Detail Office 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 .
Examiner cites particular columns and line numbers in the references as applied to the claims below for the convenience of the applicant. Although the specified citations are representative of the teachings in the art and are applied to the specific limitations within the individual claim, other passages and figures may apply as well. It is respectfully requested that, in preparing responses, the applicant fully consider the references in entirety as potentially teaching all or part of the claimed invention, as well as the context of the passage as taught by the prior art or disclosed by the examiner.
Status: Please all the replies and correspondence should be addressed to Examiner’s art unit 2629. Receipt is acknowledged of papers submitted on 03-06-2023 under new application; which have been placed of record in the file. Claims 1-20 are pending.
Information Disclosure Statement
The information disclosure statement (IDS) submitted on 06-05-2023, 08-23-2024, 11-25-2024, is in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statement is being considered by the examiner.
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.
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.
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 is/are rejected under 35 U.S.C. 103 as being unpatentable over Reda Fitsum A et al. (US-20190297326-A1) hereinafter referenced as Reda et al. in view of Vembar Deepak et al. (US-20220392116-A1) hereinafter referenced as Vembar et al.
Regarding Claim 1, Reda et al. disclosing a data processing system (Fig. 2, paras. 39-46,53) comprising: a memory device (para. 53) configured to store instructions (para. 53); a parallel processor including circuitry (para. 46) configured to perform matrix operations (para. 107 disclosing Training complex neural networks requires massive amounts of parallel computing performance and para. 108 disclosing Neural networks rely heavily on matrix math operations), the parallel processor configured to: load extrapolation weights associated with a machine learning model that is configurable to estimate an optical flow via one of extrapolation and interpolation (implicit because Fig. 2, Neural Network 220, comprises an architecture of a neural network that has been trained, which resulted in a set of weights that needs to be loaded during the inference phase in order for the network to perform its task, paras abstract, 42-44, 53 please notice Reda et al. does not explicitly recite extrapolation and interpolation); perform operations associated with the machine learning model, the operations to estimate a predicted optical flow based on the extrapolation weights, a plurality of rendered frames, and an optical flow between the plurality of rendered frames (abstract, par. 42, Fig. 2, Neural Network 220 outputting Parameters 222 comprising per pixel predicted displacement vectors, i.e. the claimed predicted optical flow, based on optical flows 212 and a sequence of frames 202; the prediction is seen as an extrapolation because it pertains to a future frame; NB the extrapolation weights are implicit because Neural Network 220 is trained paras 42-44, 53); and extrapolate a predicted frame based on a rendered frame of the plurality of rendered frames and the predicted optical flow (par.44, Fig. 2, SOC Module 230 predicting Frame 232, based on Frame 224 and Parameters 222, wherein the parameters include the displacement vectors, which constitute a predicted optical flow; see abstract, paras 42-44, 46, 53).
However, prior art of Reda et al. fails to recite extrapolation and interpolation.
However, prior art of Vembar et al. discloses a data processing system (para. 0227 and fig. 19, where the computing device 1900 corresponds to said data processing system) comprising: a memory device configured to store instructions (paras. 0231 and 0235); a parallel processor including circuitry configured to perform matrix operations (paras. 0231, 0234 and 0235, where the graphics processing unit corresponds to said parallel processor, perform matrix operations), the parallel processor configured to load extrapolation or interpolation weights associated with a machine learning model that is configurable to estimate an optical flow via one of extrapolation and interpolation (see paras. 116 discloses a pixel shader or fragment shader calculates the values of the various vertex attributes that are to be interpolated across the rasterized object, 238-239 (extrapolation), as well as fig. 20, disclosing a predicted (or inferred) optical flow generated by optical flow NN (e.g., t, t+1) is used as input to an extrapolation NN along, with the frames and the optical flow (t, t−1). An extrapolation NN considers past rendered frames and associated motion vectors and warps in order to predict a subsequent (or next) frame (e.g., for fixed frame intervals, such 30 fps or 60 fps, where the weights of the optical flow NN correspond to extrapolation weights); perform operations associated with the machine learning model, the operations to estimate a predicted optical flow based on extrapolation weights (paras. 238-239, as well as fig. 20, disclosing a predicted (or inferred) optical flow generated by optical flow NN (e.g., t, t+1) is used as input to an extrapolation NN along, with the frames and the optical flow (t, t−1). An extrapolation NN considers past rendered frames and associated motion vectors and warps in order to predict a subsequent (or next) frame (e.g., for fixed frame
intervals, such 30 fps or 60 fps), where the weights of the optical flow NN correspond to extrapolation weights); and generate a predicted frame based on a rendered frame of the plurality of rendered frames (abstract, para. 30) and the predicted optical flow via one of extrapolation and interpolation (paras. 30, 0238, 0239 discloses a predicted (or inferred) optical flow generated by optical flow NN (e.g., t, t+1) is used as input to an extrapolation NN along, with the frames and the optical flow (t, t−1). An extrapolation NN considers past rendered frames and associated motion vectors and warps in order to predict a subsequent (or next) frame (e.g., for fixed frame intervals, such 30 fps or 60 fps), where the weights of the optical flow NN correspond to extrapolation weights and paras. 240-243, as well as figs. 20 and 22, disclose generating a predicted optical flow (t) based on a received optical flow (t−1). Warp engine 2220 receives the predicted optical flow, as well as previous (or past) frames (t−1 and t−2) and performs an image reprojection of the frames to generate a predicted frame and a confidence map 2230. Predicted frame may be generated (e.g., via frame extrapolation) at a neural accelerator (e.g., accelerator
1911 in FIG. 19), while the confidence map threshold generation may occur at GPU 1910).
Reda et al. teaches a neural network that processes a sequence of video frames and optical flows to generate parameters for sampling a previous video frame. The sampling of the previous video frame, based on the parameters, is performed to generate pixel values for a predicted video frame, which is the next video frame following the sequence of video frames.
Reda et al. teaches A spatially-displaced convolution module receives the set of parameters for a pixel of the predicted video frame and samples the previous video frame by performing a convolution operation on a corresponding patch of pixels displaced from a corresponding pixel in the previous video frame. The patch of pixels is identified using the displacement vector from the set of parameters for the pixel and is offset from the corresponding pixel in the previous video frame by a magnitude of the displacement vector.
Vembar et al. teaches and recites extrapolation and interpolation as noted above.
Hence the prior art includes each element claimed, although not necessarily in a single prior art reference, with the only difference between the claimed invention and the prior art being the lack of actual combination of the elements in a single prior art reference.
In combination, Reda et al. performs the same function as it does separately of managing process operating, a neural network that processes a sequence of video frames and optical flows to generate parameters for sampling a previous video frame.
Vembar et al. performs the same neural frame extrapolation to generate a predicted next frame and a per pixel confidence map and the predicted optical flow, render a first set of the plurality of pixels in the predicted frame based on the confidence map and adding the rendered pixels to the predicted frame to generate a final frame.
Therefore one of ordinary skill in the art could have combined the elements as claimed by known methods, and that in combination, each element merely performs the same function as it does separately.
The results of the combination would have been predictable, and it would have been obvious to one of ordinary skill in the art to modify the invention of Reda et al. to include teaching and recitation of extrapolation and interpolation, as disclosed by Vembar et al. in order to render a first set of the plurality of pixels in the predicted frame based on the confidence map and adding the rendered pixels to the predicted frame to generate a final frame as Vembar et al. discusses at para. 30.
Regarding Claim 2, Vembar et al. the parallel processor (paras. 0231, 0234 and 0235, where the graphics processing unit corresponds to said parallel processor) configured to warp a rendered frame of the plurality of rendered frames based on the predicted optical flow to extrapolate the predicted frame (para. 241, fig.22)
Claim(s) 3-4 and 11 is/are rejected under 35 U.S.C. 103 as being unpatentable over Reda Fitsum A et al. (US-20190297326-A1) hereinafter referenced as Reda et al. in view of Vembar Deepak et al. (US-20220392116-A1) hereinafter referenced as Vembar et al. as applied to claims 1-2 above and further in view of Jiang Huaizu et al. (US-20190138889-A1) hereinafter referenced as Jiang et al.
Regarding Claim 3-4, Vembar et al. discloses the parallel processor (paras. 0231, 0234 and 0235, where the graphics processing unit corresponds to said parallel processor) configured to: load interpolation weights associated with a machine learning model that is configurable to estimate an optical flow via one of extrapolation and interpolation (please see para. 116 discloses a pixel shader or fragment shader calculates the values of the various vertex attributes that are to be interpolated across the rasterized object, paras. 238-239, as well as fig. 20, disclosing a predicted (or inferred) optical flow generated by optical flow NN (e.g., t, t+1) is used as input to an extrapolation NN along, with the frames and the optical flow (t, t−1). An extrapolation NN considers past rendered frames and associated motion vectors and warps in order to predict a subsequent (or next) frame (e.g., for fixed frame intervals, such 30 fps or 60 fps), where the weights of the optical flow NN correspond to extrapolation weights); perform the operations associated with the machine learning model, the operations to estimate an interpolated optical flow based on the interpolation weights, the plurality of rendered frames, an optical flow between the plurality of rendered frames; and interpolate an intermediate frame between frames of the plurality of predicted frames based on a rendered frame of the plurality of rendered frames and the interpolated optical flow (para. 240-243, fig. 22, disclosing generates a predicted optical flow (t) based on a received optical flow (t−1). Warp engine 2220 receives the predicted optical flow, as well as previous (or past) frames (t−1 and t−2) and performs an image reprojection of the frames to generate a predicted frame and a confidence map 2230. Predicted frame may be generated (e.g., via frame extrapolation) at a neural accelerator (e.g., accelerator 1911 in FIG. 19), while the confidence map threshold generation may occur at GPU 1910. Instead of two frames, the above processing is used to provide a different magnitude of previous frames (e.g., (t−1−t−3), (t−1−t−4), etc.) to generate a predicted frame (interpolation with intermediate frame) Vembar et al. does not recite interpolate an intermediate frame).
Please also see Reda et al. disclosure, para 45-46, 53, 58 disclosing the parallel processor configured to warp a rendered frame of the plurality of rendered frames based on the interpolated optical flow to interpolate the intermediate frame.
However, Reda et al. in view of Vembar et al. fails to recite interpolate an intermediate frame.
However, prior art of Jiang et al. recites Video interpolation is used to predict one or more intermediate frames at timesteps defined between two consecutive frames (abstract, paras. 41, 55).
Reda et al. teaches a neural network that processes a sequence of video frames and optical flows to generate parameters for sampling a previous video frame. The sampling of the previous video frame, based on the parameters, is performed to generate pixel values for a predicted video frame, which is the next video frame following the sequence of video frames.
Reda et al. teaches A spatially-displaced convolution module receives the set of parameters for a pixel of the predicted video frame and samples the previous video frame by performing a convolution operation on a corresponding patch of pixels displaced from a corresponding pixel in the previous video frame. The patch of pixels is identified using the displacement vector from the set of parameters for the pixel and is offset from the corresponding pixel in the previous video frame by a magnitude of the displacement vector.
Jiang et al. teaches and recites Video interpolation is used to predict one or more intermediate frames at timesteps defined between two consecutive frames.
Reda et al. in view of Vemba et al. does not recite Video interpolation is used to predict one or more intermediate frames at timesteps defined between two consecutive frames.
Hence the prior art includes each element claimed, although not necessarily in a single prior art reference, with the only difference between the claimed invention and the prior art being the lack of actual combination of the elements in a single prior art reference.
In combination, Reda et al. and Vemba et al. performs the same function as it does separately of managing process operating, a neural network that processes a sequence of video frames and optical flows to generate parameters for sampling a previous video frame.
Jiang et al. performs the same Video interpolation is used to predict one or more intermediate frames at timesteps defined between two consecutive frames. A first neural network model approximates optical flow data defining motion between the two consecutive frames. A second neural network model refines the optical flow data and predicts visibility maps for each timestep.
Therefore one of ordinary skill in the art could have combined the elements as claimed by known methods, and that in combination, each element merely performs the same function as it does separately.
The results of the combination would have been predictable, and it would have been obvious to one of ordinary skill in the art to modify the invention of Reda et al. and Vemba et al. to include teaching and recitation of using Video interpolation to predict one or more intermediate frames at timesteps defined between two consecutive frames, as disclosed by Jiang et al. in the data processing system of Reda et al. and Vembar et al. in order to advantageously reduce Artifacts caused by motion boundaries and occlusions in the predicted intermediate frames, as Jiang et al. discusses at abstract, paras. 6, 50.
Regarding Claim 11, Reda et al. disclosing a data processing system (Fig. 2, paras. 39-46,53) comprising: a memory device (para. 53) configured to store instructions (para. 53); a parallel processor including circuitry (para. 46) configured to perform matrix operations (para. 107 disclosing Training complex neural networks requires massive amounts of parallel computing performance and para. 108 disclosing Neural networks rely heavily on matrix math operations), the parallel processor configured to: load extrapolation weights associated with a machine learning model that is configurable to estimate an optical flow via one of extrapolation and interpolation (implicit because Fig. 2, Neural Network 220, comprises an architecture of a neural network that has been trained, which resulted in a set of weights that needs to be loaded during the inference phase in order for the network to perform its task, paras abstract, 42-44, 53 please notice Reda et al. does not to recite extrapolation and interpolation); perform operations associated with the machine learning model, the operations to estimate a predicted optical flow based on the extrapolation weights, a plurality of rendered frames, and an optical flow between the plurality of rendered frames (abstract, par. 42, Fig. 2, Neural Network 220 outputting Parameters 222 comprising per pixel predicted displacement vectors, i.e. the claimed predicted optical flow, based on optical flows 212 and a sequence of frames 202; the prediction is seen as an extrapolation because it pertains to a future frame; NB the extrapolation weights are implicit because Neural Network 220 is trained paras 42-44, 53); and extrapolate a predicted frame based on a rendered frame of the plurality of rendered frames and the predicted optical flow (par.44, Fig. 2, SOC Module 230 predicting Frame 232, based on Frame 224 and Parameters 222, wherein the parameters include the displacement vectors, which constitute a predicted optical flow paras abstract, 42-44, 46, 53).
Vembar et al. discloses a data processing system (para. 0227 and fig. 19, where the computing device 1900 corresponds to said data processing system) comprising: a memory device configured to store instructions (paras. 0231 and 0235); a parallel processor including circuitry configured to perform matrix operations (paras. 0231, 0234 and 0235, where the graphics processing unit corresponds to said parallel processor, perform matrix operations), the parallel processor configured to load extrapolation or interpolation weights associated with a machine learning model that is configurable to estimate an optical flow via one of extrapolation and interpolation (see paras. 116 discloses a pixel shader or fragment shader calculates the values of the various vertex attributes that are to be interpolated across the rasterized object, 238-239 (extrapolation), as well as fig. 20, disclosing a predicted (or inferred) optical flow generated by optical flow NN (e.g., t, t+1) is used as input to an extrapolation NN along, with the frames and the optical flow (t, t−1). An extrapolation NN considers past rendered frames and associated motion vectors and warps in order to predict a subsequent (or next) frame (e.g., for fixed frame intervals, such 30 fps or 60 fps, where the weights of the optical flow NN correspond to extrapolation weights); perform operations associated with the machine learning model, the operations to estimate a predicted optical flow based on extrapolation weights (paras. 238-239, as well as fig. 20, disclosing a predicted (or inferred) optical flow generated by optical flow NN (e.g., t, t+1) is used as input to an extrapolation NN along, with the frames and the optical flow (t, t−1). An extrapolation NN considers past rendered frames and associated motion vectors and warps in order to predict a subsequent (or next) frame (e.g., for fixed frame
intervals, such 30 fps or 60 fps), where the weights of the optical flow NN correspond to extrapolation weights); and generate a predicted frame based on a rendered frame of the plurality of rendered frames (abstract, para. 30) and the predicted optical flow via one of extrapolation and interpolation (paras. 30, 0238, 0239 discloses a predicted (or inferred) optical flow generated by optical flow NN (e.g., t, t+1) is used as input to an extrapolation NN along, with the frames and the optical flow (t, t−1). An extrapolation NN considers past rendered frames and associated motion vectors and warps in order to predict a subsequent (or next) frame (e.g., for fixed frame
intervals, such 30 fps or 60 fps), where the weights of the optical flow NN correspond to extrapolation weights and paras. 240-243, as well as figs. 20 and 22, discloses disclosing generates a predicted optical flow (t) based on a received optical flow (t−1). Warp engine 2220 receives the predicted optical flow, as well as previous (or past) frames (t−1 and t−2) and performs an image reprojection of the frames to generate a predicted frame and a confidence map 2230. Predicted frame may be generated (e.g., via frame extrapolation) at a neural accelerator (e.g., accelerator
1911 in FIG. 19), while the confidence map threshold generation may occur at GPU 1910, further discloses perform operations associated with the machine learning model, the operations to estimate a predicted optical flow based on loaded weights, a plurality of rendered frames at a first resolution, and an optical flow between the plurality of rendered frames at the first resolution, the optical flow upsampled to the first resolution from input optical flow at a second resolution that is less than the first resolution).
Jiang et al. recites and discloses A method comprising: performing end-to-end training of a spatiotemporal neural frame prediction network (paras. 52, 61 , 72, disclosing end-to-end training of a spatiotemporal neural frame prediction network), an interpolation dataset to generate interpolation weights for frame generation and super resolution (para. 50, 65, disclosing by applying the visibility maps to the warped images before fusion, the contribution of occluded pixels is excluded from the interpolated intermediate frame, thereby avoiding or reducing artifacts. Since none of parameters (e.g., weights) of the frame interpolation system learned during training are time-dependent, the frame interpolation system is able to produce as many intermediate frames as needed, thereby Artifacts caused by motion boundaries and occlusions are reduced in the predicted intermediate frames.(producing super or high resolution); and Video interpolation is used to predict one or more intermediate frames at timesteps defined between two consecutive frames (abstract, paras. 41, 55).
Allowable Subject Matter
Claims 18-20 are allowed.
The following is an examiner’s statement of reasons for allowance:
after further consideration as well as extensive search, all of the prior art cited on 892’s 1449’s, searched in NPL and searched in PGPUB, fails to recite or disclose all the limitations of independent claims with uniquely distinct features represented by underlined bold claim limitations recited below;
concatenate the preprocessed color data and the optical flow data into a multi-channel block of input data; estimate optical flow between a frame of the plurality of frames and a new frame, based on the multi-channel block of input data via first operations associated with a machine learning model and interpolation weights or extrapolation weights for the machine learning model; warp color data for a frame of the plurality of frames into the new frame based on estimated optical flow between the frame of the plurality of frames and the new frame; and upsample the frame data for the plurality of frames and the new frame to a target resolution via second operations associated with the machine learning model.
Any comments considered necessary by applicant must be submitted no later than the payment of the issue fee and, to avoid processing delays, should preferably accompany the issue fee. Such submissions should be clearly labeled “Comments on Statement of Reasons for Allowance.”
Claims 5-10, 12-17 are 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.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Applicant is requested to review the prior art cited on USTO 892’s.
The prior art of Liu Feng et al. (US 20200012940 A1)disclosure; paras. 37-181, discloses, Arrangement 100 may comprise a convolutional neural network (CNN or ConvNet) 105, which may be a deep neural network that transforms a three dimensional (3D) input volume into a 3D output volume. The ConvNet 105 may be an architecture for image classification tasks. A computer system/device implementing the ConvNet 105 may apply a series of filters to raw pixel data of an image to extract and learn higher-level features, which may then be used for classification. The ConvNet 105 may comprise an input layer, an output layer, and one or more hidden layers. The one or more hidden layers may include one or more convolutional layers, transposed convolution layers, one or more pooling layers, dense layers, upsampling layers, activation layers, and/or regularization and/or normalization layers. The input layer may comprise the raw pixel data of an image and the output layer may comprise a single vector of class scores of the image along the depth dimension. In this example, the input layer may hold pixels data of the input image frames 110. The three dimensions of the 3D input volume of the input layer may include height, width, and depth. The depth dimension of the input layer may be a color of one or more image frames 110 (e.g., Red, Green, Blue (RGB) channels for each frame). The convolutional layers may apply a specified number of convolutional operations to input data. This may include performing a set of operations to produce a single value in an output feature map. Each convolutional layer may include a set of learnable filters (also referred to as "kernels" or "weights"), which may be a small matrix that extends through the entire input volume, and may be used for blurring, sharpening, embossing, edge detection, and the like. Each convolutional layer may apply a convolution operation to (or "convolve") a corresponding kernel with the input volume, and may pass a result to a next layer. The data processed by individual convolutional layers may correspond to a receptive field of an image or image frame (e.g., receptive patches R1 and R2 in FIG. 1), where receptive fields of different convolutional layers may overlap with one another. Additionally, the convolutional layers may apply an activation function (e.g., ReLUs) of an activation layer to increase nonlinearity. [0041] The pooling layer(s) may be used to combine the outputs of various groupings or clusters of convolutional layers. The pooling layers may implement a form of downsampling to reduce the size of the feature map. In some implementations, the pooling layer(s) may implement max pooling, which extracts subregions of the feature map, keeps their maximum value, and discards the other values. [0042] For a pixel (x, y) in an interpolated frame, the deep neural network (e.g., ConvNet 105) may take two receptive field (input) patches R1 and R2 centered at the output pixel as an input, and may estimate a convolution kernel K. The convolution kernel K may be used to convolve with the input patches P1 and P2 to synthesize the output pixel 115. In FIG. 1, for each output pixel (x, y), a convolution kernel K may be estimated and used to convolve with patches P1 and P2 centered at (x, y) in the input frames 110 to produce color Î(x, y) of the output frame 115.
The prior art of Sun Deqing et al. (US 20200084427 A1) disclosure, paras. 21-150, discloses, scene flow represents the three-dimensional (3D) structure and movement of objects in a video sequence in three dimensions from frame-to-frame and is used to track objects and estimate speeds for autonomous driving applications. Scene flow is recovered, by a neural network system, from a video sequence captured from at least two viewpoints (e.g., cameras), such as a left-eye and right-eye of a viewer as simulated by a pair of offset cameras. An encoder portion of the system extracts features from frames of the video sequence. The features are input to a first decoder to predict optical flow and a second decoder to predict disparity. The optical flow represents pixel movement in (x,y) and the disparity represents pixel movement in z (depth). In an embodiment, when combined, the optical flow and disparity represent the scene flow. In an embodiment, the scene flow is 3D pixel movement (e.g., 3D motion vectors) that is inferred using the optical flow and the disparity. A method, computer readable medium, and system are disclosed for estimating scene flow. A first set of features for a first stereo image pair in a sequence of image pairs and a second set of features for a second stereo image pair in the sequence of image pairs are received. A first task-specific decoder neural network computes disparity estimates for the sequence of image pairs based on the first set of features. A second task-specific encoder neural network computes optical flow estimates for the sequence of image pairs based on the first set of features and the second set of features. The disparity estimates and the optical flow estimates are combined to produce a scene flow estimate for the sequence of image pairs.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to PRABODH M DHARIA whose telephone number is (571)272-7668. The examiner can normally be reached Monday -Friday 9:00 AM to 5:30 PM.
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, Benjamin Lee can be reached on 571-272-2963. 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.
Any response to this action should be mailed to:
Commissioner of Patents and Trademarks
P.O. Box 1450
Alexandria VA 22313-1450
/Prabodh M Dharia/
Primary Examiner
Art Unit 2629
04-01-2026
Prosecution Insights