Object-centric video redaction

1. A computer-implemented method for video redaction, the method comprising:

receiving a video for redaction having a plurality of frames;

analyzing the video to generate an appearance model for the video;

providing a user interface (UI) allowing a user to modify the appearance model with a plurality of user-placed marks;

receiving the plurality of user-placed marks corresponding with a plurality of sample frames selected from the plurality of frames;

interpolating a bounding box on the plurality of sample frames based on the received plurality of user-placed marks;

responsive to a user selecting an object from the appearance model based on the interpolated bounding box, extending and completing a trajectory of the selected object within a plurality of remainder frames from the plurality of frames, and placing automatically generated marks to enclose the selected object within each remainder frame of the plurality of remainder frames based on the received plurality of user-placed marks and the extended and completed trajectory of the selected object; and

redacting the plurality of frames of the video based on the modified appearance model.

2. The method of claim 1 , further comprising:

designating weightings for trajectory observations in the appearance model.

3. The method of claim 2 , further comprising:

designating a higher weighting for the trajectory observations of the selected object than the weighting for observations that have not been selected.

4. The method of claim 1 , further comprising:

receiving user input selecting a boundary for identify an object in the appearance model.

5. The method of claim 1 , further comprising:

receiving further user input and making further modifications to the appearance model iteratively.

6. The method of claim 1 , wherein the video is a video used for law enforcement.

7. A computer system comprising:

one or more processors, one or more computer-readable memories, and one or more computer-readable, tangible storage devices;

program instructions, stored on at least one of the one or more storage devices for execution by at least one of the one or more processors via at least one of the one or more memories to perform operations comprising:

receiving a video for redaction having a plurality of frames;

analyzing the video to generate an appearance model for the video;

providing a user interface (UI) allowing a user to modify the appearance model with a plurality of user-placed marks;

receiving the plurality of user-placed marks corresponding with a plurality of sample frames selected from the plurality of frames;

interpolating a bounding box on the plurality of sample frames based on the received plurality of user-placed marks;

responsive to a user selecting an object from the appearance model based on the interpolated bounding box, extending and completing a trajectory of the selected object within a plurality of remainder frames from the plurality of frames, and placing automatically generated marks to enclose the selected object within each remainder frame of the plurality of remainder frames based on the received plurality of user-placed marks and the extended and completed trajectory of the selected object; and

redacting the plurality of frames of the video based on the modified appearance model.

8. The system of claim 7 , wherein the operations further comprise:

designating weightings for trajectory observations in the appearance model.

9. The system of claim 8 , wherein the operations further comprise:

designating a higher weighting for the trajectory observations of the selected object than the weighting for observations that have not been selected.

10. The system of claim 7 , wherein the operations further comprise:

receiving user input selecting a boundary for identify an object in the appearance model.

11. The system of claim 7 , wherein the operations further comprise:

receiving further user input and making further modifications to the appearance model iteratively.

12. A computer program product comprising a non-transitory computer-readable storage medium having program code embodied therewith, the program code executable by a computing device to perform operations comprising:

receiving a video for redaction having a plurality of frames;

analyzing the video to generate an appearance model for the video;

providing a user interface (UI) allowing a user to modify the appearance model with a plurality of user-placed marks;

receiving the plurality of user-placed marks corresponding with a plurality of sample frames selected from the plurality of frames;

interpolating a bounding box on the plurality of sample frames based on the received plurality of user-placed marks;

responsive to a user selecting an object from the appearance model based on the interpolated bounding box, extending and completing a trajectory of the selected object within a plurality of remainder frames from the plurality of frames, and placing automatically generated marks to enclose the selected object within each remainder frame of the plurality of remainder frames based on the received plurality of user-placed marks and the extended and completed trajectory of the selected object; and

redacting the plurality of frames of the video based on the modified appearance model.

13. The program product of claim 12 , wherein the operations further comprise:

designating weightings for trajectory observations in the appearance model.

14. The program product of claim 13 , wherein the operations further comprise:

designating a higher weighting for the trajectory observations of the selected object than the weighting for observations that have not been selected.

15. The program product of claim 12 , wherein the operations further comprise:

receiving user input selecting a boundary for identify an object in the appearance model.

16. The program product of claim 12 , wherein the operations further comprise:

receiving further user input and making further modifications to the appearance model iteratively.

17. The program product of claim 12 , wherein the video is a video used for law enforcement.