DETAILED ACTIONS
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 .
Priority
Acknowledgment is made of applicant’s claim this application being in benefit of foreign priority from Korean Patent Application No. KR10-2022-0066401 filed on May 30, 2022 and Korean Patent Application No. KR10-2023-0001007 filed on January 4, 2023.
Drawings
The 9-page drawings have been considered and placed on record in the file.
Status of Claims
Claims 2-6, 8-14, and 18 are pending. Claims 1, 7, and 15-17 are canceled.
Response to Amendment
The amendment filed 12/21/2025 has been entered. Claims 2-6, 8-14, and 18 remain pending in the application.
Response to Arguments
Applicant’s arguments, see pages 12-15 of Remarks, filed 12/21/2025, with respect to the rejections od claims 1, 11, and 18 under 35 USC 103 have been fully considered and are persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made in view of training the object recognition model by using the object of interest and object of non-interest, as well as the user designating the object of interest in the acquired image.
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 2, 5-6, 8, 11-14, and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Chen et al., (US 2018/0341813 A1, published 11/29/2018, previously cited in an IDS filed by applicant), hereinafter referred to as Chen, in view of Carmena et al., (US 2021/0365739 A1, published 11/25/2021), hereinafter referred to as Carmena.
Claim 5
Chen discloses a false detection removal method (Chen, [0007], “The techniques and systems described herein provide a false positive removal mechanism that can be applied to remove false positives”) of an image processing device (Chen, Fig. 1) , the method comprising:
detecting an object of interest (Chen, [0109], “The blob detection system 104 can detect one or more blobs in video frames (e.g., video frames 102) of a video sequence, and the object tracking system 106 can track the one or more blobs across the frames of the video sequence”) based on an object of interest recognition model (Chen, Fig. 3) in an image (Fig. 3, video frames 302) acquired from an image capture device (Chen, [0029], “the apparatus includes a camera for capturing the one or more video frames.”);
removing feature-based false detection of the object of interest (Chen, [0192], “the strong pixel level analysis may include both AMFR and a Correlation based False positive Removal (CFR) to determine whether a current object is a true positive or a false positive.” [0197], “A high correlation value (e.g., higher than a correlation threshold) between the texture (pixel values) of a current frame and the co-located texture (pixel values) of the mean picture indicates that the textures are expected to be statistically similar (e.g., linearly or other statistical relationship) and thus may be considered as false positives because the background texture is similar to the current texture. For example, the background texture of the mean picture indicates the texture of the background area, and the high correlation indicates the similarity of the two textures. If the texture of the patch in the current frame (of the current blob) is statistically similar to the texture of the co-located patch (based on the high correlation value) in the background represented by the mean picture, the patch in the current frame is also statistically likely to be a background patch, in which case the object represented by the current blob (and the corresponding tracker) can be considered as a false positive.”, [0216], “if a positive result is determined at step 620, the tracker may be kept in the memory for the next round of analysis (instead of being output), and, if a negative result is found at step 620, the tracker can be killed and removed from the trackers maintained by the video analytics system (instead of being held for the next round)”) based on a first false detection filtering model (Chen, [0193], “To perform CFR, the true-false positive detection process can determine a correlation between the current texture of the blob (representing the object) associated with the current tracker in the current frame and the co-located texture of a mean background picture (referred to herein as a mean picture).”, In the Detailed Description of the application, [0140] “The processor 250 may calculate a similarity indicating a correlation between a first feature vector of the object of interest and a second feature vector of the recognized object as a result of the object recognition operation as a numerical value. The processor 250 may check whether false detection is detected depending on the similarity.”, and image feature vector is an abstraction of an image used to characterize and numerically quantify the contents of an image which means the pixel value is a feature vector, [0196], “A correlation can be calculated between the texture of the current blob associated with the current tracker (for the current picture) and a corresponding, co-located texture of the mean picture. For example, the pixel values of the current blob in the current picture can be compared to pixel values in the mean picture that are in the same coordinate locations as the pixels in the current picture (e.g., a correlation can be calculated for pixels located at a 0,0 coordinate of the current frame and the mean picture). A correlation (also referred to as a correlation coefficient) is a number that quantifies some type of correlation and dependence, meaning statistical relationships between two or more random variables or observed data values.”);
removing color-based false detection of the object of interest (Chen, [0192], “the strong pixel level analysis may include both AMFR and a Correlation based False positive Removal (CFR) to determine whether a current object is a true positive or a false positive.” [0164], “The appearance based true-false positive detection process can apply an Appearance Model based False positive Removal (AMFR) process, which itself works as a strong indication of whether a current object is true positive or a false positive.”, [0165], “The appearance characteristics used by the AMFR process can include color characteristics (e.g., an appearance model, a color mass center, and/or other color characteristics) that are determined for a tracker. [0213], “f the AMFR process returns a negative result (the distance between the mass centers C.sub.0 and C.sub.t is below the threshold distance), the tracker is killed) based on a second false detection filtering model (Chen, [0192], “the strong pixel level analysis may include both AMFR and a Correlation based False positive Removal (CFR) to determine whether a current object is a true positive or a false positive.” [0164], “The appearance based true-false positive detection process can apply an Appearance Model based False positive Removal (AMFR) process, which itself works as a strong indication of whether a current object is true positive or a false positive.”, [0165], “The appearance characteristics used by the AMFR process can include color characteristics (e.g., an appearance model, a color mass center, and/or other color characteristics) that are determined for a tracker. The color characteristics can be based on the pixels in a region of a frame included in a tracker's bounding region (e.g., a bounding box of the tracker). Color characteristics of a tracker can be determined in multiple frames of a video sequence, and can be compared to determine a difference between the color characteristics in the multiple frames. The status of the blob tracker can then be determined based on the difference between the color characteristics. For example, a determined status of a tracker can include outputting the tracker immediately, removing the tracker from a list of trackers maintained for the video sequence (also referred to as killing the tracker), and transitioning the tracker to an intermediate status. A tracker with an intermediate status can be kept or maintained for future iterations of the true-false positive detection process, in which case there is no need to either kill or output the tracker immediately.”); and
acquiring a final object of interest without the false detection (Chen, [0136], “as shown in FIG. 6A, if the strong pixel level analysis returns a positive result at step 620 (e.g., AMFR determines the distance is greater than the threshold distance and the correlation is determined to be below the correlation threshold), the tracker can be output at step 626 (e.g., immediately for the current frame, once all trackers for the current frame have been processed, or at some other time)”).
Chen does not explicitly disclose further comprising training the object of interest recognition model, wherein the training of the object of interest recognition model comprising: receiving a user input designating the object of interest (OB1) in accordance with a user selecting displayed objects in the image, wherein the image is acquired from the image capture device; generating an object of non-interest (Nr) in at least a portion of a region except for the object of interest (OB1) in the image; and training the object of interest recognition model using the object of interest (OB1) and the object of non-interest (Nr) as training data.
However, Carmena teaches training the object of interest recognition model (Carmena,[0026], “the detection model may be a machine-learning network trained for object detection”), wherein the training of the object of interest recognition model (Carmena,[0026], “the detection model may be a machine-learning network trained for object detection”) comprising: receiving a user input designating the object of interest (OB1) in accordance with a user selecting displayed objects in the image, wherein the image is acquired from the image capture device (Carmena, [0038], “an image obtained by the input source 102 is shown to a user. The user can select portions of the image as background and portions of the image as non-background elements. The selections by the user can be used to determine the background of the image obtained by the input source 102 as well as similar images obtained by the input source 102 or other input sources in a system such as the system 100”, [0096], “For example, a user can take an image of a scene. The user can manually assign identifiers to the various elements in the scene. The manually assigned identifiers can be sent to the validation engine 116. The validation engine 116 can receive the detections 112 and cross-reference known identities with the identities of the detections 112”); generating an object of non-interest (Nr) in at least a portion of a region except for the object of interest (OB1) in the image (Carmena, Abstract, “for false detection removal using adversarial masks”, [0011], “In some implementations, images with multiple false detections are used to compute one or more adversarial masks. For example, an image containing a tree, a house, and a car, each falsely identified, can create three adversarial masks corresponding to each false identification that can be added to any new image that contains the corresponding element that has been falsely identified. The adversarial mask generated based on the false identification of the tree can be added to new images containing the tree. The adversarial mask generated based on the false identification of the house can be added to new images containing the house. The adversarial mask generated based on the false identification of the car can be added to new images containing the car. In general, any object detected using a detection model can have a corresponding adversarial mask generated.”, [0049], “the detections 112 include coordinates for each false detection detected within the training image 104. For example, the detection engine 108, which can be a form of deep object detector, obtains one or more coordinates that correspond to one or more of the detections 112. The detections 112 can include coordinates as well as a classifier for each detection. In some cases, a false detection object class can be used to distinguish one or more detections of the detections 112 as false detections. In some implementations, a first false detection includes the false detection object class, a first x coordinate, a first y coordinate, a second x coordinate, and a second y coordinate in a given two-dimensional (2D) x-y plane of the training image 104. The first x coordinate and the first y coordinate correspond to a corner of a bounding box that bounds a first false detection. The second x coordinate and the second y coordinate correspond to another corner of a bounding box that bounds the first false detection.”); and training the object of interest recognition model (Carmena, as shown I Fig. 1, the training image 104 has both the object of interest which is the human and the object of non-interest which is the tree, [0042], “Data related to the false detection determination made by the validation engine 116 is sent to the mask generation engine 120. The mask generation engine 120 is an alternative to fine-tuning a given detection model based on false or invalid detections. In some cases, one or more visual images can include background objects that can be detected by machine-learning-based object detectors. In some cases, background objects are detected as foreground objects. In some cases, background objects or foreground objects are falsely identified.”) using the object of interest (OB1) (Carmena, [0077], “The user can provide feedback that can be used by the validation engine 116 to label one or more detections as true or false”, true detection is analogous to the object of interest) and the object of non-interest (Nr) as training data. (Carmena, [0055], “the false detections include a class object that identifies the false detection as well as a location corresponding to the location of the false detection with regard to the input image or the training image 104”, Fig. 1, Figure 148 shows the final detection where the human is the object of interest and the tree is the object of non-interest which is removed by using an adversarial mask).
Chen and Carmena are both considered to be analogous to the claimed invention because they are in the same field of false-positive detection. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method as taught by Chen to incorporate the teachings of Carmena of training the object of interest recognition model, wherein the training of the object of interest recognition model comprising: receiving a user input designating the object of interest (OB1) in accordance with a user selecting displayed objects in the image, wherein the image is acquired from the image capture device; generating an object of non-interest (Nr) in at least a portion of a region except for the object of interest (OB1) in the image; and training the object of interest recognition model using the object of interest (OB1) and the object of non-interest (Nr) as training data. Such a modification is the result of combining prior art elements according to known methods to yield predictable results. The motivation for the proposed modification would have been to so that the detection engine can be trained based on selections made by the user to improve detection of backgrounds within obtained images (Carmena, [0039]).
Claim 2
The combination of Chen in view of Carmena discloses the false detection removal method of claim 5 (Chen, [0007], “The techniques and systems described herein provide a false positive removal mechanism that can be applied to remove false positives”), further comprising training the first false detection filtering model (Chen, [0193], “To perform CFR, the true-false positive detection process can determine a correlation between the current texture of the blob (representing the object) associated with the current tracker in the current frame and the co-located texture of a mean background picture (referred to herein as a mean picture).”, In the Detailed Description of the application, [0140] “The processor 250 may calculate a similarity indicating a correlation between a first feature vector of the object of interest and a second feature vector of the recognized object as a result of the object recognition operation as a numerical value. The processor 250 may check whether false detection is detected depending on the similarity.”, and image feature vector is an abstraction of an image used to characterize and numerically quantify the contents of an image which means the pixel value is a feature vector, [0196], “A correlation can be calculated between the texture of the current blob associated with the current tracker (for the current picture) and a corresponding, co-located texture of the mean picture. For example, the pixel values of the current blob in the current picture can be compared to pixel values in the mean picture that are in the same coordinate locations as the pixels in the current picture (e.g., a correlation can be calculated for pixels located at a 0,0 coordinate of the current frame and the mean picture). A correlation (also referred to as a correlation coefficient) is a number that quantifies some type of correlation and dependence, meaning statistical relationships between two or more random variables or observed data values.”), wherein the training of the first false detection filtering model includes: extracting a specific vector of the object of interest designated in advance; and modeling a distribution of the object of interest on a coordinate space based on the feature vector (Chen, [0194], “a mean picture can be generated using a Gaussian mixture model (GMM) for each pixel location. For example, a pixel value of a synthesis mean picture for a pixel location can be set as the expectation (or average or mean) of a model from the GMM for that pixel location, without taking into account whether the current pixel belongs to a background pixel or foreground pixel. In some examples, the model is chosen as the most probable model, which is the model with a highest weight from the GMM for a current pixel location can be used to synthesize the mean picture for that pixel location. The model with the highest weight from a GMM is referred to herein as the most probable model. In some examples, the model from the GMM for a current pixel location whose distance to the current input pixel (in a current frame) is the smallest among all the existing models in the GMM for the current pixel location can be used to synthesize the mean picture for that pixel location. The model from a GMM for a pixel location with the smallest distance to the current input pixel is referred to herein as the closest model.”).
Claim 6
The combination of Chen in view of Carmena discloses the false detection removal method of claim 5 (Chen, [0007], “The techniques and systems described herein provide a false positive removal mechanism that can be applied to remove false positives”) comprising: performing first learning using the trained object recognition model (Carmena,[0026], “the detection model may be a machine-learning network trained for object detection”); and additionally performing training N times after the first learning (Carmena, [0056], “The BCE loss 122 is back-propagated through the detection model 110 of the system 100. By back-propagating the BCE loss 122, shown in item 123, the mask generation engine 120 determines gradients of the BCE loss 122. The gradients of the BCE loss 122 are determined with respect to the input image corresponding to the training image 104. The mask generation engine 120 uses the gradients within the process of adding noise to the input image..”); and automatically extracting location information of an erroneously detected object based on an immediately previous learning result for each learning and changing the erroneously detected object into the object of non-interest (Carmena, [0066], “The locations of the one or more false detections received by the mask generation engine 120 are used to obtain portions of the adversarial auxiliary image that corresponds to locations of the one or more false detections. The portions of the adversarial auxiliary image are used to create an adversarial mask 128 shown visually in visual representation 130. The adversarial mask's visual representation 130 shows a modified portion 130a of the input image as a rectangular shape cropped from the generated adversarial auxiliary image. The rectangular shape of the modified portion 130a corresponds to the location of the false detection shown in the detection region 114a. The values of pixels corresponding to the modified portion 130a can be added to a corresponding portion of a second input image, such as the new sample image 134.”).
Chen and Carmena are both considered to be analogous to the claimed invention because they are in the same field of false-positive detection. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method as taught by Chen to incorporate the teachings of Carmena of performing first learning using the trained object recognition model, and additionally performing training N times after the first learning), and automatically extracting location information of an erroneously detected object based on an immediately previous learning result for each learning and changing the erroneously detected object into the object of non-interest. Such a modification is the result of combining prior art elements according to known methods to yield predictable results. The motivation for the proposed modification would have been to so that the detection engine can be trained based on selections made by the user to improve detection of backgrounds within obtained images (Carmena, [0039]).
Claim 11-12 and 8 are rejected for similar reasons as those described in claims 5-6 and 2, respectively. The additional elements in Claims 11 and 8 (the combination of Chen in view of Carmena) discloses includes: an image processing device (Chen, Fig. 1), comprising: an image acquisitor (Chen, [0029], “the apparatus includes a camera for capturing the one or more video frames.”); a storage configured to store (Chen, [0014], “an apparatus for maintaining blob trackers for one or more video frames is provided that includes a memory configured to store video data and a processor”) a previously trained object of interest recognition model (Chen, Fig. 1), a first false detection filtering model (Chen, [0192], “the strong pixel level analysis may include both AMFR and a Correlation based False positive Removal (CFR) to determine whether a current object is a true positive or a false positive”), and a second false detection filtering model (Chen, [0192], “the strong pixel level analysis may include both AMFR and a Correlation based False positive Removal (CFR) to determine whether a current object is a true positive or a false positive”); and a processor (Chen, [0014], “an apparatus for maintaining blob trackers for one or more video frames is provided that includes a memory configured to store video data and a processor”). The proposed combination as well as the motivation for combining the Chen and Carmena references presented in the rejection of Claim 5, apply to Claims 11-12 and 8 and are incorporated herein by reference. Thus, the device recited in Claims 11 -12 and 8 is met by Chen and Carmena.
Claim 13
The combination of Chen in view of Carmena discloses the image processing device of claim 11 (Chen, Fig. 1), further comprising: a wireless communication unit, and wherein the image acquisition unit is configured to obtain a captured image from an external image capture device through the wireless communication unit (Chen, [0378], “some cases, the source device and the destination device may be equipped for wireless communication.”, [0379], “The encoded video data may be modulated according to a communication standard, such as a wireless communication protocol, and transmitted to destination device. The communication medium may comprise any wireless or wired communication medium, such as a radio frequency (RF) spectrum or one or more physical transmission lines. The communication medium may form part of a packet-based network, such as a local area network, a wide-area network, or a global network such as the Internet. The communication medium may include routers, switches, base stations, or any other equipment that may be useful to facilitate communication from source device to destination device.”).
Claim 14
The combination of Chen in view of Carmena discloses the image processing device of claim 11 (Chen, Fig. 1), further comprising a wireless communication unit (Chen, [0378], “some cases, the source device and the destination device may be equipped for wireless communication.”, [0379], “The encoded video data may be modulated according to a communication standard, such as a wireless communication protocol, and transmitted to destination device. The communication medium may comprise any wireless or wired communication medium, such as a radio frequency (RF) spectrum or one or more physical transmission lines. The communication medium may form part of a packet-based network, such as a local area network, a wide-area network, or a global network such as the Internet. The communication medium may include routers, switches, base stations, or any other equipment that may be useful to facilitate communication from source device to destination device.”), wherein the processor (Chen, [0014], “an apparatus for maintaining blob trackers for one or more video frames is provided that includes a memory configured to store video data and a processor”) is configured to transmit the object of interest recognition model, the first false detection filtering model, and the second false detection filtering model stored in the storage unit to an image capture device through the wireless communication unit (Chen, [0103], “An IP camera can be used to send and receive data via a computer network and the Internet. In some cases, IP camera systems can be used for two-way communications. For example, data (e.g., audio, video, metadata, or the like) can be transmitted by an IP camera using one or more network cables or using a wireless network, allowing users to communicate with what they are seeing. In one illustrative example, a gas station clerk can assist a customer with how to use a pay pump using video data provided from an IP camera (e.g., by viewing the customer's actions at the pay pump). Commands can also be transmitted for pan, tilt, zoom (PTZ) cameras via a single network or multiple networks. Furthermore, IP camera systems provide flexibility and wireless capabilities.”).
Claim 18 are rejected for similar reasons as those described in claim 5. The additional elements in Claim 18 (the combination of Chen in view of Carmena) discloses includes: an image processing device (Chen, Fig. 1), comprising: an image processing device of comprising (Chen, Fig. 1): an image acquisitor (Chen. [0029], “the apparatus includes a camera for capturing the one or more video frames”; a storage (Chen, [0100], “term “computer-readable medium” includes, but is not limited to, portable or non-portable storage devices, optical storage devices, and various other mediums capable of storing, containing, or carrying instruction(s) and/or data”) configured to store a previously trained object of interest recognition model (Chen, Fig. 3, [0105], “video analytics can be trained”), a first false detection filtering model (Chen, [0192], “the strong pixel level analysis may include both AMFR and a Correlation based False positive Removal (CFR) to determine whether a current object is a true positive or a false positive.”), and a second false detection filtering model (Chen, [0192], “the strong pixel level analysis may include both AMFR and a Correlation based False positive Removal (CFR) to determine whether a current object is a true positive or a false positive.”); a processor (Chen, [0303], “the computing device or apparatus may include a processor, microprocessor, microcomputer, or other component of a device that is configured to carry out the steps of process 1100”) configured to detect an object of interest (Chen, [0109], “The blob detection system 104 can detect one or more blobs in video frames (e.g., video frames 102) of a video sequence, and the object tracking system 106 can track the one or more blobs across the frames of the video sequence”) based on an object of interest recognition model (Chen, Fig. 3) from an image (Fig. 3, video frames 302) acquired from an image acquisitor (Chen, [0029], “the apparatus includes a camera for capturing the one or more video frames.”). The proposed combination as well as the motivation for combining the Chen and Carmena references presented in the rejection of Claim 5, apply to Claim 18 and are incorporated herein by reference. Thus, the device recited in Claim 18 is met by Chen and Carmena.
Claims 3 and 9 are rejected under 35 U.S.C. 103 as being unpatentable over Chen in view of Carmena in further view of Poole (US 2013/0044927 A1, published 02/21/2013), hereinafter referred to as Poole.
Claim 3
The combination of Chen in view of Carmena discloses the false detection removal method of claim 2 (Chen, [0007], “The techniques and systems described herein provide a false positive removal mechanism that can be applied to remove false positives”).
The combination of Chen in view of Carmena does not explicitly disclose further comprising determining an object as an erroneously detected object when a Mahalanobis distance of a feature vector of the object of interest detected by the object of interest recognition model is greater than or equal to a threshold distance.
However, Poole teaches determining an object as an erroneously detected object when a Mahalanobis distance of a feature vector of the object of interest detected by the object of interest recognition model is greater than or equal to a threshold distance (Poole, [0070], “the outlier probability for each point is determined using Hotelling's T.sup.2 statistic, which is a multi-dimensional generalization of the univariate t-test. For the case of testing a single M-dimensional sample against a distribution estimated from a sufficiently large sample, the probability reduces to a .chi..sup.2 function of the Mahalanobis distance squared, D.sub.j.sup.2, with M degrees of freedom. Thus the threshold on Mahalanobis distance can be set given M and the chosen false positive operating threshold. The false positive operating threshold is selected by an operator in some embodiments.”, [0072], “The selected threshold defines ellipsoids in feature space representing `normality` as illustrated in FIG. 6. FIG. 6 shows the feature space for a single atlas position. Sample points from the training data sets representing normal anatomy for the atlas position are shown as + signs. The dotted line represents a contour of equal Mahalonobis distance of selected threshold value D. A feature vector measured for a corresponding voxel in a patient image data set is show by an X. In this case it can be seen that the point X has a Mahalanobis distance D.sub.x greater than the selected threshold Mahalanobis distance D, and the voxel would therefore be identified as potentially representing an abnormality.”).
Chen, Carmena, and Poole are both considered to be analogous to the claimed invention because they are in the same field of false-positive detection. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method as taught by Chen to incorporate the teachings of Poole of determining an object as an erroneously detected object when a Mahalanobis distance of a feature vector of the object of interest detected by the object of interest recognition model is greater than or equal to a threshold distance. Such a modification is the result of combining prior art elements according to known methods to yield predictable results. The motivation for the proposed modification would have been because the Mahalanobis distance method can provide an accurate transformation across all regions covered by the training data sets (Poole, [0057]).
Claim 9 is rejected for similar reasons as those described in claims 3. The additional elements in Claim 9 (Chen, Carmena, and Poole) discloses includes: an image processing device (Chen, Fig. 1), comprising: an image acquisitor (Chen, [0029], “the apparatus includes a camera for capturing the one or more video frames.”); a storage configured to store (Chen, [0014], “an apparatus for maintaining blob trackers for one or more video frames is provided that includes a memory configured to store video data and a processor”) a previously trained object of interest recognition model (Chen, Fig. 1), a first false detection filtering model (Chen, [0192], “the strong pixel level analysis may include both AMFR and a Correlation based False positive Removal (CFR) to determine whether a current object is a true positive or a false positive”), and a second false detection filtering model (Chen, [0192], “the strong pixel level analysis may include both AMFR and a Correlation based False positive Removal (CFR) to determine whether a current object is a true positive or a false positive”); and a processor (Chen, [0014], “an apparatus for maintaining blob trackers for one or more video frames is provided that includes a memory configured to store video data and a processor”). The proposed combination as well as the motivation for combining the Chen, Carmena, and Poole references presented in the rejection of Claim 3, apply to Claim 9 and are incorporated herein by reference. Thus, the device in Claim 9 is met by Chen, Carmena, and Poole.
Claims 4 and 10 rejected under 35 U.S.C. 103 as being unpatentable over Chen in view of Carmena in further view of Nanni et al., ("Face Detection Ensemble with Methods Using Depth Information to Filter False Positives", published 2019), hereinafter referred to as Nanni.
Claim 4
The combination of Chen in view of Carmena discloses the false detection removal method of claim 1 (Chen, [0007], “The techniques and systems described herein provide a false positive removal mechanism that can be applied to remove false positives”), further comprising raining the second false detection filtering model (Chen, [0192], “the strong pixel level analysis may include both AMFR and a Correlation based False positive Removal (CFR) to determine whether a current object is a true positive or a false positive.” [0164], “The appearance based true-false positive detection process can apply an Appearance Model based False positive Removal (AMFR) process, which itself works as a strong indication of whether a current object is true positive or a false positive.”, [0165], “The appearance characteristics used by the AMFR process can include color characteristics (e.g., an appearance model, a color mass center, and/or other color characteristics) that are determined for a tracker. The color characteristics can be based on the pixels in a region of a frame included in a tracker's bounding region (e.g., a bounding box of the tracker). Color characteristics of a tracker can be determined in multiple frames of a video sequence, and can be compared to determine a difference between the color characteristics in the multiple frames. The status of the blob tracker can then be determined based on the difference between the color characteristics. For example, a determined status of a tracker can include outputting the tracker immediately, removing the tracker from a list of trackers maintained for the video sequence (also referred to as killing the tracker), and transitioning the tracker to an intermediate status. A tracker with an intermediate status can be kept or maintained for future iterations of the true-false positive detection process, in which case there is no need to either kill or output the tracker immediately.”).
The combination of Chen in view of Carmena does not explicitly disclose wherein the training of the second false detection filtering model may include: extracting color information of the object of interest designated in advance; and acquiring a primary color of the object of interest by analyzing colors in the CIE-LAB color space based on the color information.
However, Nanni teaches wherein the training of the second false detection filtering model (Nanni, Section 2, Section 2.3, Filtering Steps, “As noted in Figure 1, some of the false positives generated by the ensemble of classifiers are extracted by applying several filtering approaches that take advantage of the depth maps. The filters tested in this work are the set of six tested in (viz. SIZE, STD, SEG, ELL, EYE, and SEC) and a new filter proposed here (viz. WAV), which is based on processing the image with different wavelets.”) may include: extracting color information of the object of interest designated in advance; and acquiring a primary color of the object of interest by analyzing colors in the CIE-LAB color space based on the color information (Nanni, Section 2.1, “Every sample in the Kinetic depth map corresponds to a 3D point, pi , i = 1, . . . , N, with N the number of points. The joint calibration of the depth and color cameras, as described in [57], allows a reprojection of the depth samples over the corresponding pixels in the color image so that each point is associated with the 3D spatial coordinates (x, y, and z) of pi and its RGB color components. Since these two representations lie in entirely different spaces, they cannot be compared directly, and all components must be comparable to extract multidimensional vectors that are appropriate for the mean shift clustering algorithm. Thus, a conversion is performed so that the color values lie in the CIELAB uniform color space, which represents color in three dimensions expressed by values representing lightness (L) from black (0) to white (100), a value (a) from green (−) to red (+), and a value (b) from blue (−) to yellow (+). This introduces a perceptual significance to the Euclidean distance between the color vectors that can be used in the mean shift algorithm.”).
Chen, Carmena, and Nanni are both considered to be analogous to the claimed invention because they are in the same field of false-positive detection. Therefore, it would have been obvious to someone of ordinary skill in the art before the effective filing date of the claimed invention to have modified the method as taught by Chen to incorporate the teachings of Nanni wherein the training of the second false detection filtering model may include: extracting color information of the object of interest designated in advance; and acquiring a primary color of the object of interest by analyzing colors in the CIE-LAB color space based on the color information. Such a modification is the result of combining prior art elements according to known methods to yield predictable results. The motivation for the proposed modification would have been to reduce the number of false positives without decreasing the detection rate (Nanni, Abstract).
Claim 10 is rejected for similar reasons as those described in claim 4. The additional elements in Claim 10 (Chen, Carmena, and Nanni) discloses includes: an image processing device (Chen, Fig. 1), comprising: an image acquisitor (Chen, [0029], “the apparatus includes a camera for capturing the one or more video frames.”); a storage configured to store (Chen, [0014], “an apparatus for maintaining blob trackers for one or more video frames is provided that includes a memory configured to store video data and a processor”) a previously trained object of interest recognition model (Chen, Fig. 1), a first false detection filtering model (Chen, [0192], “the strong pixel level analysis may include both AMFR and a Correlation based False positive Removal (CFR) to determine whether a current object is a true positive or a false positive”), and a second false detection filtering model (Chen, [0192], “the strong pixel level analysis may include both AMFR and a Correlation based False positive Removal (CFR) to determine whether a current object is a true positive or a false positive”); and a processor (Chen, [0014], “an apparatus for maintaining blob trackers for one or more video frames is provided that includes a memory configured to store video data and a processor”). The proposed combination as well as the motivation for combining the Chen, Carmena and Nanni references presented in the rejection of Claim 4, apply to Claim 10 and are incorporated herein by reference. Thus, the device in Claim 10 is met by Chen, Carmena, and Nanni.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to DENISE G ALFONSO whose telephone number is (571)272-1360. The examiner can normally be reached Monday - Friday 7:30 - 5:30.
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, Amandeep Saini can be reached at (571)272-3382. 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.
/DENISE G ALFONSO/Examiner, Art Unit 2662
/AMANDEEP SAINI/Supervisory Patent Examiner, Art Unit 2662