Classification using multiframe analysis

1. A system for video analysis, comprising:

an interface configured to receive a trigger indication, wherein the trigger indication comprises a door opening signal following an engine running for greater than a cruising period, wherein the cruising period corresponds to 1 or more minutes; and

a processor configured to:

determine a time sequence set of video frames associated with the trigger indication;

determine a decision based at least in part on an analysis of the time sequence set of video frames, wherein the decision comprises a passenger pickup decision; and

indicate the decision.

2. The system as in claim 1 , wherein the analysis of the time sequence set of video frames comprises parallel processing of a plurality of frames of the time sequence set of video frames.

3. The system as in claim 2 , wherein the parallel processing is processed using a neural net.

4. The system as in claim 1 , wherein the analysis of the time sequence set of video frames includes classification of a frame of the time sequence set of video frames.

5. The system as in claim 4 , wherein the classification of the frame is combined with other classifications to determine the decision.

6. The system as in claim 1 , wherein the interface is configured to receive the time sequence set of video frames.

7. The system as in claim 1 , further comprising a data storage device, wherein the data storage device stores the time sequence set of video frames.

8. The system as in claim 1 , wherein the time sequence set of video frames is captured using a video camera mounted to a vehicle.

9. The system as in claim 8 , wherein the vehicle comprises one of the following: a car, a van, a truck, a truck with a trailer, or a delivery truck.

10. The system as in claim 8 , wherein the time sequence set of video frames is transmitted to a server from the vehicle.

11. The system as in claim 10 , wherein the analysis is performed by the server.

12. The system as in claim 1 , wherein the analysis is performed by a computing system in a vehicle.

13. The system as in claim 1 , wherein the trigger indication is received using a vehicle bus signal.

14. The system as in claim 1 , wherein the trigger indication further comprises one or more of the following: a seat weight signal, an audio detection signal, an engine RPM signal, a braking signal, or a vehicle bus signal.

15. The system as in claim 1 , wherein the trigger indication further comprises a signal generated using computer vision or a signal generated using data processing.

16. A method for video analysis, comprising:

receiving a trigger indication, wherein the trigger indication comprises a door opening signal following an engine running for greater than a cruising period, wherein the cruising period corresponds to 1 or more minutes;

determining, using a processor, a time sequence set of video frames associated with the trigger indication;

determining a decision based at least in part on an analysis of the time sequence set of video frames, wherein the decision comprises a passenger pickup decision; and

indicating the decision.

17. A computer program product for video analysis, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:

receiving a trigger indication, wherein the trigger indication comprises a door opening signal following an engine running for greater than a cruising period, wherein the cruising period corresponds to 1 or more minutes;

determining, using a processor, a time sequence set of video frames associated with the trigger indication;

determining a decision based at least in part on an analysis of the time sequence set of video frames, wherein the decision comprises a passenger pickup decision; and

indicating the decision.