DETAILED ACTION
Notice of AIA Status
1. The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Arguments
2. Applicant’s remarks received on 03/05/2026 with respect to the amended independent claims have been acknowledged and are moot in view of a new ground of rejection necessitated by the corresponding amendment. Currently claims 1-18, 20, and 21 are rejected; claim 19 is rejected.
Response to Amendment
Claim Rejections - 35 USC § 103
3. 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 of this title, 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.
41066.. Claims 1-9, 11-18, 20, and 21 are rejected under 35 U.S.C. 103 as being unpatentable over Alcantarilla et al (Learning Visibility of Landmarks for Vision-Based Localization, 2010) and in further view of Parisotto et al (NEURAL MAP: STRUCTURED MEMORY FOR DEEP REINFORCEMENT LEARNING, 02/27/2017).
Regarding claim 1 (currently amended), Alcantarilla et al teaches: An apparatus for modelling an environment, comprising: at least one memory; and at least one processor coupled to the at least one memory and configured to: receive a location for a device and a map of feature points in the environment [abstract (“we model the visibility of every map feature w.r.t. the camera pose using a non-parametric distribution model…”)]; predict whether one or more feature points of the map of feature points are visible by at least one sensor from the location based on history information related to at least a plurality of feature points included in the map of feature points [page 1: I. Introduction: p04 (“predicting whether a feature will be visible or not in the image, based on a rough estimate of the camera pose…”); page 2: p01], wherein the history information indicates previous locations of the device and feature points of the map visible by the at least one sensor from the previous locations [page 3: p05 (“By positive examples, we mean the camera poses from which the particular map element was visible, and by negative examples the rest of the set…”)]; and determine a location of the device in the environment based on the one or more feature points predicted to be visible by the at least one sensor from the location [page 4: Results And Discussion, abstract].
Alcantarilla et al does disclose a learned embedding of the environment. In the same field of endeavor, Parisotto et al teaches: predict whether one or more feature points of the map of feature points are visible by at least one sensor from the location using a learned embedding of the environment, wherein the learned embedding is generated based on previous locations of the device in the environment and wherein the location for the device differs from the previous location of the device in the environment [abstract, page 3: 3 neural map (environment representation is location based feature vector.); page 4: 3.3, 3.4 (Encode current location into a new map feature vector to update the learned embedding based on current location.)].
Therefore, given Parisotto et al’s teaching on a learned embedded representation of an environment in the form of a neural map and updating the learned environment embedding based on previous location/observation, it would have been obvious for an ordinary skilled in the art before the effective filing date of the claimed invention to combine the teaching of the two to substitute Alcantarilla et al’s stored feature pose history with a learned neural map from Parisotto et al to store environment information to reduce memory and computation burden.
Regarding claim 2 (original), the rationale applied to the rejection of claim 1 has been incorporated herein. Parisotto et al further teaches: The apparatus of claim 1, wherein the at least one processor is further configured to: encode the determined location and an indication that the one or more feature points are visible by the at least one sensor from the location into an embedded representation of the environment [abstract, page 2: 3 neural map (3.3, 3.4), “Let st be the current state embedding, Mt be the current neural map, and (xt, yt) be the current position of the agent within the neural map…”The update operation creates the neural map for the next time step. The new neural map Mt+1 is equal to the old neural map Mt, except at the current agent position (xt, yt), where the current write candidate vector w (xt,yt) t+1 is store…” (The current position, (xt, yt) and what is visible at the location: the state embedded st and the write vector into the neural map Mt, which embedded representation of the environment.)]. Therefore, it would have been obvious for an ordinary skilled in the art before the effective filing date of the claimed invention to combine the teaching of the two to feed Alcantarilla’s pose conditioned visibility data into Parisotto et al’s neural map style embedded representation of the environment and keep updating the representation with new visibilities/poses overtime for efficiency and compactness.
Regarding claim 3 (original), the rationale applied to the rejection of claim 2 has been incorporated herein. Alcantarilla et al in view of Parisotto et al further teaches: The apparatus of claim 2, wherein the embedded representation is generated based on one or more previous predictions of whether one or more previous features of the map are visible by the at least one sensor at one or more locations in the environment [Alcantarilla: page 1: I. Introduction: p04; page 2: p01, page 3: p05 (Use history of camera poses and feature visibility labels to predict whether a feature is visible from a pose.)] and wherein, to encode the determined location and the indication that the one or more feature points are visible from the location into the embedded representation of the environment, the at least one processor is further configured to update the embedded representation of the environment [Parisotto: abstract, page 2: 3 neural map (3.3, 3.4)]. Therefore, given Alcantarilla et al’s disclosure on modeling and predicting visibilities for each feature based on a history of poses where the feature was visible or not and Parisotto et al’s teaching on neural map that writes a new feature vector at a given location with a state embedding of what is observed and continuous update of the map, it would have been obvious for an ordinary skilled in the art before the effective filing date of the claimed invention to combine the teaching of the two to feed visibility prediction into a state embedding at each pose and use a neural map to update stored embeddings at an agent’s location in the map to improve long-term memory efficiency.
Regarding claim 4 (original), the rationale applied to the rejection of claim 3 has been incorporated herein. Parisotto et al further teaches: The apparatus of claim 3, wherein the embedded representation is generated by an embedder comprising at least one of a temporal machine learning (ML) model, recurrent ML model, or transformer network ML model [abstract, page 5: 3.5.3 (GRU based local write is a recurrent ML model.)].
Regarding claim 5 (original), the rationale applied to the rejection of claim 4 has been incorporated herein. Parisotto et al further teaches: The apparatus of claim 4, wherein the location and the indication that the one or more feature points are visible by the at least one sensor from the location are received by the embedder along with a previously-generated embedded representation [page 4: 3.3 and 3.4 (The write network receives current location, observation encoded in a vector, and previously embedded representation at the location.)]. Again, the combined teaching of the two would have made it obvious to apply Alcantarilla’s feature visibility data to Parisotto’s embedder.
Regarding claim 6 (original), the rationale applied to the rejection of claim 5 has been incorporated herein. Alcantarilla et al in view of Parisotto et al further teaches: The apparatus of claim 5, wherein the previously-generated embedded representation includes an indication of whether the feature point was previously determined to be visible by the at least one sensor from a previous location [Alcantarilla: page 1: I. Introduction: p04; page 2: p01, page 3: p05 (Predict visibility of map features as a function of device pose based on history of poses associated with each feature visibility.); Parisotto: page 3: 3 (Each cell in neural map Mt stores a feature vector from a previous write at the same spatial position.)].
Regarding claim 7 (original), the rationale applied to the rejection of claim 6 has been incorporated herein. Alcantarilla et al further teaches: The apparatus of claim 6, wherein the location differs from the previous location in the previous embedded representation [page 2: p01, page 3: p03].
Regarding claim 8 (original), the rationale applied to the rejection of claim 1 has been incorporated herein. Alcantarilla et al further teaches: The apparatus of claim 1, wherein whether the one or more feature points are visible by the at least one sensor from the location is predicted as a calibrated uncertainty value [page 3: p02 (Model probabilistic visibility distribution for each feature)].
Regarding claim 9 (original), the rationale applied to the rejection of claim 8 has been incorporated herein. Alcantarilla et al further teaches: The apparatus of claim 8, wherein the calibrated uncertainty value indicates a probability that the one or more feature points are visible by the at least one sensor [page 3: p02 (Model probabilistic visibility distribution for each feature)].
Regarding claim 11 (original), the rationale applied to the rejection of claim 1 has been incorporated herein. Alcantarilla et al further teaches: The apparatus of claim 1, wherein, to determine the location of the device, the at least one processor is configured to exclude using one or more feature points predicted not to be visible by the at least one sensor from the location [page 2: III Visibility].
Regarding claim 12 (original), the rationale applied to the rejection of claim 1 has been incorporated herein. Alcantarilla et al further teaches: The apparatus of claim 1, wherein the at least one sensor includes an image sensor [abstract (camera)].
Claim 13 (currently amended) is a method version of claim 1 and has been analyzed and rejected with regard to claim 1.
Regarding claims 14-17 (original) and 20 (original), the rationale applied to the rejection of claim 13 has been incorporated herein. Claims 14-17 and 20 have been analyzed and rejected with regard to claims 2-5 and 8 respectively.
Regarding claim 18 (currently amended), the rationale applied to the rejection of claim 17 has been incorporated herein. Claim 18 has been analyzed and rejected with regard to claims 6 and 7.
Regarding claim 21 (New), the rationale applied to the rejection of claim 1 has been incorporated herein. Parisotto et al further teaches: The apparatus of claim 1, wherein the learned embedding of the environment is a fixed length embedding [page 3: 3 (neural map is a fixed size embedding with fixed C x H x W feature block.)].
51066.. Claim 10 is rejected under 35 U.S.C. 103 as being unpatentable over Alcantarilla et al (Learning Visibility of Landmarks for Vision-Based Localization, 2010) and Parisotto et al (NEURAL MAP: STRUCTURED MEMORY FOR DEEP REINFORCEMENT LEARNING, 2017); and in further view of Srinivasan et al (NeRV: Neural Reflectance and Visibility Fields for Relighting and View Synthesis, 2020).
Regarding claim 10 (original), the rationale applied to the rejection of claim 1 has been incorporated herein. Alcantarilla et al in view of Parisotto et al does not disclose a MLP model. In the same field of endeavor, Srinivasan et al teaches: The apparatus of claim 1, wherein whether the feature point is visible by the at least one sensor from the location is predicted by a multi-layer perceptron (MLP) ML model [abstract, page 2: p04]. Therefore, given Srinivasan et al’s prescription on a learned visibility MLP as a fast lookup table, it would have been obvious for an ordinary skilled in the art before the effective filing date of the claimed invention to combine the teaching of all to replace Alcantarilla’s non-parametric visibility predictor with a parametric visibility MLP taught by Srinivasan et al to improve computational efficiency.
Conclusion
6. There is a new ground of rejection necessitated by the corresponding amendment presented in this Office Action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP 706.07(a).
Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a). A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any extension fee pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Contact
7. Any inquiry concerning this communication or earlier communications from the examiner should be directed to FAN ZHANG whose telephone number is (571)270-3751. The examiner can normally be reached on Mon-Fri 9:00-5:00.
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, Benny Tieu can be reached on 571-272-7490. 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). 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.
/Fan Zhang/
Patent Examiner, Art Unit 2682
Prosecution Insights