System and method for dynamic redaction
Techniques for dynamic redaction are provided. At least one object within a video to be redacted is identified. At least one region including the at least one object is redacted. The at least one redacted region including the at least one object and being larger than the at least one object. A tracked region being separate from the at least one object is identified. When the tracked region intersects with the at least one redacted region, the at least one redacted region is modified to un-redact portions of the at least one redacted region that overlap the tracked region.
1 . A method of dynamic redaction comprising:
identifying at least one object within a video to be redacted;
redacting at least one region including the at least one object, the at least one redacted region including the at least one object and being larger than the at least one object;
identifying a tracked region, the tracked region being separate from the at least one object; and
when the tracked region intersects with the at least one redacted region:
modifying the at least one redacted region to un-redact portions of the at least one redacted region that overlap the tracked region, wherein the modifying of the at least one redacted region is limited by a boundary of the at least one object.
2 . The method of claim 1 wherein the tracked region comprises an object of interest.
3 . The method of claim 1 further comprising:
when the un-redacted portions of the at least one redacted region no longer intersect the tracked region:
modifying the at least one redacted region to re-redact portions of the at least one redacted region that were un-redacted.
4 . The method of claim 1 further comprising:
detecting an object of interest in the video, the object of interest substantially occluded by the at least one redacted region; and
detecting a user has zoomed in on the object of interest, wherein the tracked region includes the object of interest.
5 . The method of claim 2 wherein the tracked region is congruous with the object of interest.
6 . A system for dynamic redaction comprising:
a processor; and
a memory coupled to the processor, the memory containing a set of instructions thereon that when executed by the processor cause the processor to:
identify at least one object within a video to be redacted;
redact at least one region including the at least one object, the at least one redacted region including the at least one object and being larger than the at least one object;
identify a tracked region, the tracked region being separate from the at least one object; and
when the tracked region intersects with the at least one redacted region:
modify the at least one redacted region to un-redact portions of the at least one redacted region that overlap the tracked region, wherein the instructions to modify the at least one redacted region is limited by a boundary of the at least one object.
7 . The system of claim 6 wherein the tracked region comprises an object of interest.
8 . The system of claim 6 wherein the memory further includes instructions to:
when the un-redacted portions of the at least one redacted region no longer intersect the tracked region:
modify the at least one redacted region to re-redact portions of the at least one redacted region that were un-redacted.
9 . The system of claim 6 wherein the memory further includes instructions to:
detect an object of interest in the video, the object of interest substantially occluded by the at least one redacted region; and
detect a user has zoomed in on the object of interest, wherein the tracked region includes the object of interest.
10 . The system of claim 7 wherein the tracked region is congruous with the object of interest.
11 . A non-transitory processor readable medium containing a set of instructions thereon that when executed by a processor cause the processor to:
identify at least one object within a video to be redacted;
redact at least one region including the at least one object, the at least one redacted region including the at least one object and being larger than the at least one object;
identify a tracked region, the tracked region being separate from the at least one object; and
when the tracked region intersects with the at least one redacted region:
modify the at least one redacted region to un-redact portions of the at least one redacted region that overlap the tracked region, wherein the instructions to modify the at least one redacted region is limited by a boundary of the at least one object.
12 . The medium of claim 11 wherein the tracked region comprises an object of interest.
13 . The medium of claim 11 wherein the medium further includes instructions to:
when the un-redacted portions of the at least one redacted region no longer intersect the tracked region:
modify the at least one redacted region to re-redact portions of the at least one redacted region that were un-redacted.
14 . The medium of claim 11 wherein the medium further includes instructions to:
detect an object of interest in the video, the object of interest substantially occluded by the at least one redacted region; and
detect a user has zoomed in on the object of interest, wherein the tracked region includes the object of interest.
15 . The medium of claim 12 wherein the tracked region is congruous with the object of interest.