing units further to apply a weight to the confidence level for the individual predictions based at least on weights corresponding to types of sensor data (Friedman and
Podkamien supply the types of sensor data, camera and RADAR) used to generate corresponding predictions of the different types of predictions (Weng C33:L18-25 discloses “The processors 106a-106n may determine the age of the occupant 452' based on an aggregation of various characteristics detected using the computer vision operations performed by the CNN module 150. Statistical weight may be adjusted for various characteristics. For example, a higher confidence level of an accurate detection of particular characteristic may have more statistical weight than a lower confidence level of accurate detection.”; C33:L29-33 discloses "The various factors and/or the statistical weights applied by the processors 106a-106n for each type of factor used for determining the age of the occupant 452' may be varied according to the design criteria of a particular implementation.”).
Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to combine Friedman, Podkamien, Gronau,
Luo and Weng, because Weng gives a higher confidence level more statistical weight than a lower confidence level, and camera, depth, and radar modalities differ in how accurately each one indicates age, so calibrating the combination to the known accuracy of each sensing modality rather than treating all modalities the same yields a more accurate age assesment.
Allowable Subject Matter
Claim 9 is objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims. Previously this claim was rejected over MOUSTAFA et al. (WO 2020205597 A1); however, upon further review, MOUSTAFA ¶447 generates image from text keywords, not synthetic faces, and does not supply and age range to the generator.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Ross Varndell whose telephone number is (571)270-1922. The examiner can normally be reached M-F, 9-5 EST.
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.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, O’Neal Mistry can be reached at (313)446-4912. 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 https://ppair-my.uspto.gov/pair/PrivatePair. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/Ross Varndell/Primary Examiner, Art Unit 2674
Prosecution Insights