DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
This action is in response to the Amendment filed on 5/29/2026.
Claims 1-15, 17-21 are pending. Claim 1, 10, 17 have been amended. Claim 16 has been cancelled, Claim 21 is newly added.
Continued Examination Under 37 CFR 1.114
A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant's submission filed on 5/29/2026 has been entered.
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-4, 6-8, 10-12, 14-16, 17-19, 21 are rejected under 35 U.S.C. § 103 as being unpatentable over Cowburn et al. (US 20200312008 A1) in view of Feng et al. (Interactive Segmentation on RGB-0 Images via Cue Selection, CVPR 2016, hereinafter Feng), further in view of Yang et al. (US 20160217552 A1, hereinafter Yang).
Regarding Claim 10, Cowburn teaches A system comprising: a memory component; and a processing device coupled to the memory component, the processing device to perform operations comprising: (Cowburn, Fig. 1, Element 100, messaging system; Paragraph [0015], "FIG. 10 is a block diagram illustrating components of a machine ... able to read instructions from a machine-readable medium" <read on memory component and processing device>); receiving a textured mesh representing multiple virtual materials present on a surface of an object; (Cowburn, Paragraph [0017], "accessing an image that comprises a set of image features"; Paragraph [0017], "generating a 3D model, such as a 3D mesh model, based on the set of image features of the image, the 3D model comprising a plurality of vertices" <read on textured mesh>; Paragraph [0020], "Semantic features may for example include ... material parameters" <read on multiple virtual materials>; Paragraph [0060], "the semantic labeling module 308 assigns a material parameter to a feature class that corresponds with a semantic label from among the set of semantic label ... The material parameters may be defined based on a set of interaction variables" <read on materials present on a surface of an object>); [[ receiving an input indicating one or more regions of the textured mesh corresponding to a material type; ]] determining one or more three-dimensional features of the object represented by the textured mesh; (Cowburn, Paragraph [0017], "generating a 3D model, such as a 3D mesh model, based on the set of image features of the image, the 3D model comprising a plurality of vertices" <read on determining one or more three-dimensional features of the object, as the vertices of the 3D mesh define three-dimensional features of the modeled object>; Paragraph [0024], "based on the coordinates of the vertices" <read on determining three-dimensional features of the object>); segmenting pixels corresponding to the material type from other pixels of the multiple virtual materials of the textured mesh, using an algorithm to identify an additional region of the textured mesh corresponding to the material type by comparing three-dimensional positions, [[ color values, and normal properties ]] of pixels on the one or more three-dimensional features of the textured mesh (Cowburn, Paragraph [0024], "The semantic texture map system projects the semantic segmentation image that corresponds with the image upon the 3D model based on the coordinates of the vertices" <read on comparing three-dimensional positions of pixels on the one or more three-dimensional features>; Paragraph [0024], "maps the semantic feature labels of semantic segmentation image to the 2D texture coordinates of the UV map" <read on segmenting pixels corresponding to the material type>); [[ identifying a boundary between the material type and a different material type depicted on the textured mesh based on the segmenting pixels; and ]] generating a material map that includes a selectable portion corresponding to the material type of the one or more regions of the textured mesh and the additional region of the textured mesh [[ and identified by the boundary. ]] (Cowburn, Paragraph [0017], "generating a semantic texture map" <read on material map>; Paragraph [0024], "the resulting semantic texture map therefor comprises a set of texels, where each texel comprises one or more semantic feature labels" <read on selectable portion corresponding to the material type, as each texel labeled with a semantic material label constitutes a portion of the map corresponding to that material type that is capable of being selected>).
But Cowburn does not explicitly disclose receiving an input indicating one or more regions of the textured mesh corresponding to a material type; and comparing... color values, and normal properties.
However, Feng teaches receiving an input indicating one or more regions of the textured mesh corresponding to a material type (Feng, Page 5, Section 3.2.4, "user inputs typically can take one of the following forms: 1) foreground / background clicks"); comparing... color values, and normal properties of pixels on the one or more three-dimensional features of the textured mesh (Feng, Page 3, Section 3.2, "Color is useful for identifying foregrounds with different appearance from their backgrounds" <read on comparing color values>; Feng, Page 3, Section 3.2, "normal vectors are computed from a depth-projected 3D point cloud" <read on normal properties of pixels on the one or more three-dimensional features>).
Feng and Cowburn are analogous since both address segmentation and labeling of regions on 3D meshes or image data for material or semantic classification. Cowburn provides a way of generating a semantic texture map based on automated semantic segmentation that assigns semantic labels to different regions of a 3D mesh based on visual features derived from image inputs. Feng provides a way of performing interactive segmentation in which a user provides input clicks or strokes to indicate which regions should be assigned to a particular foreground/material class, with per-pixel cues including color and surface normals that are inherently geometric properties tied to the three-dimensional features of the object. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate the user-guided input mechanism and the color/normal cue evaluation taught by Feng into the modified invention of Cowburn such that instead of relying solely on automated semantic segmentation the system accepts a user-provided input indicating one or more regions of the textured mesh corresponding to a material type and additional regions of the same material type are identified by jointly comparing color values and normal properties of pixels on the three-dimensional features already provided by Cowburn's mesh. The motivation is to reduce reliance on extensive training data, to provide user control over the material assignment process, and to improve segmentation accuracy on textured RGB-D / 3D surfaces, as discussed by Feng in Page 1, Abstract, "with significant fewer user inputs".
But the combination of Cowburn and Feng does not explicitly disclose identifying a boundary between the material type and a different material type depicted on the textured mesh based on the segmenting pixels; or that the generated material map is identified by the boundary.
However, Yang teaches identifying a boundary between the material type and a different material type depicted on the textured mesh based on the segmenting pixels (Yang, Paragraph [0038], "the control strengths are often blurred at boundaries between two different regions, e.g., a texture region and a non-texture region" <read on boundary between the material type and a different material type>; Yang, Paragraph [0004], "A directional edge map is generated based on a data correlation analysis by removing a correlation response in texture regions based on the texture map and the label map" <read on identifying a boundary based on the segmenting pixels>); generating a material map that includes a selectable portion corresponding to the material type of the one or more regions of the textured mesh and the additional region of the textured mesh and identified by the boundary (Yang, Paragraph [0004], "A gain map with boundaries and smoothing is generated using localized pixels based on information from one or more of the label map, the texture map, and the directional edge map" <read on material map that includes a selectable portion corresponding to the material type and identified by the boundary, as the labeled region bounded by the identified boundary constitutes a discrete portion of the map that is capable of being selected>; Yang, Paragraph [0065], "the processor uses the gain map to control enhancement strength of pixel details in super-resolution by distinguishing boundaries between a non-texture pixel region and a texture pixel region").
Yang and the combination of Cowburn and Feng are analogous since all deal with pixel-level segmentation and classification of surface regions based on their material/texture properties in image processing systems, and each generates a map that assigns classification labels to identified pixel regions. The combination of Cowburn and Feng provides a way of generating a semantic texture map that segments and labels regions of a 3D mesh surface according to material type using user input together with color and normal cues. Yang provides a way of identifying boundaries between differently classified pixel regions arising from the segmentation and generating a map that captures those boundaries together with the classified regions so that each labeled region is bounded and independently addressable. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate the boundary identification and boundary-indicating map generation taught by Yang into the modified semantic texture mapping system of Cowburn and Feng, such that the system additionally identifies a boundary between the material type and a different material type on the textured mesh based on the pixel segmentation result and generates a material map that includes, for the material type, a portion made up of the one or more marked regions and the additional identified region that is delimited (and thereby selectable) by the identified boundary. The motivation is to preserve boundary accuracy between different material type regions and to produce discrete, addressable per-material regions in the output map, as discussed by Yang in Paragraph [0038].
Regarding Claim 1, it recites limitations similar in scope to the limitations of Claim 10 but as a method and the combination of Cowburn, Feng and Yang teaches all the limitations as of Claim 10. Therefore is rejected under the same rationale.
Regarding Claim 2, The combination of Cowburn, Feng and Yang teaches the invention in Claim 1.
The combination further teaches further comprising identifying the additional region of the textured mesh by comparing color values and normal properties of the pixels of the textured mesh using the algorithm (Feng, Page 3, Section 3.2, "Color is useful for identifying foregrounds with different appearance from their backgrounds"; Feng, Page 4, Section 3.2.1, "For color, we convert RGB value into LAB space and use L2 norm as distance"; Feng, Page 3, Section 3.2, "normal vectors <read on normal properties> are computed from a depth-projected 3D point cloud"; Feng, Page 4, Section 3.2.1, "As for normal, cosine similarity is used to compute the similarity score between two unit normal vectors and then converted to a distance measure").
Feng and Cowburn are analogous since both address segmentation of image/mesh regions based on appearance features, including color and surface orientation. Feng provides an explicit teaching to measure similarity using color distances (e.g., LAB space with L2 norm) and normal-vector similarity (cosine similarity), while Cowburn's semantic texture map system relies on feature-based semantic segmentation without detailing such cue metrics. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Cowburn's semantic texture map generation in view of Feng so that additional regions of the textured mesh corresponding to a material type are identified by jointly evaluating color values and normal properties of pixels/vertices, as Feng teaches, in order to improve segmentation accuracy in textured RGB-D scenes.
Regarding Claim 3, the combination of Cowburn, Feng and Yang teaches the invention in Claim 2.
The combination further teaches further comprising receiving an additional input to change a relative importance of three-dimensional positions, the color values, and the normal properties of the pixels of the textured mesh and updating the algorithm based on the additional input (Feng, Page 5, Section 4.4, "The pairwise weight A is set to 0.1 which we find produces the best results for our algorithm").
Feng and Cowburn are analogous since both involve algorithms whose performance depends on relative weighting of pairwise or cue-based terms in a segmentation of energy. Feng explicitly teaches adjusting a pairwise weight A to obtain improved results and identifies that tuning such weights affects the relative importance of different cues (e.g., color, normals, spatial relationships) in the segmentation. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Feng's teaching into Cowburn's semantic texture map system so as to allow receipt of additional user input to change relative importance among three-dimensional positions, color values, and normal properties, and to update the algorithm, accordingly, thereby providing controllable cue weighting to refine material segmentation quality.
Regarding Claim 4, the combination of Cowburn, Feng and Yang teaches the invention in Claim 1.
The combination further teaches wherein the material map includes texture IDs corresponding to the multiple virtual materials of the textured mesh (Cowburn, Paragraph[0024], "the semantic texture map generated by the semantic texture map system may be applied to one or more objects … each texel comprises one or more semantic feature labels <read on texture IDs>, and 2D texture coordinates"; [0060], “the semantic labeling module 308 assigns a material parameter to a feature class that corresponds with a semantic label from among the set of semantic label… The material parameters may be defined based on a set of interaction variables”).
Regarding Claim 6, the combination of Cowburn, Feng and Yang teaches the invention in Claim 1.
The combination further teaches wherein the marker is a scribble drawn over the textured mesh in the user interface (Feng, Page 5, Section 3.2.4, "user inputs typically can take one of the following forms: 1) foreground / background clicks: this gives the least amount of user inputs; 2) foreground/background strokes").
Feng and the combination of Cowburn and Chen are analogous in that they each address interactive, pixel- or vertex-level labeling based on local feature relationships. Feng teaches that user inputs can take forms such as foreground/background clicks and scribbles over image regions to drive the segmentation, while the modified Cowburn-and-Chen system uses
K-nearest-neighbors matting over 3D/appearance features to assign material labels. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Feng's scribble-based marker input into the modified Cowburn and Chen system so that
a user can indicate regions of the textured mesh corresponding to material types by drawing scribbles in the UI, thereby leveraging known interactive scribble paradigms to guide KNN-based material matting on 3D textured meshes.
Regarding Claim 7, the combination of Cowburn, Feng and Yang teaches the invention in Claim 1.
The combination further teaches wherein the material map visually designates different material types of the multiple virtual materials (Cowburn, Paragraph [0025], "semantic feature labels of the texels of the texture map may provide classification information for a feature"; Paragraph [0020], "Semantic features may for example include...material parameters" <read on multiple virtual materials>).
Regarding Claim 8, the combination of Cowburn, Feng and Yang teaches the invention in Claim 7.
The combination further teaches wherein the different material types are identified by different colors on the material map (Cowburn, Paragraph[0020], "material parameters that define properties of a surface or object and which may include a roughness value, a metallic value, a specular value, and a base color value" <read on different colors>).
Regarding Claim 11, it recites limitations similar in scope to the limitations of Claim 3 and therefore is rejected under the same rationale.
Regarding Claim 12, it recites limitations similar in scope to the limitations of Claim 4 and therefore is rejected under the same rationale.
Regarding Claim 14, it recites limitations similar in scope to the limitations of Claim 6 and therefore is rejected under the same rationale.
Regarding Claim 15, it recites limitations similar in scope to the limitations of Claim 7 and therefore is rejected under the same rationale.
Regarding Claim 11, it recites limitations similar in scope to the limitations of claim 1 and the combination of Cowburn, Feng and Yang teaches all the limitations as of Claim 1. And Cowburn discloses these features can be implemented on a computer readable storage medium (Cowburn , Paragraph [0015], FIG. 10 is a block diagram illustrating components of a machine, according to some example embodiments, able to read instructions from a machine-readable medium (e.g., a machine-readable storage medium) and perform any one or more of the methodologies discussed herein.).
Regarding Claim 18, it recites limitations similar in scope to the limitations of Claim 3 and therefore is rejected under the same rationale.
Regarding Claim 19, it recites limitations similar in scope to the limitations of Claim 4 and therefore is rejected under the same rationale.
Regarding Claim 21, the combination of Cowburn, Feng, and Yang teaches the invention of Claim 1.
The combination further teaches wherein the one or more three-dimensional features are determined based on geometry of one or more surfaces of the object (Cowburn, Paragraph [0017], "generating a 3D model, such as a 3D mesh model, based on the set of image features of the image, the 3D model comprising a plurality of vertices" <read on three-dimensional features determined based on geometry of one or more surfaces of the object, as the vertices of the 3D mesh model define the surface geometry of the modeled object>; Cowburn, Paragraph [0024], "based on the coordinates of the vertices" <read on determined based on geometry of surfaces>; Feng, Page 3, Section 3.2, "normal vectors are computed from a depth-projected 3D point cloud" <read on three-dimensional features determined based on geometry of one or more surfaces, as surface normal vectors are by definition derived from the local geometry of the surface at each point>; Feng, Page 3, Section 3.2, "to be able to distinguish objects with different geometry but similar distance, normal vectors are computed from a depth-projected 3D point cloud" <read on determined based on geometry of surfaces of the object>).
Feng and Cowburn are analogous since both are directed to characterizing regions of a three-dimensional object surface for the purpose of segmentation/labeling by material or semantic class. Cowburn provided a way to constructs a 3D mesh model made up of vertices whose coordinates define the geometry of the object's surfaces and uses those vertex coordinates to project semantic information onto the surface. Feng provided a way by deriving per-pixel three-dimensional features (surface normals) from a depth-projected 3D point cloud in order to distinguish regions of differing surface geometry, and Feng expressly ties this computation to "objects with different geometry." Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate surface-normal features from the object's surface geometry taught by Feng into modified invention of Cowburn such that geometry-derived per-pixel features improve the ability of the segmentation algorithm to discriminate regions that differ in surface shape. The motivation is to improve segmentation accuracy on textured 3D surfaces where different material regions coincide with different local surface geometry, as taught by Feng in Section 3.2 ("to be able to distinguish objects with different geometry").
Claims 5, 9, 13, 20 are rejected under 35 U.S.C. § 103 as being unpatentable over Cowburn et al. (US 20200312008 A1) in view of Feng et al. (Interactive Segmentation on RGB-0 Images via Cue Selection, CVPR 2016, hereinafter Feng), further in view of Yang et al. (US 20160217552 A1, hereinafter Yang) as applied to Claim 1, 10, 17 above respectively and further in view of Chen, Q., Li, D., & Tang, C.-K. (2012). KNN Matting. 2012 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 869-876, hereinafter Chen).
Regarding Claim 5, the combination of Cowburn, Feng and Yang teaches the invention in Claim 1.
The combination further teaches normal properties (Feng, Section 3.2, "to be able to distinguish objects with different geometry but similar distance, normal vectors are computed from a depth-projected 3D point cloud <read on: normal properties>"; Feng, Section 3.2, "we use three different cues to infer foreground confidence for each pixel ... normal"; it is noted since normals are a per-pixel property used by the segmentation algorithm to characterize each pixel on the surface of a 3D scene).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Feng's teaching of normal properties as a per-pixel feature into the material segmentation system of Cowburn, because Feng demonstrates that using surface normals as a pixel-level cue improves segmentation accuracy on 3D surfaces, and Cowburn's semantic texture map is applied to the surface of a 3D mesh where normals are well-defined geometric properties of each texel's surface position. The motivation is to improve the accuracy of material-type boundary identification on the 3D mesh surface, as taught by Feng in Section 4.5.
The combination does not explicitly disclose but Chen teaches the algorithm is a k-nearest neighbors matting algorithm (Chen, Abstract, "our matting technique, aptly called KNN matting, capitalizes on the nonlocal principle by using K nearest neighbors (KNN) in matching nonlocal neighborhoods" <read on: k-nearest neighbors matting algorithm>); that determines three-dimensional positions (Chen, Section 4.2, "a feature vector X(i) at a given pixel i that includes spatial coordinates to reinforce spatial coherence can be defined as X(i) = (cos(h), sin(h), s, v, x, y)i" <read on: three-dimensional positions> "where ... (x, y) are the spatial coordinates of pixel i); color values (Chen, Section 4.2, "X(i) = (cos(h), sin(h), s, v, x, y)i" <read on: color values> "where h, s, v are the respective HSV coordinates" <read on: color values, as HSV coordinates are color values of the pixel>; it is noted HSV coordinates represents brightness or intensity of the color).
Chen and Cowburn are analogous since both of them are dealing with pixel-level feature-based identification and extraction of distinct material regions from a surface, where each pixel is characterized by a set of per-pixel properties used to determine its material layer membership. Cowburn provides a way of assigning semantic material labels to texels of a UV-mapped 3D mesh based on per-pixel semantic feature properties of the surface. Chen provides a way of identifying and extracting distinct material layers by operating a k-nearest neighbors matting algorithm on per-pixel feature vectors that capture the color values and spatial coordinates of each pixel. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention was made to incorporate the k-nearest neighbors matting algorithm that determines three-dimensional positions and color values for the pixels taught by Chen into the modified invention of Cowburn such that the algorithm used to identify the additional region of the textured mesh corresponding to a material type is a k-nearest neighbors matting algorithm that determines three-dimensional positions and color values for the pixels. The motivation is to improve the accuracy and efficiency of material-layer identification on a textured mesh using sparse user input, as discussed by Chen in the Abstract
Regarding Claim 9, the combination of Cowburn, Feng, and Chen teaches the invention in Claim 7.
The combination does not explicitly disclose identifying the boundary using a k-nearest neighbors matting algorithm.
However, Chen teaches identifying the boundary using a k-nearest neighbors matting algorithm (Chen, Abstract, "our matting technique, aptly called KNN matting, capitalizes on the nonlocal principle by using K nearest neighbors (KNN) in matching nonlocal neighborhoods";Figure 1 caption, "KNN matting produces better results. Top row: clearer and cleaner boundary" ⟨read on identifying the boundary using a k-nearest neighbors matting algorithm, as the boundary between material layers is identified as a result of the KNN matting process⟩; Chen, Section 5.1, "five overlapping material mattes are produced; despite that the matte for 'blue paper' has several disconnected components one click is all it takes for matting the material").
Chen and Cowburn are analogous since both deal with pixel-level material segmentation and extraction of distinct material regions with identified boundaries. Cowburn provides a way of segmenting regions of a textured mesh by material type using pixel-level feature analysis. Chen provides a way of identifying the boundaries between different material regions using a k-nearest neighbors matting algorithm operating on per-pixel feature vectors. Therefore, it would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate the boundary-identifying KNN matting algorithm taught by Chen into the modified system of Cowburn, such that the boundary between material types on the textured mesh is identified using a k-nearest neighbors matting algorithm. The motivation is to achieve cleaner and more accurate boundary identification between material regions with minimal user interaction, as discussed by Chen in Figure 1.
Regarding Claim 13, it recites limitations similar in scope to the limitations of Claim 5 and therefore is rejected under the same rationale.
Regarding Claim 20, it recites limitations similar in scope to the limitations of Claim 5 and therefore is rejected under the same rationale.
Response to Arguments
Applicant’s arguments with respect to claim 1, 10, 17, filed on 5/29/2026, with respect to rejection under 35 USC § 103 have been fully considered but are not persuasive.
Regarding Claim 1, applicant asserts that the prior art does not teach the limitation "determining one or more three-dimensional features of the object represented by the textured mesh" and "identifying an additional region of the textured mesh corresponding to the material type using an algorithm by comparing positions of pixels on the one or more three-dimensional features of the object represented by the textured mesh." In response to the argument, as described in the rejection of Claim 1 above, this limitation is taught by the combination of Cowburn, Feng, and Yang. In particular, Cowburn in Paragraph [0017] expressly teaches "generating a 3D model, such as a 3D mesh model, based on the set of image features of the image, the 3D model comprising a plurality of vertices," and Paragraph [0024] teaches that the semantic segmentation image is projected upon the 3D model "based on the coordinates of the vertices." The vertices of Cowburn's 3D mesh constitute three-dimensional features of the object represented by the textured mesh, and the projection of the semantic segmentation onto the mesh using vertex coordinates constitutes comparing positions of pixels on those three-dimensional features. Applicant's argument that Cowburn "merely retrieves an already-segmented image" mischaracterizes the scope of Cowburn's disclosure — Cowburn actively generates the 3D mesh model and actively projects segmentation labels onto the mesh using the mesh's three-dimensional vertex coordinates, which is the very act of comparing pixel positions on the three-dimensional features of the object. Furthermore, Feng, Page 3, Section 3.2, teaches that "normal vectors are computed from a depth-projected 3D point cloud," which expressly determines a three-dimensional feature (surface normals) of the object surface and uses it as a per-pixel cue in the segmentation algorithm. The claim does not require the "determining" step to be performed by any particular mechanism, nor does it exclude the determination being performed as part of the mesh generation itself. Under the broadest reasonable interpretation, both Cowburn's mesh vertex generation and Feng's normal-vector computation from the 3D point cloud read on "determining one or more three-dimensional features of the object." Hence the combination of prior arts fully anticipates the limitation. Therefore, applicant's remark cannot be considered persuasive.
Regarding Claim 1, applicant further asserts that the prior art does not teach "presenting, by the processing device, the material map that includes a selectable portion corresponding to the material type and identified by the boundary in a user interface." In response to the argument, as described in the rejection of Claim 1 above, this limitation is taught by the combination of Cowburn, Feng, and Yang. In particular, Cowburn in Paragraph [0024] teaches that "the resulting semantic texture map therefor comprises a set of texels, where each texel comprises one or more semantic feature labels," and Paragraph [0060] teaches that "the semantic labeling module 308 assigns a material parameter to a feature class ... The material parameters may be defined based on a set of interaction variables." A discretely labeled region of the semantic texture map, bounded by transitions between differently labeled texels, is a portion of the map corresponding to a material type that is capable of being selected — which is all the claim language "selectable portion" requires under the broadest reasonable interpretation. The claim does not recite any particular selection mechanism, nor does it require any specific user-interface interaction with the portion beyond the map being presented in a user interface. Yang further teaches at Paragraph [0004] that "A gain map with boundaries and smoothing is generated using localized pixels based on information from one or more of the label map, the texture map, and the directional edge map," expressly generating a map whose classified regions are delimited by identified boundaries. Feng at Page 5, Section 3.2.4 teaches user inputs including "foreground / background clicks" and "foreground/background strokes" in the user interface, evidencing that maps generated in this class of systems are presented in a user interface amenable to selection. The applicant's argument that "a categorization of a background or a foreground, as described by Feng, does not involve presenting a material map that includes a selectable portion" attacks Feng individually, whereas the rejection relies on the combined teachings of Cowburn (semantic texture map with per-texel material labels), Yang (map with regions delimited by identified boundaries), and Feng (interactive user-interface presentation). Hence the combination of prior arts fully anticipates the limitation. Therefore, applicant's remark cannot be considered persuasive.
In regard to Claims 2-4, 6-8, 11-12, 14-15, 18-19, they directly/indirectly depends on independent Claim 1, 10, 17 respectively. Applicant does not argue anything other than the independent claim 1, 10, 17. The limitations in those claims in conjunction with combination previously established as explained.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
US 11790610 B2 Systems and methods for selective image compositing
US 20140160114 A1 Transformation of a Three-Dimensional Flow Image
US 20040242987 A1 Methods of predicting musculoskeletal disease
Any inquiry concerning this communication or earlier communications from the examiner should be directed to YUJANG TSWEI whose telephone number is (571)272-6669. The examiner can normally be reached 8:30am-5:30pm EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Kent Chang can be reached at (571)272-7667. 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.
/YuJang Tswei/Primary Examiner, Art Unit 2614
Prosecution Insights