Three-way video visitation detection using frame detection
A system and method for preventing unauthorized inmate to inmate communications comprises establishing a video conferencing session between a first device and a second device. Video content is extracted from video conferencing session data received from the second device. One or more artifacts are identified in the video content. The artifacts indicate that a third device is a source of at least a portion of the video content. The artifacts comprise at least three layers of objects in the video content, wherein the layers are associated with different rates of variation in linear measurements of the objects. In other embodiments, multiple backgrounds are detected in the video conferencing session data.
1. A method, comprising:
establishing a video conferencing session between a first device and a second device;
extracting video content from video conferencing session data received from the second device;
identifying three layers of physical objects in the video content, the layers each associated with different rates of variation in linear measurements of the physical objects, presence of the third of these layers identifying that a third device is a source of at least a portion of the video content.
2. The method of claim 1 , wherein the linear measurements of the physical objects are measured as a number of video pixels.
3. The method of claim 1 , wherein the linear measurements of the objects are selected from the group consisting of a width, a height, a length, and a distance.
4. The method of claim 1 , wherein at least one of the physical objects comprises a body part.
5. The method of claim 1 , wherein at least one of the physical objects in the video content comprises a case, frame, display screen, or function buttons for a video conferencing device.
6. The method of claim 1 , wherein the physical objects comprise a participant comprising one of the three layers, the participant surrounded almost entirely by a first background comprising a second of the three layers, and the first background surrounded almost entirely by a second background comprising a third of the three layers.
7. The method of claim 6 , wherein the first background and the second background are also differentiated from each other using color, lighting, or motion.
8. The method of claim 6 , wherein the first background and the second background are also separated by a border object in the video content.
9. The method of claim 1 , wherein a first of the three layers comprise a participant object and a second of the three layers comprises a first background that rotate together in the video content, and a third of the layers comprises a second background in the video content that does not rotate with the participant object and the first background.
10. The method of claim 9 , wherein the third layer second background rotates in a direction opposite to the first layer participant object and the second layer first background or at a different speed than the first layer participant object and the first background.
11. The method of claim 1 , further comprising:
generating an alert when the third device is identified as the source of at least a portion of in the video content.
12. The method of claim 1 , further comprising:
terminating the video conferencing session when the third device is identified as the source of at least a portion of the video content.
13. The method of claim 1 , further comprising:
flagging a video conferencing session record for further analysis when the third device is identified as the source of at least a portion of the video content.
14. A video conferencing system, comprising:
inbound video circuitry configured to extract video content from received video conferencing session data; and
video analysis circuitry configured to identify three layers of physical objects in the video content, the layers each associated with different rates of variation in linear measurements of the physical objects, presence of the third of these layers identifying that a third device is a source of at least a portion of the video content.
15. The video conferencing system of claim 14 , further comprising:
an alert system configured to perform one or more of the following when the third device is identified as the source of at least a portion of the video content:
generate an alert;
terminate the video conferencing session; and
flag a video conferencing session record for further analysis.
16. The video conferencing system of claim 14 , wherein the linear measurements of the physical objects are measured as a number of video pixels.
17. The video conferencing system of claim 14 , wherein at least one of the physical objects comprises a body part.
18. The video conferencing system of claim 14 , wherein at least one of the physical objects in the video content comprises a case, frame, display screen, or function buttons for a video conferencing device.
19. The video conferencing system of claim 14 , wherein the physical objects comprise a participant comprising one of the three layers, the participant surrounded almost entirely by a first background comprising a second of the three layers, and the first background surrounded almost entirely by a second background comprising a third of the three layers.
20. The video conferencing system of claim 14 , wherein a first of the three layers comprise a participant object and a second of the three layers comprises a first background that rotate together in the video content, and a third of the layers comprises a second background in the video content that does not rotate with the participant object and the first background.
21. A computer readable non-transitory medium having program instructions stored thereon that upon execution by a video conferencing system, cause the video conferencing system to:
establish a video conferencing session between a first device and a second device;
extract video content from video conferencing session data received from the second device; and
identify three layers of physical objects in the video content, the layers each associated with different rates of variation in linear measurements of the physical objects, presence of the third of these layers identifying that a third device is a source of at least a portion of the video content.