# US 2025/0095157 A1 · System and method for field line reconstruction within video of american football > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 2025/0095157 A1 - Document ID: US20250095157A1 - Status: Published application - Kind: A1 - Application number: 18/369,944 - Filed: 2023-09-19 - Published: 2025-03-20 - Earliest claimed: 2023-09-19 - Assignees: Jered Donald Aasheim - Inventors: Jered Donald Aasheim; Ethan John Aasheim - Web page: https://patentgrep.com/patent/US20250095157A1 ## CPC - G06T 7/13: Edge detection - G06V 10/25: Determination of region of interest [ROI] or a volume of interest [VOI] - G06V 10/762: using clustering, e.g. of similar faces in social networks - G06V 10/764: using classification, e.g. of video objects - G06V 10/82: using neural networks - G06V 20/42: {of sport video content} - G06V 20/70: Labelling scene content, e.g. deriving syntactic or semantic representations - G06T 2207/20081: Training; Learning - G06T 2207/20084: Artificial neural networks [ANN] - G06V 2201/07: Target detection - G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract Systems and methods for reconstructing field lines within video may include one or more processor configured by machine-readable instructions. Exemplary implementations may include identifying a plurality of candidate field lines in one or more frame of video; training a neural network to detect one or more field object; detecting, using the neural network, one or more field object in the one or more fame of video; assigning, using the detected one or more field object, a confidence value to each of the plurality of candidate field lines; removing one or more low confidence candidate field line from the plurality of candidate field lines; and retaining one or more high confidence candidate field line from the plurality of candidate field lines. ## Claims ### Claim 1 A system for reconstructing field lines within video, comprising: one or more processor configured by machine-readable instructions for: identifying a plurality of candidate field lines in one or more frame of video; training a neural network to detect one or more field object; detecting, using the neural network, one or more field object in the one or more fame of video; assigning, using the detected one or more field object, a confidence value to each of the plurality of candidate field lines; removing one or more low confidence candidate field line from the plurality of candidate field lines; and retaining one or more high confidence candidate field line from the plurality of candidate field lines. ### Claim 2 (depends on 1) The system of claim 1, wherein identifying a plurality of candidate field lines comprises: applying a clustering algorithm to group similar line segments into a cluster; and iterating through the cluster to identify a single candidate field line. ### Claim 3 (depends on 1) The system of claim 1, wherein the one or more field object includes one or more hashmark and field number. ### Claim 4 (depends on 1) The system of claim 1, wherein the one or more processor is further configured by machine-readable instructions for: recalibrating the alignment of the one or more high confidence candidate field line to intercept the middle of the detected one or more field object. ### Claim 5 (depends on 1) The system of claim 1, wherein the one or more processor is further configured by machine-readable instructions for: constructing one or more orthogonal hashmark field line and sideline. ### Claim 6 (depends on 1) The system of claim 1, wherein the one or more processor is further configured by machine-readable instructions for: assigning the one or more high confidence candidate field line a value of an intercepted field number. ### Claim 7 (depends on 1) The system of claim 1, wherein the one or more processor is further configured by machine-readable instructions for: identifying and replacing one or more missing field line. ### Claim 8 (depends on 1) The system of claim 1, wherein the one or more processor is further configured by machine-readable instructions for: overlaying the one or more high confidence candidate field line on the one or more frame of video. ### Claim 9 (depends on 1) The system of claim 1, wherein the identifying a plurality of candidate field lines utilizes a Canny edge or Hough line algorithm. ### Claim 10 (depends on 1) The system of claim 1, wherein the one or more processor is further configured by machine-readable instructions for: creating a field model including the one or more high confidence candidate field line, one or more hashmark, one or more field number, and one or more sideline. ### Claim 11 (depends on 2) The system of claim 2, wherein the clustering algorithm groups line segments having similar axis intercepts. ### Claim 12 (depends on 1) The system of claim 1, wherein training a neural network includes annotating field markings or field objects on samples of video. ### Claim 13 (depends on 1) The system of claim 1, wherein the one or more processor is further configured by machine-readable instructions for: overlaying the one or more high confidence candidate field line along with bounding boxes surrounding the detected one or more field object. ### Claim 14 (depends on 1) The system of claim 1, wherein assigning a confidence value includes determining whether each of the plurality of candidate field lines intercepts one or more hashmark. ### Claim 15 (depends on 1) The system of claim 1, wherein assigning a confidence value includes determining whether each of the plurality of candidate field lines intercepts a field number. ### Claim 16 A method for reconstructing field lines within video, comprising: obtaining sets of football video information, wherein individual ones of the sets of football video information reflect gameplay; training a machine-learning model with the obtained sets of football video information such that the machine-learning model predicts the locations of one or more hashmark and field number; and storing the trained machine-learning model. ### Claim 17 (depends on 16) The method of claim 16, further comprising: overlaying one or more bounding box for one or more hashmark and field number. ### Claim 18 (depends on 17) The method of claim 17, further comprising: assigning a confidence value to each of a plurality of candidate field lines, wherein a hashmark intercept or a field number intercept increases the confidence value. ### Claim 19 (depends on 18) The method of claim 18, further comprising: reconciling the alignment of one or more candidate field line having a high confidence value to intercept one or more hashmark or field number. ### Claim 20 (depends on 18) The method of claim 18, further comprising: assigning each of the plurality of candidate field lines a value associated with an intercepted field number. ### Claim 21 (depends on 16) The method of claim 16, further comprising: identifying and replacing missing field lines. ### Claim 22 (depends on 15) The method of claim 15, further comprising: assigning location names to the candidate field lines based upon intercepting identified field numbers and the geometrical definitions of the field of play.