Restoration of a kinetic event using video

{
  "publication_number": "US12094041B2",
  "country": "US",
  "kind": "B2",
  "title": "Restoration of a kinetic event using video",
  "abstract": "A method for restoration of a kinetic event using video. The method includes obtaining video involving the kinetic event including a plurality of frames. An absolute location of a frame of the plurality of frames is determined. Movement of a feature point cloud corresponding to an object in motion in the plurality of frames is analyzed. A 3D object model with substantially similar dimensions to the object in motion is selected. The 3D object model is displayed emulating the analysed movement on the 3D map based on the absolute location.",
  "claims": [
    "1. A computer system for restoration of a kinetic event using video, the system comprising: one or more computer processors, one or more non-transitory computer-readable storage media, and program instructions stored on the one or more non-transitory computer-readable storage media for execution by at least one of the one or more processors capable of performing a method, the method comprising: obtaining video involving the kinetic event, wherein the video includes a plurality of frames; determining an absolute location of a frame of the plurality of frames; analyzing movement of a feature point cloud corresponding to an object in motion in the plurality of frames compared to another object in motion at a known speed; selecting a 3D object model with substantially similar dimensions to the object in motion; and displaying the 3D object model emulating the analysed movement on a 3D map based on the absolute location.",
    "2. The computer system of claim 1, wherein the determining the absolute location based on the frame in the video is based on matching a frame of the video to an image in the 3D map, wherein the kinetic event is a car accident, wherein the object in motion is a car, and wherein the 3D object model corresponds to a make and model of the car.",
    "3. The computer system of claim 1, further comprising: generating a plurality of feature point clouds for the plurality of frames in the video; wherein the analysing the movement of the feature point cloud for the object in motion is performed relative to another feature point cloud.",
    "4. The computer system of claim 1, wherein the 3D object model exhibiting the analysed movement on the 3D map is displayed from a different perspective of the absolute location than the perspective of the matched frame used to determine the absolute location.",
    "5. The method of claim 1, wherein the video is plural, wherein each video includes a different object in motion and corresponding 3D object model, and wherein the displaying of the 3D object models exhibiting the analysed movements on the 3D map based on the absolute location are synced by time.",
    "6. The method of claim 1, wherein the displaying the 3D object model exhibiting the analysed movement on the 3D map based on the absolute location includes annotations describing features of the analysed movement and a collision.",
    "7. A computer program product for restoration of a kinetic event using video, the computer program product comprising: one or more non-transitory computer-readable storage media and program instructions stored on the one or more non-transitory computer-readable storage media capable of performing a method, the method comprising: obtaining video involving the kinetic event, wherein the video includes a plurality of frames; determining an absolute location of a frame of the plurality of frames; analyzing movement of a feature point cloud corresponding to an object in motion in the plurality of frames compared to another object in motion at a known speed; selecting a 3D object model with substantially similar dimensions to the object in motion; and displaying the 3D object model emulating the analysed movement on a 3D map based on the absolute location.",
    "8. The computer program product of claim 7, wherein the determining the absolute location based on the frame in the video is based on matching a frame of the video to an image in the 3D map, wherein the kinetic event is a car accident, wherein the object in motion is a car, and wherein the 3D object model corresponds to a make and model of the car.",
    "9. The computer program product of claim 7, further comprising: generating a plurality of feature point clouds for the plurality of frames in the video; wherein the analysing the movement of the feature point cloud for the object in motion is performed relative to another feature point cloud.",
    "10. The method of claim 7, wherein the 3D object model exhibiting the analysed movement on the 3D map is displayed from a different perspective of the absolute location than the perspective of the matched frame used to determine the absolute location.",
    "11. The method of claim 7, wherein the video is plural, wherein each video includes a different object in motion and corresponding 3D object model, and wherein the displaying of the 3D object models exhibiting the analysed movements on the 3D map based on the absolute location are synced by time.",
    "12. The method of claim 7, wherein the displaying the 3D object model exhibiting the analysed movement on the 3D map based on the absolute location includes annotations describing features of the analysed movement and a collision.",
    "13. The method of claim 12, wherein the annotations include indications relating to deviations from analysed traffic regulations.",
    "14. A method for restoration of a kinetic event using video, the method comprising: obtaining video involving the kinetic event, wherein the video includes a plurality of frames; determining an absolute location of a frame of the plurality of frames; analyzing movement of a feature point cloud corresponding to an object in motion in the plurality of frames compared to another object in motion at a known speed; selecting a 3D object model with substantially similar dimensions to the object in motion; and displaying the 3D object model emulating the analysed movement on a 3D map based on the absolute location.",
    "15. The method of claim 1, wherein the determining the absolute location based on the frame in the video is based on matching a frame of the video to an image in the 3D map, wherein the kinetic event is a car accident, wherein the object in motion is a car, and wherein the 3D object model corresponds to a make and model of the car.",
    "16. The method of claim 1, further comprising: generating a plurality of feature point clouds for the plurality of frames in the video; wherein the analysing the movement of the feature point cloud for the object in motion is performed relative to another feature point cloud.",
    "17. The method of claim 1, wherein the 3D object model exhibiting the analysed movement on the 3D map is displayed from a different perspective of the absolute location than the perspective of the matched frame used to determine the absolute location.",
    "18. The method of claim 1, wherein the video is plural, wherein each video includes a different object in motion and corresponding 3D object model, and wherein the displaying of the 3D object models exhibiting the analysed movements on the 3D map based on the absolute location are synced by time.",
    "19. The method of claim 1, wherein the displaying the 3D object model exhibiting the analysed movement on the 3D map based on the absolute location includes annotations describing features of the analysed movement and a collision.",
    "20. The method of claim 19, wherein the annotations include indications relating to deviations from analysed traffic regulations."
  ],
  "description_excerpt": "Exemplary embodiments of the present inventive concept relate to visualization of a kinetic event, and more particularly, to restoration of a kinetic event using video.\n\nWhen a kinetic event occurs, such as a car accident, involved parties (e.g., drivers, pedestrians, passengers, etc.) often offer differing accounts of the events that precipitated the kinetic event. Consequently, in the case of a car accident, video is manually reviewed by various interested parties. However, different video source perspectives (e.g., car 1, car 2, pedestrian, passenger, traffic camera, etc.) may lead to inconsistent inferences of fault among the involved parties. However, at present, video from different sources cannot be consolidated into a comprehensive visualization for evaluation. In some instances, there is also no traffic camera available at the scene of the car accident, hindering an aerial visualization of the car accident. Animation re-enactments of car accidents can be produced, but these animations rely on replication by generalization rather than restoration. Furthermore, they are often self-serving because they are commissioned by involved parties based on the video or rendering that is most consistent with their account. Thus, car accident animations are often biased and imprecise.\n\nExemplary embodiments of the present inventive concept relate to a method, a computer program product, and a system for kinetic event restoration using video.\n\nAccording to an exemplary embodiment of the present inventive concept, a method is provided for kinetic event restoration using video.",
  "cpc": [
    "G06T 13/20",
    "G06Q 10/10",
    "G06Q 40/08",
    "G06Q 50/18",
    "G06Q 50/26",
    "G06Q 50/40",
    "G06T 19/00",
    "G06T 2207/10016",
    "G06T 2207/10028",
    "G06T 2219/004",
    "G06T 7/248"
  ],
  "ipc": [
    "G06T 13/20",
    "G06T 19/00",
    "G06T 7/246"
  ],
  "assignees": [
    "International Business Machines Corp"
  ],
  "inventors": [
    "Nuo XU",
    "Yuan Yuan Ding",
    "Ke Yong Zhang",
    "Tian Tian Chai",
    "Yi Chen Zhong",
    "Hong Bing Zhang"
  ],
  "filing_date": "2022-07-26",
  "publication_date": "2024-09-17",
  "grant_date": "2024-09-17",
  "priority_date": "2022-07-26",
  "application_number": "US-202217814938-A",
  "family_id": "89664584",
  "cited_by_count": 2,
  "citations": [
    "US20150029308A1",
    "US9098753B1",
    "US9607226B2",
    "US11068995B1",
    "TWI630132B",
    "CN105046731A",
    "WO2018147329A1",
    "US10884409B2",
    "US20180365888A1",
    "US10417816B2",
    "US10528851B2",
    "US11210859B1",
    "US11815623B2",
    "US20220044024A1",
    "CN112132993A"
  ]
}