# US 2025/0095200 A1 · Semantic visual feature sharing > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2025/0095200 A1 - Document ID: US20250095200A1 - Status: Published application - Kind: A1 - Application number: 18/467,526 - Filed: 2023-09-14 - Published: 2025-03-20 - Earliest claimed: 2023-09-14 - Assignees: QUALCOMM Incorporated - Applicants: QUALCOMM Incorporated - Inventors: Anantharaman Balasubramanian; Stelios Stefanatos; Kapil Gulati - Web page: https://patentgrep.com/patent/US20250095200A1 ## CPC - G06T 7/74: {involving reference images or patches} - G06V 20/56: exterior to a vehicle by using sensors mounted on the vehicle - G06V 10/751: {Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching} - G06V 20/58: Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads - G06V 20/70: Labelling scene content, e.g. deriving syntactic or semantic representations - G06T 2207/10016: Video; Image sequence - G06T 2207/30241: Trajectory - G06T 2207/30252: Vehicle exterior; Vicinity of vehicle - G06V 10/462: {Salient features, e.g. scale invariant feature transforms [SIFT]} - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract An example device for processing image data includes a processing unit configured to determine a set of criteria representing keypoints of interest for processed image data; extract a plurality of keypoints from the image; determine which of the plurality of keypoints match the set of criteria representing the keypoints of interest to determine a matched set of keypoints; and output data representative of the matched set of keypoints. ## Claims ### Claim 1 A method of processing image data, the method comprising: determining, by processing circuitry of an image analysis unit, a set of criteria representing keypoints of interest for processed image data; extracting, by the processing circuitry, a plurality of keypoints from the image; determining, by the processing circuitry, which of the plurality of keypoints match the set of criteria representing the keypoints of interest to determine a matching set of keypoints; and outputting, by the processing circuitry, data representative of the matching set of keypoints. ### Claim 2 (depends on 1) The method of claim 1, wherein outputting the data representative of the matching set of keypoints comprises outputting, for each of the matching set of keypoints, a keypoint location and a descriptor for the keypoint at the keypoint location. ### Claim 3 (depends on 1) The method of claim 1, wherein determining the set of criteria representing the keypoints of interest comprises: receiving one or more identifiers; and determining a mapping for each of the identifiers to one of the keypoints of interest. ### Claim 4 (depends on 1) The method of claim 1, wherein determining the set of criteria representing the keypoints of interest comprises: receiving data representing a set of one or more objects of interest; and for each of the objects in the set of one or more objects of interest, receiving data representing one or more keypoints of interest for the corresponding object. ### Claim 5 (depends on 4) The method of claim 4, further comprising defining a first level of hierarchy with respect to the corresponding object, and defining a second level of hierarchy with respect to one or more features of the corresponding object. ### Claim 6 (depends on 4) The method of claim 4, further comprising defining a first level of hierarchy with respect to the corresponding object, and defining a second level of hierarchy with respect to a finer identity of the corresponding object. ### Claim 7 (depends on 4) The method of claim 4, further comprising associating each of the one or more keypoints of interest with a corresponding semantic configuration identifier specifying a corresponding feature category. ### Claim 8 (depends on 4) The method of claim 4, further comprising performing a coarse matching procedure for matching the one or more keypoints of interest with the corresponding object, and subsequently performing a fine matching procedure for matching the one or more keypoints of interest with the corresponding object. ### Claim 9 (depends on 4) The method of claim 4, wherein determining which of the keypoints match the set of criteria representing the keypoints of interest comprises: identifying objects in the image; for each of the identified objects, determining whether the object matches one of the set of one or more objects of interests to determine matching objects; and for each of the matching objects, determining whether keypoints for the matching object match one of the one or more keypoints of interest to determine the matching set of keypoints. ### Claim 10 (depends on 1) The method of claim 1, wherein determining the set of criteria representing the keypoints of interest comprises: receiving data representing a set of one or more objects of interest; and for each of the objects in the set of objects of interest, receiving data representing attributes of interest for the corresponding object; and for each of the objects in the set of objects having the representative attributes, receiving data representing one or more keypoints of interest for the corresponding object. ### Claim 11 (depends on 10) The method of claim 10, wherein determining which of the keypoints match the set of criteria representing the keypoints of interest comprises: identifying objects in the image; for each of the identified objects, determining whether the object has the attributes of interest to determine matching objects; and for each of the matching objects, determining whether keypoints for the matching object match one of the one or more keypoints of interest to determine the matching set of keypoints. ### Claim 12 (depends on 1) The method of claim 1, wherein the image analysis unit is included in an ego vehicle, and wherein outputting the data representative of the matching set of keypoints comprises outputting the data representative of the matching set of keypoints to a target vehicle. ### Claim 13 (depends on 1) The method of claim 1, wherein the image analysis unit is included in an ego vehicle, and wherein determining the set of criteria representing keypoints of interest comprises receiving the set of criteria representing the keypoints of interest from a target vehicle. ### Claim 14 A device for processing image data, wherein the device includes processing circuitry configured to: determine a set of criteria representing keypoints of interest for processed image data; extract a plurality of keypoints from the image; determine which of the plurality of keypoints match the set of criteria representing the keypoints of interest to determine a matched set of keypoints; and output data representative of the matched set of keypoints. ### Claim 15 (depends on 14) The device of claim 14, wherein output the data comprises output, for each of the matched set of keypoints, a keypoint location and a descriptor for the keypoint at the keypoint location. ### Claim 16 (depends on 14) The device of claim 14, wherein determine the set of criteria representing the keypoints of interest comprises: receive one or more identifiers; and determine a mapping for each of the identifiers to one of the keypoints of interest. ### Claim 17 (depends on 14) The device of claim 14, wherein determine the set of criteria representing the keypoints of interest comprises: receive data representing a set of one or more objects of interest; and for each of the objects in the set of one or more objects of interest, receive data representing one or more keypoints of interest for the corresponding object. ### Claim 18 (depends on 17) The device of claim 17, wherein the device includes processing circuitry configured to define a first level of hierarchy with respect to the corresponding object, and to define a second level of hierarchy with respect to one or more features of the corresponding object. ### Claim 19 (depends on 17) The device of claim 17, wherein the device includes processing circuitry configured to define a first level of hierarchy with respect to the corresponding object, and to define a second level of hierarchy with respect to a finer identity of the corresponding object. ### Claim 20 (depends on 17) The device of claim 17, wherein the device includes processing circuitry configured to associate each of the one or more keypoints of interest with a corresponding semantic configuration identifier specifying a corresponding feature category. ### Claim 21 (depends on 17) The device of claim 17, wherein the device includes processing circuitry configured to perform a coarse matching procedure for matching the one or more keypoints of interest with the corresponding object, and to subsequently perform a fine matching procedure for matching the one or more keypoints of interest with the corresponding object. ### Claim 22 A device for processing image data, the device comprising: means for determining a set of criteria representing keypoints of interest for processed image data; means for extracting a plurality of keypoints from the image; means for determining which of the plurality of keypoints match the set of criteria representing the keypoints of interest to determine a matching set of keypoints; and means for outputting data representative of the matching set of keypoints.